Order changes in PG16 since ICU introduction

From 3d2791af0a236cbc7ce7f29d988e8ac7fd3fd389 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 21 Apr 2023 14:03:57 -0700
Subject: [PATCH] ICU: do not convert locale 'C' to 'en-US-u-va-posix'.

The conversion was intended to be for convenience, but it's more
likely to be confusing than useful.

The user can still directly specify 'en-US-u-va-posix' if desired.

Discussion: https://postgr.es/m/f83f089ee1e9acd5dbbbf3353294d24e1f196e95.camel@j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 19 +------------------
 src/bin/initdb/initdb.c                       | 17 +----------------
 .../regress/expected/collate.icu.utf8.out     |  8 ++++++++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 ++++
 4 files changed, 14 insertions(+), 34 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 51df570ce9..58c4c426bc 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2782,26 +2782,10 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		if (elevel > 0)
-			ereport(elevel,
-					(errmsg("could not get language from locale \"%s\": %s",
-							loc_str, u_errorName(status))));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2889,8 +2873,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2c208ead01..4086834458 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2238,24 +2238,10 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		pg_fatal("could not get language from locale \"%s\": %s",
-				 loc_str, u_errorName(status));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2327,8 +2313,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..99f12d2e73 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1020,6 +1020,7 @@ CREATE ROLE regress_test_role;
 CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
@@ -1034,17 +1035,24 @@ BEGIN
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
+RESET icu_validation_level;
 RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+WARNING:  ICU locale "c" has unknown language "c"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..d9778faacc 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -358,6 +358,7 @@ CREATE SCHEMA test_schema;
 
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 
 do $$
 BEGIN
@@ -373,13 +374,16 @@ BEGIN
 END
 $$;
 
+RESET icu_validation_level;
 RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
-- 
2.34.1

From c768e040dc92b033e4eb0e69f08b59d8d1ffe1e4 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 21 Apr 2023 14:03:57 -0700
Subject: [PATCH v2 1/3] ICU: do not convert locale 'C' to 'en-US-u-va-posix'.

The conversion was intended to be for convenience, but it's more
likely to be confusing than useful.

The user can still directly specify 'en-US-u-va-posix' if desired.

Discussion: https://postgr.es/m/f83f089ee1e9acd5dbbbf3353294d24e1f196e95.camel@j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 19 +------------------
 src/bin/initdb/initdb.c                       | 17 +----------------
 .../regress/expected/collate.icu.utf8.out     |  8 ++++++++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 ++++
 4 files changed, 14 insertions(+), 34 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 51df570ce9..58c4c426bc 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2782,26 +2782,10 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		if (elevel > 0)
-			ereport(elevel,
-					(errmsg("could not get language from locale \"%s\": %s",
-							loc_str, u_errorName(status))));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2889,8 +2873,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2c208ead01..4086834458 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2238,24 +2238,10 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		pg_fatal("could not get language from locale \"%s\": %s",
-				 loc_str, u_errorName(status));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2327,8 +2313,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..99f12d2e73 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1020,6 +1020,7 @@ CREATE ROLE regress_test_role;
 CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
@@ -1034,17 +1035,24 @@ BEGIN
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
+RESET icu_validation_level;
 RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+WARNING:  ICU locale "c" has unknown language "c"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..d9778faacc 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -358,6 +358,7 @@ CREATE SCHEMA test_schema;
 
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 
 do $$
 BEGIN
@@ -373,13 +374,16 @@ BEGIN
 END
 $$;
 
+RESET icu_validation_level;
 RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
-- 
2.34.1

From 1302a4b65e4e12753ae15e732dab059afe69dbd9 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v2 2/3] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 +++++----
 src/backend/utils/adt/pg_locale.c             | 86 ++++++++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 12 +--
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +-
 5 files changed, 131 insertions(+), 59 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c91fe66d9b..7e69a889fb 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (pg_strcasecmp(colliculocale, "C") == 0 ||
+				pg_strcasecmp(colliculocale, "POSIX") == 0)
 			{
-				char *langtag = icu_language_tag(colliculocale,
-												 icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char *langtag = icu_language_tag(colliculocale,
+													 icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 2e242eeff2..8ef33871f0 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (pg_strcasecmp(dbiculocale, "C") == 0 ||
+			pg_strcasecmp(dbiculocale, "POSIX") == 0)
 		{
-			char *langtag = icu_language_tag(dbiculocale,
-											 icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char *langtag = icu_language_tag(dbiculocale,
+												 icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 58c4c426bc..06e7530247 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,14 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (pg_strcasecmp(iculocstr, "C") == 0 ||
+		pg_strcasecmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1686,10 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (pg_strcasecmp("C", collcollate) ||
+		pg_strcasecmp("POSIX", collcollate))
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1708,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2495,14 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (pg_strcasecmp(loc_str, "C") == 0 ||
+		pg_strcasecmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 99f12d2e73..53ab496bfe 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1042,21 +1042,21 @@ ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
-ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
-WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-WARNING:  ICU locale "c" has unknown language "c"
-HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 RESET icu_validation_level;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index d9778faacc..63d5352ee6 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,14 +379,17 @@ RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
+
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
 
-- 
2.34.1

From 1c30ea67e48bab60b7e96847ff3c24880e954471 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v2 3/3] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 ++--
 doc/src/sgml/ref/createdb.sgml                |  5 ++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++--
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++---
 src/bin/initdb/initdb.c                       | 31 ++++++++++++-------
 src/bin/scripts/createdb.c                    | 13 +++-----
 src/bin/scripts/t/020_createdb.pl             |  4 +--
 src/test/icu/t/010_database.pl                | 23 +++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 22 ++++++-------
 10 files changed, 77 insertions(+), 51 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 7e69a889fb..e481f20dc8 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 8ef33871f0..b447dc55f3 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 4086834458..1ef028617e 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2157,7 +2157,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2391,7 +2395,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2407,6 +2411,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	/*
@@ -2443,14 +2449,18 @@ setlocales(void)
 			printf(_("Using default ICU locale \"%s\".\n"), icu_locale);
 		}
 
-		/* canonicalize to a language tag */
-		langtag = icu_language_tag(icu_locale);
-		printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
-			   langtag, icu_locale);
-		pg_free(icu_locale);
-		icu_locale = langtag;
-
-		icu_validate_locale(icu_locale);
+		if (pg_strcasecmp(icu_locale, "C") != 0 &&
+			pg_strcasecmp(icu_locale, "POSIX") != 0)
+		{
+			/* canonicalize to a language tag */
+			langtag = icu_language_tag(icu_locale);
+			printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
+				   langtag, icu_locale);
+			pg_free(icu_locale);
+			icu_locale = langtag;
+
+			icu_validate_locale(icu_locale);
+		}
 
 		/*
 		 * In supported builds, the ICU locale ID will be opened during
@@ -3282,7 +3292,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..9ca86a3e53 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
@@ -219,6 +211,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index af3b1492e3..3db9fe931f 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -126,7 +126,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -134,7 +134,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 715b1bffd6..df4af00afe 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,16 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 53ab496bfe..ecceb6d10c 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1202,9 +1202,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1212,7 +1212,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1229,12 +1229,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1243,7 +1243,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1271,13 +1271,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1327,9 +1327,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 6c0251c584edea64148604da52c8e55e43fe36e6 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 21 Apr 2023 14:03:57 -0700
Subject: [PATCH v3 1/4] ICU: do not convert locale 'C' to 'en-US-u-va-posix'.

The conversion was intended to be for convenience, but it's more
likely to be confusing than useful.

The user can still directly specify 'en-US-u-va-posix' if desired.

Discussion: https://postgr.es/m/f83f089ee1e9acd5dbbbf3353294d24e1f196e95.camel@j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 19 +------------------
 src/bin/initdb/initdb.c                       | 17 +----------------
 .../regress/expected/collate.icu.utf8.out     |  8 ++++++++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 ++++
 4 files changed, 14 insertions(+), 34 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 51df570ce9..58c4c426bc 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2782,26 +2782,10 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		if (elevel > 0)
-			ereport(elevel,
-					(errmsg("could not get language from locale \"%s\": %s",
-							loc_str, u_errorName(status))));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2889,8 +2873,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2c208ead01..4086834458 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2238,24 +2238,10 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		pg_fatal("could not get language from locale \"%s\": %s",
-				 loc_str, u_errorName(status));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2327,8 +2313,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..99f12d2e73 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1020,6 +1020,7 @@ CREATE ROLE regress_test_role;
 CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
@@ -1034,17 +1035,24 @@ BEGIN
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
+RESET icu_validation_level;
 RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+WARNING:  ICU locale "c" has unknown language "c"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..d9778faacc 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -358,6 +358,7 @@ CREATE SCHEMA test_schema;
 
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 
 do $$
 BEGIN
@@ -373,13 +374,16 @@ BEGIN
 END
 $$;
 
+RESET icu_validation_level;
 RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
-- 
2.34.1

From 22a8ba5748953fbc577f7aeb8d8d85d185364fb7 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v3 2/4] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 +++++----
 src/backend/utils/adt/pg_locale.c             | 86 ++++++++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 12 +--
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +-
 5 files changed, 131 insertions(+), 59 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c91fe66d9b..7e69a889fb 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (pg_strcasecmp(colliculocale, "C") == 0 ||
+				pg_strcasecmp(colliculocale, "POSIX") == 0)
 			{
-				char *langtag = icu_language_tag(colliculocale,
-												 icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char *langtag = icu_language_tag(colliculocale,
+													 icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 2e242eeff2..8ef33871f0 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (pg_strcasecmp(dbiculocale, "C") == 0 ||
+			pg_strcasecmp(dbiculocale, "POSIX") == 0)
 		{
-			char *langtag = icu_language_tag(dbiculocale,
-											 icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char *langtag = icu_language_tag(dbiculocale,
+												 icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 58c4c426bc..3e19b21122 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,14 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (pg_strcasecmp(iculocstr, "C") == 0 ||
+		pg_strcasecmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1686,10 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (pg_strcasecmp("C", collcollate) == 0 ||
+		pg_strcasecmp("POSIX", collcollate) == 0)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1708,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2495,14 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (pg_strcasecmp(loc_str, "C") == 0 ||
+		pg_strcasecmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 99f12d2e73..53ab496bfe 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1042,21 +1042,21 @@ ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
-ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
-WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-WARNING:  ICU locale "c" has unknown language "c"
-HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 RESET icu_validation_level;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index d9778faacc..63d5352ee6 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,14 +379,17 @@ RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
+
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
 
-- 
2.34.1

From b33dc56960378a1047ccf9c0387a1fe333912140 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 28 Apr 2023 12:22:41 -0700
Subject: [PATCH v3 3/4] ICU: fix up old libc-style locale strings.

Before transforming a locale string into a language tag, fix up old
libc-style locale strings such as 'de__PHONEBOOK' or
'fr_FR@EURO'. Older ICU versions did this automatically, but ICU
version 64 removed that support.
---
 src/backend/utils/adt/pg_locale.c             | 59 ++++++++++++++++-
 src/bin/initdb/initdb.c                       | 63 ++++++++++++++++++-
 .../regress/expected/collate.icu.utf8.out     | 11 ++++
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +++
 4 files changed, 138 insertions(+), 2 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 3e19b21122..9f2c139b0b 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2812,6 +2812,60 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 	pfree(lower_str);
 }
 
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = palloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pstrdup(loc_str);
+}
+
 #endif
 
 /*
@@ -2828,6 +2882,7 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
@@ -2844,7 +2899,7 @@ icu_language_tag(const char *loc_str, int elevel)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2864,6 +2919,8 @@ icu_language_tag(const char *loc_str, int elevel)
 		break;
 	}
 
+	pfree(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pfree(langtag);
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 4086834458..600c8d93f3 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2229,6 +2229,64 @@ check_icu_locale_encoding(int user_enc)
 	return true;
 }
 
+#ifdef USE_ICU
+
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = pg_malloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pg_strdup(loc_str);
+}
+
+#endif
+
 /*
  * Convert to canonical BCP47 language tag. Must be consistent with
  * icu_language_tag().
@@ -2238,6 +2296,7 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
@@ -2254,7 +2313,7 @@ icu_language_tag(const char *loc_str)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2273,6 +2332,8 @@ icu_language_tag(const char *loc_str)
 		break;
 	}
 
+	pg_free(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pg_free(langtag);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 53ab496bfe..5f5b61d036 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1048,15 +1048,26 @@ CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W');
 ERROR:  RULES not supported for C or POSIX locale
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
+ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
+WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-cu-eur" for ICU locale "@EURO"
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-pinyin" for ICU locale "@pinyin"
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-stroke" for ICU locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 63d5352ee6..e4bbd2c009 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -382,14 +382,21 @@ CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
 CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 RESET icu_validation_level;
 
 CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
 
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
 
-- 
2.34.1

From 066eac039f86d95ad853aa1e5de2b34fdf688f2e Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v3 4/4] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 ++--
 doc/src/sgml/ref/createdb.sgml                |  5 ++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++--
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++---
 src/bin/initdb/initdb.c                       | 31 ++++++++++++-------
 src/bin/scripts/createdb.c                    | 13 +++-----
 src/bin/scripts/t/020_createdb.pl             |  4 +--
 src/test/icu/t/010_database.pl                | 23 +++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 22 ++++++-------
 10 files changed, 77 insertions(+), 51 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 7e69a889fb..e481f20dc8 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 8ef33871f0..b447dc55f3 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 600c8d93f3..7e316c8ba9 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2157,7 +2157,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2452,7 +2456,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2468,6 +2472,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	/*
@@ -2504,14 +2510,18 @@ setlocales(void)
 			printf(_("Using default ICU locale \"%s\".\n"), icu_locale);
 		}
 
-		/* canonicalize to a language tag */
-		langtag = icu_language_tag(icu_locale);
-		printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
-			   langtag, icu_locale);
-		pg_free(icu_locale);
-		icu_locale = langtag;
-
-		icu_validate_locale(icu_locale);
+		if (pg_strcasecmp(icu_locale, "C") != 0 &&
+			pg_strcasecmp(icu_locale, "POSIX") != 0)
+		{
+			/* canonicalize to a language tag */
+			langtag = icu_language_tag(icu_locale);
+			printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
+				   langtag, icu_locale);
+			pg_free(icu_locale);
+			icu_locale = langtag;
+
+			icu_validate_locale(icu_locale);
+		}
 
 		/*
 		 * In supported builds, the ICU locale ID will be opened during
@@ -3343,7 +3353,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..9ca86a3e53 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
@@ -219,6 +211,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index af3b1492e3..3db9fe931f 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -126,7 +126,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -134,7 +134,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 715b1bffd6..df4af00afe 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,16 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 5f5b61d036..566e91d2d9 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1213,9 +1213,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1223,7 +1223,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1240,12 +1240,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1254,7 +1254,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1282,13 +1282,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1338,9 +1338,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From ddda683963959a175dff17ab0e3d8519641498b9 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 21 Apr 2023 14:03:57 -0700
Subject: [PATCH v4 1/4] ICU: do not convert locale 'C' to 'en-US-u-va-posix'.

The conversion was intended to be for convenience, but it's more
likely to be confusing than useful.

The user can still directly specify 'en-US-u-va-posix' if desired.

Discussion: https://postgr.es/m/f83f089ee1e9acd5dbbbf3353294d24e1f196e95.camel@j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 19 +------------------
 src/bin/initdb/initdb.c                       | 17 +----------------
 .../regress/expected/collate.icu.utf8.out     |  8 ++++++++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 ++++
 4 files changed, 14 insertions(+), 34 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index f0b6567da1..51b4221a39 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2782,26 +2782,10 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		if (elevel > 0)
-			ereport(elevel,
-					(errmsg("could not get language from locale \"%s\": %s",
-							loc_str, u_errorName(status))));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2889,8 +2873,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2c208ead01..4086834458 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2238,24 +2238,10 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
-	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
 
-	status = U_ZERO_ERROR;
-	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
-	{
-		pg_fatal("could not get language from locale \"%s\": %s",
-				 loc_str, u_errorName(status));
-		return NULL;
-	}
-
-	/* C/POSIX locales aren't handled by uloc_getLanguageTag() */
-	if (strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
-		return pstrdup("en-US-u-va-posix");
-
 	/*
 	 * A BCP47 language tag doesn't have a clearly-defined upper limit
 	 * (cf. RFC5646 section 4.4). Additionally, in older ICU versions,
@@ -2327,8 +2313,7 @@ icu_validate_locale(const char *loc_str)
 
 	/* check for special language name */
 	if (strcmp(lang, "") == 0 ||
-		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0 ||
-		strcmp(lang, "c") == 0 || strcmp(lang, "posix") == 0)
+		strcmp(lang, "root") == 0 || strcmp(lang, "und") == 0)
 		found = true;
 
 	/* search for matching language within ICU */
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..99f12d2e73 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1020,6 +1020,7 @@ CREATE ROLE regress_test_role;
 CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
@@ -1034,17 +1035,24 @@ BEGIN
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
+RESET icu_validation_level;
 RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+WARNING:  ICU locale "c" has unknown language "c"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..d9778faacc 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -358,6 +358,7 @@ CREATE SCHEMA test_schema;
 
 -- We need to do this this way to cope with varying names for encodings:
 SET client_min_messages TO WARNING;
+SET icu_validation_level = disabled;
 
 do $$
 BEGIN
@@ -373,13 +374,16 @@ BEGIN
 END
 $$;
 
+RESET icu_validation_level;
 RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 RESET icu_validation_level;
 
-- 
2.34.1

From 3db6abd0fe56e9c4b6653e04e28f6f77381c2fc8 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 28 Apr 2023 12:22:41 -0700
Subject: [PATCH v4 2/4] ICU: fix up old libc-style locale strings.

Before transforming a locale string into a language tag, fix up old
libc-style locale strings such as 'fr_FR@euro'. Older ICU versions did
this automatically, but ICU version 64 removed that support.

Discussion: https://postgr.es/m/654a49f7ff7461bcf47be4181430678d45f93858.camel%40j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 59 ++++++++++++++++-
 src/bin/initdb/initdb.c                       | 63 ++++++++++++++++++-
 .../regress/expected/collate.icu.utf8.out     | 11 ++++
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +++
 4 files changed, 138 insertions(+), 2 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 51b4221a39..0e7343b28b 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2766,6 +2766,60 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 	pfree(lower_str);
 }
 
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = palloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pstrdup(loc_str);
+}
+
 #endif
 
 /*
@@ -2782,6 +2836,7 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
@@ -2798,7 +2853,7 @@ icu_language_tag(const char *loc_str, int elevel)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2818,6 +2873,8 @@ icu_language_tag(const char *loc_str, int elevel)
 		break;
 	}
 
+	pfree(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pfree(langtag);
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 4086834458..600c8d93f3 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2229,6 +2229,64 @@ check_icu_locale_encoding(int user_enc)
 	return true;
 }
 
+#ifdef USE_ICU
+
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = pg_malloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pg_strdup(loc_str);
+}
+
+#endif
+
 /*
  * Convert to canonical BCP47 language tag. Must be consistent with
  * icu_language_tag().
@@ -2238,6 +2296,7 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
 	const bool	 strict = true;
@@ -2254,7 +2313,7 @@ icu_language_tag(const char *loc_str)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2273,6 +2332,8 @@ icu_language_tag(const char *loc_str)
 		break;
 	}
 
+	pg_free(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pg_free(langtag);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 99f12d2e73..d520674edf 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1046,6 +1046,8 @@ CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
+ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
@@ -1056,7 +1058,16 @@ HINT:  To disable ICU locale validation, set parameter icu_validation_level to D
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
+WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 RESET icu_validation_level;
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-pinyin" for locale "@pinyin"
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-stroke" for locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index d9778faacc..ab9a8484b9 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -381,12 +381,19 @@ CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, nee
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 RESET icu_validation_level;
 
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
 
-- 
2.34.1

From 05597bea2f48cb1ef78a745401bcabdd29245b84 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v4 3/4] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 +++++----
 src/backend/utils/adt/pg_locale.c             | 86 ++++++++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 12 +--
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +-
 5 files changed, 131 insertions(+), 59 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c91fe66d9b..7e69a889fb 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (pg_strcasecmp(colliculocale, "C") == 0 ||
+				pg_strcasecmp(colliculocale, "POSIX") == 0)
 			{
-				char *langtag = icu_language_tag(colliculocale,
-												 icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char *langtag = icu_language_tag(colliculocale,
+													 icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 2e242eeff2..8ef33871f0 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (pg_strcasecmp(dbiculocale, "C") == 0 ||
+			pg_strcasecmp(dbiculocale, "POSIX") == 0)
 		{
-			char *langtag = icu_language_tag(dbiculocale,
-											 icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char *langtag = icu_language_tag(dbiculocale,
+												 icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 0e7343b28b..76ca42441d 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,14 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (pg_strcasecmp(iculocstr, "C") == 0 ||
+		pg_strcasecmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1686,10 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (pg_strcasecmp("C", collcollate) == 0 ||
+		pg_strcasecmp("POSIX", collcollate) == 0)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1708,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2495,14 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (pg_strcasecmp(loc_str, "C") == 0 ||
+		pg_strcasecmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index d520674edf..f217658151 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1042,8 +1042,10 @@ ERROR:  parameter "locale" must be specified
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
-ERROR:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
@@ -1051,16 +1053,14 @@ ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMEN
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
-WARNING:  could not convert locale name "c" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-WARNING:  ICU locale "c" has unknown language "c"
-HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 RESET icu_validation_level;
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
 NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index ab9a8484b9..e4bbd2c009 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,16 +379,19 @@ RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-CREATE COLLATION testx (provider = icu, locale = 'c'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'c', rules = '&V << w <<< W'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
-CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 RESET icu_validation_level;
 
+CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
+
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
-- 
2.34.1

From 310bdcd136e44bfca1eea4da5181886eac02d52d Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v4 4/4] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 ++--
 doc/src/sgml/ref/createdb.sgml                |  5 ++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++--
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++---
 src/bin/initdb/initdb.c                       | 31 ++++++++++++-------
 src/bin/scripts/createdb.c                    | 13 +++-----
 src/bin/scripts/t/020_createdb.pl             |  4 +--
 src/test/icu/t/010_database.pl                | 23 +++++++++-----
 .../regress/expected/collate.icu.utf8.out     | 28 ++++++++---------
 10 files changed, 80 insertions(+), 54 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 7e69a889fb..e481f20dc8 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 8ef33871f0..b447dc55f3 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 600c8d93f3..7e316c8ba9 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2157,7 +2157,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2452,7 +2456,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2468,6 +2472,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	/*
@@ -2504,14 +2510,18 @@ setlocales(void)
 			printf(_("Using default ICU locale \"%s\".\n"), icu_locale);
 		}
 
-		/* canonicalize to a language tag */
-		langtag = icu_language_tag(icu_locale);
-		printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
-			   langtag, icu_locale);
-		pg_free(icu_locale);
-		icu_locale = langtag;
-
-		icu_validate_locale(icu_locale);
+		if (pg_strcasecmp(icu_locale, "C") != 0 &&
+			pg_strcasecmp(icu_locale, "POSIX") != 0)
+		{
+			/* canonicalize to a language tag */
+			langtag = icu_language_tag(icu_locale);
+			printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
+				   langtag, icu_locale);
+			pg_free(icu_locale);
+			icu_locale = langtag;
+
+			icu_validate_locale(icu_locale);
+		}
 
 		/*
 		 * In supported builds, the ICU locale ID will be opened during
@@ -3343,7 +3353,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..9ca86a3e53 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
@@ -219,6 +211,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index af3b1492e3..3db9fe931f 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -126,7 +126,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -134,7 +134,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 715b1bffd6..df4af00afe 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,16 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index f217658151..566e91d2d9 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1063,11 +1063,11 @@ CREATE COLLATION testx (provider = icu, locale = 'c'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'posix'); DROP COLLATION testx;
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
+NOTICE:  using standard form "und-u-cu-eur" for ICU locale "@EURO"
 CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-co-pinyin" for locale "@pinyin"
+NOTICE:  using standard form "und-u-co-pinyin" for ICU locale "@pinyin"
 CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-co-stroke" for locale "@stroke"
+NOTICE:  using standard form "und-u-co-stroke" for ICU locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
@@ -1213,9 +1213,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1223,7 +1223,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1240,12 +1240,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1254,7 +1254,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1282,13 +1282,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1338,9 +1338,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From b09515bfaf5e9de330138ec4a627d02a7947de1a Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 27 Apr 2023 14:43:46 -0700
Subject: [PATCH v1] Doc improvements for language tags and custom ICU
 collations.

Separate the documentation for language tags from the documentaiton
for the available collation settings which can be included in a
language tag.

Include tables of the available options, more details about the
effects of each option, and additional examples.

Also include an explanation of the "levels" of textual features and
how they relate to collation.
---
 doc/src/sgml/charset.sgml | 656 +++++++++++++++++++++++++++++++-------
 1 file changed, 535 insertions(+), 121 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 6dd95b8966..be74064168 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -377,7 +377,125 @@ initdb --locale-provider=icu --icu-locale=en
     variants and customization options.
    </para>
   </sect2>
+  <sect2 id="icu-locales">
+   <title>ICU Locales</title>
+   <sect3 id="icu-locale-names">
+    <title>ICU Locale Names</title>
+    <para>
+     The ICU format for the locale name is a <link
+     linkend="icu-language-tag">Language Tag</link>.
+
+<programlisting>
+CREATE COLLATION mycollation1 (PROVIDER = icu, LOCALE = 'ja-JP);
+CREATE COLLATION mycollation2 (PROVIDER = icu, LOCALE = 'fr');
+</programlisting>
+    </para>
+   </sect3>
+   <sect3 id="icu-canonicalization">
+    <title>Locale Canonicalization and Validation</title>
+    <para>
+     When defining a new ICU collation object or database with ICU as the
+     provider, the given locale name is transformed ("canonicalized") into a
+     language tag if not already in that form. For instance,
+
+<screen>
+CREATE COLLATION mycollation3 (PROVIDER = icu, LOCALE = 'en-US-u-kn-true');
+NOTICE:  using standard form "en-US-u-kn" for locale "en-US-u-kn-true"
+CREATE COLLATION mycollation4 (PROVIDER = icu, LOCALE = 'de_DE.utf8');
+NOTICE:  using standard form "de-DE" for locale "de_DE.utf8"
+</screen>
+
+     If you see such a message, ensure that the <symbol>PROVIDER</symbol> and
+     <symbol>LOCALE</symbol> are as you expect, and consider specifying
+     directly as the canonical language tag instead of relying on the
+     transformation.
+    </para>
+    <note>
+     <para>
+      ICU can transform most libc locale names, as well as some other formats,
+      into language tags for easier transition to ICU. If a libc locale name
+      is used in ICU, it may not have precisely the same behavior as in libc.
+     </para>
+    </note>
+    <para>
+     If there is some problem interpreting the locale name, or if it represents
+     a language or region that ICU does not recognize, a message will be reported:
 
+<screen>
+SET icu_validation_level = ERROR;
+CREATE COLLATION nonsense (PROVIDER = icu, LOCALE = 'nonsense');
+ERROR:  ICU locale "nonsense" has unknown language "nonsense"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+</screen>
+
+     <xref
+     linkend="guc-icu-validation-level"/> controls how the message is
+     reported. If set below <literal>ERROR</literal>, the collation will still
+     be created, but the behavior may not be what the user intended.
+    </para>
+   </sect3>
+   <sect3 id="icu-language-tag">
+    <title>Language Tag</title>
+    <para>
+     Basic language tags are simply
+     <replaceable>language</replaceable><literal>-</literal><replaceable>region</replaceable>;
+     or even just <replaceable>language</replaceable>. The
+     <replaceable>language</replaceable> is a language code
+     (e.g. <literal>fr</literal> for French or <literal>und</literal> for
+     "undefined"), and <replaceable>region</replaceable> is a region code
+     (e.g. <literal>CA</literal> for Canada). Examples:
+     <literal>ja-JP</literal>, <literal>de</literal>, or
+     <literal>fr-CA</literal>.
+    </para>
+    <para>
+     Collation settings may be included in the language tag to customize
+     collation behavior. ICU allows extensive customization, such as
+     sensitivity (or insensitivity) to accents, case, and punctuation;
+     treatment of digits within text; and many other options to satisfy a
+     variety of uses.
+    </para>
+    <para>
+     To include this additional collation information in a language tag,
+     append <literal>-u</literal>, followed by one or more
+     <literal>-</literal><replaceable>key</replaceable><literal>-</literal><replaceable>value</replaceable>
+     pairs, where <replaceable>key</replaceable> is the key for a collation
+     setting and <replaceable>value</replaceable> is a valid value for that
+     setting. For boolean settings, the
+     <literal>-</literal><replaceable>key</replaceable> may be specified
+     without a corresponding
+     <literal>-</literal><replaceable>value</replaceable>, which implies a
+     value of <literal>true</literal>.
+    </para>
+    <para>
+     For example, the language tag <literal>en-US-u-kn-ks-level2</literal>
+     means the locale with the English language in the US region, with
+     collation settings <literal>kn</literal> set to <literal>true</literal>
+     and <literal>ks</literal> set to <literal>level2</literal>. Those
+     settings mean the collation will be case-insensitive and treat a sequence
+     of digits as a single number:
+
+<screen>
+CREATE COLLATION mycollation5 (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'en-US-u-kn-ks-level2');
+SELECT 'aB' = 'Ab' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+
+SELECT 'N-45' &lt; 'N-123' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+</screen>
+    </para>
+    <para>
+     See <xref linkend="icu-custom-collations"/> for details and additional
+     examples of using language tags with custom collation information for the
+     locale.
+    </para>
+   </sect3>
+  </sect2>
   <sect2 id="locale-problems">
    <title>Problems</title>
 
@@ -658,6 +776,13 @@ SELECT * FROM test1 ORDER BY a || b COLLATE "fr_FR";
     code byte values.
    </para>
 
+   <note>
+    <para>
+     The <literal>C</literal> and <literal>POSIX</literal> locales may behave
+     differently depending on the database encoding.
+    </para>
+   </note>
+
    <para>
     Additionally, two SQL standard collation names are available:
 
@@ -869,132 +994,23 @@ CREATE COLLATION german (provider = libc, locale = 'de_DE');
    <sect4 id="collation-managing-create-icu">
     <title>ICU Collations</title>
 
-   <para>
-    ICU allows collations to be customized beyond the basic language+country
-    set that is preloaded by <command>initdb</command>.  Users are encouraged
-    to define their own collation objects that make use of these facilities to
-    suit the sorting behavior to their requirements.
-    See <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
-    and <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink> for
-    information on ICU locale naming.  The set of acceptable names and
-    attributes depends on the particular ICU version.
-   </para>
-
-   <para>
-    Here are some examples:
-
-    <variablelist>
-     <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de@collation=phonebook');</literal></term>
-      <listitem>
-       <para>German collation with phone book collation type</para>
-       <para>
-        The first example selects the ICU locale using a <quote>language
-        tag</quote> per BCP 47.  The second example uses the traditional
-        ICU-specific locale syntax.  The first style is preferred going
-        forward, and is used internally to store locales.
-       </para>
-       <para>
-        Note that you can name the collation objects in the SQL environment
-        anything you want.  In this example, we follow the naming style that
-        the predefined collations use, which in turn also follow BCP 47, but
-        that is not required for user-defined collations.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = '@collation=emoji');</literal></term>
-      <listitem>
-       <para>
-        Root collation with Emoji collation type, per Unicode Technical Standard #51
-       </para>
-       <para>
-        Observe how in the traditional ICU locale naming system, the root
-        locale is selected by an empty string.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en@colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en@colCaseFirst=upper');</literal></term>
-      <listitem>
-       <para>
-        Sort upper-case letters before lower-case letters.  (The default is
-        lower-case letters first.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-    <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en@colCaseFirst=upper;colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Combines both of the above options.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kn-true">
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en-u-kn-true');</literal></term>
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en@colNumeric=yes');</literal></term>
-      <listitem>
-       <para>
-        Numeric ordering, sorts sequences of digits by their numeric value,
-        for example: <literal>A-21</literal> &lt; <literal>A-123</literal>
-        (also known as natural sort).
-       </para>
-      </listitem>
-     </varlistentry>
-    </variablelist>
-
-    See <ulink url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
-    Technical Standard #35</ulink>
-    and <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink> for
-    details.  The list of possible collation types (<literal>co</literal>
-    subtag) can be found in
-    the <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
-    repository</ulink>.
-   </para>
+    <para>
+     ICU collations can be created like:
 
-   <para>
-    Note that while this system allows creating collations that <quote>ignore
-    case</quote> or <quote>ignore accents</quote> or similar (using the
-    <literal>ks</literal> key), in order for such collations to act in a
-    truly case- or accent-insensitive manner, they also need to be declared as not
-    <firstterm>deterministic</firstterm> in <command>CREATE COLLATION</command>;
-    see <xref linkend="collation-nondeterministic"/>.
-    Otherwise, any strings that compare equal according to the collation but
-    are not byte-wise equal will be sorted according to their byte values.
-   </para>
+<programlisting>
+CREATE COLLATION german (provider = icu, locale = 'de-DE');
+</programlisting>
 
-   <note>
+     ICU locales are specified as a <link linkend="icu-language-tag">Language
+     Tag</link>, but can also accept most libc-style locale names (which will
+     be transformed into language tags if possible).
+    </para>
     <para>
-     By design, ICU will accept almost any string as a locale name and match
-     it to the closest locale it can provide, using the fallback procedure
-     described in its documentation.  Thus, there will be no direct feedback
-     if a collation specification is composed using features that the given
-     ICU installation does not actually support.  It is therefore recommended
-     to create application-level test cases to check that the collation
-     definitions satisfy one's requirements.
+     New ICU collations can customize collation behavior extensively by
+     including collation attributes in the langugage tag. See <xref
+     linkend="icu-custom-collations"/> for details and examples.
     </para>
-   </note>
    </sect4>
-
    <sect4 id="collation-copy">
    <title>Copying Collations</title>
 
@@ -1072,6 +1088,404 @@ CREATE COLLATION ignore_accents (provider = icu, locale = 'und-u-ks-level1-kc-tr
     </tip>
    </sect3>
   </sect2>
+  <sect2 id="icu-custom-collations">
+   <title>ICU Custom Collations</title>
+
+   <para>
+    ICU allows extensive control over collation behavior by defining new
+    collations with collation settings as a part of the language tag. These
+    settings can modify the collation order to suit a variety of needs. For
+    instance:
+
+<programlisting>
+-- ignore differences in accents and case
+CREATE COLLATION ignore_accent_case (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ks-level1');
+SELECT 'Å' = 'A' COLLATE ignore_accent_case; -- true
+SELECT 'z' = 'Z' COLLATE ignore_accent_case; -- true
+
+-- upper case letters sort before lower case.
+CREATE COLLATION upper_first (PROVIDER=icu, LOCALE = 'und-u-kf-upper');
+SELECT 'B' &lt; 'b' COLLATE upper_first; -- true
+
+-- treat digits numerically and ignore punctuation
+CREATE COLLATION num_ignore_punct (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ka-shifted-kn');
+SELECT 'id-45' &lt; 'id-123' COLLATE num_ignore_punct; -- true
+SELECT 'w;x*y-z' = 'wxyz' COLLATE num_ignore_punct; -- true
+</programlisting>
+
+    Many of the available options are described in <xref
+    linkend="icu-collation-settings"/>, or see <xref
+    linkend="icu-external-references"/> for more details.
+   </para>
+   <sect3 id="icu-collation-comparison-levels">
+    <title>ICU Comparison Levels</title>
+    <para>
+     Comparison of two strings (collation) in ICU is determined by a
+     multi-level process, where textual features are grouped into
+     "levels". Treatment of each level is controlled by the <link
+     linkend="icu-collation-settings-table">collation settings</link>. Higher
+     levels correspond to finer textual features.
+    </para>
+    <para>
+     <table id="icu-collation-levels">
+      <title>ICU Collation Levels</title>
+      <tgroup cols="3">
+       <thead>
+        <row>
+         <entry>Level</entry>
+         <entry>Description</entry>
+         <entry><literal>'f' = 'f'</literal></entry>
+         <entry><literal>'ab' = U&amp;'a\2063b'</literal></entry>
+         <entry><literal>'x-y' = 'x_y'</literal></entry>
+         <entry><literal>'g' = 'G'</literal></entry>
+         <entry><literal>'n' = 'ñ'</literal></entry>
+         <entry><literal>'y' = 'z'</literal></entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry>level1</entry>
+         <entry>Base Character</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level2</entry>
+         <entry>Accents</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level3</entry>
+         <entry>Case/Variants</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level4</entry>
+         <entry>Punctuation</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>identic</entry>
+         <entry>All</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+
+     The above table shows which textual feature differences are
+     considered significant when determining equality at the given level. The
+     unicode character <literal>U+2063</literal> is an invisible separator,
+     and as seen in the table, is ignored for at all levels of comparison less
+     than <literal>identic</literal>.
+    </para>
+    <para>
+     Examples:
+
+<programlisting>
+CREATE COLLATION level3 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level3');
+CREATE COLLATION level4 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level4');
+CREATE COLLATION identic (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-identic');
+
+-- invisible separator ignored at all levels except identic
+SELECT 'ab' = U&amp;'a\2063b' COLLATE level4; -- true
+SELECT 'ab' = U&amp;'a\2063b' COLLATE identic; -- false
+
+-- punctuation ignored at level3 but not at level 4
+SELECT 'x-y' = 'x_y' COLLATE level3; -- true
+SELECT 'x-y' = 'x_y' COLLATE level4; -- false
+</programlisting>
+
+    </para>
+    <note>
+     <para>
+      For many collation settings, you must create the collation with
+      <option>DETERMINISTIC</option> set to <literal>false</literal> for the
+      setting to have the desired effect. Additionally, some settings only
+      take effect when the key <literal>ka</literal> is set to
+      <literal>shifted</literal> (see <xref
+      linkend="icu-collation-settings-table"/>).
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="icu-collation-settings">
+    <title>Collation Settings for an ICU Locale</title>
+    <para>
+     <table id="icu-collation-settings-table">
+      <title>ICU Collation Settings</title>
+      <tgroup cols="4">
+       <thead>
+        <row>
+         <entry>Key</entry>
+         <entry>Values</entry>
+         <entry>Default</entry>
+         <entry>Description</entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry><literal>co</literal></entry>
+         <entry><literal>emoji</literal>, <literal>phonebk</literal>, <literal>standard</literal>, <replaceable>...</replaceable></entry>
+         <entry><literal>standard</literal></entry>
+         <entry>
+          Collation type. See <xref linkend="icu-external-references"/> for additional options and details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ks</literal></entry>
+         <entry><literal>level1</literal>, <literal>level2</literal>, <literal>level3</literal>, <literal>level4</literal>, <literal>identic</literal></entry>
+         <entry><literal>level3</literal></entry>
+         <entry>
+          Sensitivity when determining equality, with
+          <literal>level1</literal> the least sensitive and
+          <literal>identic</literal> the most sensitive. See <xref
+          linkend="icu-collation-levels"/> for details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ka</literal></entry>
+         <entry><literal>noignore</literal>, <literal>shifted</literal></entry>
+         <entry><literal>noignore</literal></entry>
+         <entry>
+          If set to <literal>shifted</literal>, causes some characters
+          (e.g. punctuation or space) to be ignored in comparison. Key
+          <literal>ks</literal> must be set to <literal>level3</literal> or
+          lower to take effect. Set key <literal>kv</literal> to control which
+          character classes are ignored.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kb</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          Backwards comparison for the level 2 differences. For example,
+          locale <literal>und-u-kb</literal> sorts <literal>'àe'</literal>
+          before <literal>'aé'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kk</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Enable full normalization; may affect performance. Basic
+           normalization is performed even when set to
+           <literal>false</literal>.
+          </para>
+          <para>
+           Full normalization is important in some cases, such as when
+           multiple accents are applied to a single character (e.g. in
+           Vietnamese or Arabic). Locales for languages that require full
+           normalization typically enable it by default.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kc</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Separates case into a "level 2.5" that falls between accents and
+           other level 3 features.
+          </para>
+          <para>
+           If set to <literal>true</literal> and <literal>ks</literal> is set
+           to <literal>level1</literal>, will ignore accents but take case
+           into account.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kf</literal></entry>
+         <entry>
+          <literal>upper</literal>, <literal>lower</literal>,
+          <literal>false</literal>
+         </entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>upper</literal>, upper case sorts before lower
+          case. If set to <literal>lower</literal>, lower case sorts before
+          upper case. If set to <literal>false</literal>, it depends on the
+          locale.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kn</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>true</literal>, numbers within a string are
+          treated as a single numeric value rather than a sequence of
+          digits. For example, <literal>'id-45'</literal> sorts before
+          <literal>'id-123'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kr</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>,
+          <literal>digit</literal>, <replaceable>script-id</replaceable>
+         </entry>
+         <entry></entry>
+         <entry>
+          <para>
+           Set to one or more of the valid values, or any BCP 47
+           <replaceable>script-id</replaceable>, e.g. <literal>latn</literal>
+           ("Latin") or <literal>grek</literal> ("Greek"). Multiple values are
+           separated by "<literal>-</literal>".
+          </para>
+          <para>
+           Redefines the ordering of classes of characters; those characters
+           belonging to a class earlier in the list sort before characters
+           belonging to a class later in the list. For instance, the value
+           <literal>digit-currency-space</literal> (as part of a language tag
+           like <literal>und-u-kr-digit-currency-space</literal>) sorts
+           punctuation before digits and spaces.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kv</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>
+         </entry>
+         <entry><literal>punct</literal></entry>
+         <entry>
+          Classes of characters ignored during comparison at level 3. Setting
+          to a later value includes earlier values;
+          e.g. <literal>symbol</literal> also includes
+          <literal>punct</literal> and <literal>space</literal> in the
+          characters to be ignored. Key <literal>ka</literal> must be set to
+          <literal>shifted</literal> and key <literal>ks</literal> must be set
+          to <literal>level3</literal> or lower to take effect.
+         </entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+      Defaults may depend on locale. The above table is not meant to be
+      complete. See <xref linkend="icu-external-references"/> for additinal
+      options and details.
+    </para>
+   </sect3>
+   <sect3 id="icu-locale-examples">
+    <title>Examples</title>
+    <para>
+     <variablelist>
+      <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
+       <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
+       <listitem>
+        <para>German collation with phone book collation type</para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
+       <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
+       <listitem>
+        <para>
+         Root collation with Emoji collation type, per Unicode Technical Standard #51
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
+       <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
+       <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
+       <listitem>
+        <para>
+         Sort upper-case letters before lower-case letters.  (The default is
+         lower-case letters first.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
+       <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Combines both of the above options.
+        </para>
+       </listitem>
+      </varlistentry>
+     </variablelist>
+    </para>
+   </sect3>
+   <sect3 id="icu-external-references">
+    <title>External References for ICU</title>
+    <para>
+     This section (<xref linkend="icu-custom-collations"/>) is only a brief
+     overview of ICU behavior and language tags. Refer to the following
+     documents for technical details, additional options, and new behavior:
+    </para>
+    <itemizedlist>
+     <listitem>
+      <para>
+       <ulink
+           url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
+       Technical Standard #35</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
+       repository</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink>
+      </para>
+     </listitem>
+    </itemizedlist>
+   </sect3>
+  </sect2>
  </sect1>
 
  <sect1 id="multibyte">
-- 
2.34.1

From b548a671ad02a5c851a4984db6e4535a0b70f881 Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <peter@eisentraut.org>
Date: Thu, 11 May 2023 13:02:02 +0200
Subject: [PATCH] Remove read-only server settings lc_collate and lc_ctype

The GUC settings lc_collate and lc_ctype are from a time when those
locale settings were cluster-global.  When those locale settings were
made per-database (PG 8.4), the settings were kept as read-only.  As
of PG 15, you can use ICU as the per-database locale provider, so
examining these settings is already meaningless, since you need to
look into pg_database to find out what is really happening.

Discussion: https://www.postgresql.org/message-id/696054d1-bc88-b6ab-129a-18b8bce6a6f0@enterprisedb.com
---
 contrib/citext/expected/citext_utf8.out       |  4 +--
 contrib/citext/expected/citext_utf8_1.out     |  4 +--
 contrib/citext/sql/citext_utf8.sql            |  4 +--
 doc/src/sgml/config.sgml                      | 32 -------------------
 src/backend/utils/init/postinit.c             |  4 ---
 src/backend/utils/misc/guc_tables.c           | 26 ---------------
 .../regress/expected/collate.icu.utf8.out     |  4 +--
 .../regress/expected/collate.linux.utf8.out   |  6 ++--
 .../expected/collate.windows.win1252.out      |  6 ++--
 src/test/regress/sql/collate.icu.utf8.sql     |  4 +--
 src/test/regress/sql/collate.linux.utf8.sql   |  6 ++--
 .../regress/sql/collate.windows.win1252.sql   |  6 ++--
 12 files changed, 22 insertions(+), 84 deletions(-)

diff --git a/contrib/citext/expected/citext_utf8.out b/contrib/citext/expected/citext_utf8.out
index 77b4586d8f..6630e09a4d 100644
--- a/contrib/citext/expected/citext_utf8.out
+++ b/contrib/citext/expected/citext_utf8.out
@@ -8,8 +8,8 @@
  * to the "tr-TR-x-icu" collation where it will succeed.
  */
 SELECT getdatabaseencoding() <> 'UTF8' OR
-       current_setting('lc_ctype') = 'C' OR
-       (SELECT datlocprovider='i' FROM pg_database
+       (SELECT (datlocprovider = 'c' AND datctype = 'C') OR datlocprovider = 'i'
+        FROM pg_database
         WHERE datname=current_database())
        AS skip_test \gset
 \if :skip_test
diff --git a/contrib/citext/expected/citext_utf8_1.out b/contrib/citext/expected/citext_utf8_1.out
index d1e1fe1a9d..3caa7a00d4 100644
--- a/contrib/citext/expected/citext_utf8_1.out
+++ b/contrib/citext/expected/citext_utf8_1.out
@@ -8,8 +8,8 @@
  * to the "tr-TR-x-icu" collation where it will succeed.
  */
 SELECT getdatabaseencoding() <> 'UTF8' OR
-       current_setting('lc_ctype') = 'C' OR
-       (SELECT datlocprovider='i' FROM pg_database
+       (SELECT (datlocprovider = 'c' AND datctype = 'C') OR datlocprovider = 'i'
+        FROM pg_database
         WHERE datname=current_database())
        AS skip_test \gset
 \if :skip_test
diff --git a/contrib/citext/sql/citext_utf8.sql b/contrib/citext/sql/citext_utf8.sql
index 8530c68dd7..1f51df134b 100644
--- a/contrib/citext/sql/citext_utf8.sql
+++ b/contrib/citext/sql/citext_utf8.sql
@@ -9,8 +9,8 @@
  */
 
 SELECT getdatabaseencoding() <> 'UTF8' OR
-       current_setting('lc_ctype') = 'C' OR
-       (SELECT datlocprovider='i' FROM pg_database
+       (SELECT (datlocprovider = 'c' AND datctype = 'C') OR datlocprovider = 'i'
+        FROM pg_database
         WHERE datname=current_database())
        AS skip_test \gset
 \if :skip_test
diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index 909a3f28c7..3e9030e3d7 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -10788,38 +10788,6 @@ <title>Preset Options</title>
       </listitem>
      </varlistentry>
 
-     <varlistentry id="guc-lc-collate" xreflabel="lc_collate">
-      <term><varname>lc_collate</varname> (<type>string</type>)
-      <indexterm>
-       <primary><varname>lc_collate</varname> configuration parameter</primary>
-      </indexterm>
-      </term>
-      <listitem>
-       <para>
-        Reports the locale in which sorting of textual data is done.
-        See <xref linkend="locale"/> for more information.
-        This value is determined when a database is created.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="guc-lc-ctype" xreflabel="lc_ctype">
-      <term><varname>lc_ctype</varname> (<type>string</type>)
-      <indexterm>
-       <primary><varname>lc_ctype</varname> configuration parameter</primary>
-      </indexterm>
-      </term>
-      <listitem>
-       <para>
-        Reports the locale that determines character classifications.
-        See <xref linkend="locale"/> for more information.
-        This value is determined when a database is created.
-        Ordinarily this will be the same as <varname>lc_collate</varname>,
-        but for special applications it might be set differently.
-       </para>
-      </listitem>
-     </varlistentry>
-
      <varlistentry id="guc-max-function-args" xreflabel="max_function_args">
       <term><varname>max_function_args</varname> (<type>integer</type>)
       <indexterm>
diff --git a/src/backend/utils/init/postinit.c b/src/backend/utils/init/postinit.c
index 53420f4974..df81e35eb8 100644
--- a/src/backend/utils/init/postinit.c
+++ b/src/backend/utils/init/postinit.c
@@ -483,10 +483,6 @@ CheckMyDatabase(const char *name, bool am_superuser, bool override_allow_connect
 							 quote_identifier(name))));
 	}
 
-	/* Make the locale settings visible as GUC variables, too */
-	SetConfigOption("lc_collate", collate, PGC_INTERNAL, PGC_S_DYNAMIC_DEFAULT);
-	SetConfigOption("lc_ctype", ctype, PGC_INTERNAL, PGC_S_DYNAMIC_DEFAULT);
-
 	ReleaseSysCache(tup);
 }
 
diff --git a/src/backend/utils/misc/guc_tables.c b/src/backend/utils/misc/guc_tables.c
index 5f90aecd47..23d4b38e72 100644
--- a/src/backend/utils/misc/guc_tables.c
+++ b/src/backend/utils/misc/guc_tables.c
@@ -563,8 +563,6 @@ static char *syslog_ident_str;
 static double phony_random_seed;
 static char *client_encoding_string;
 static char *datestyle_string;
-static char *locale_collate;
-static char *locale_ctype;
 static char *server_encoding_string;
 static char *server_version_string;
 static int	server_version_num;
@@ -4050,30 +4048,6 @@ struct config_string ConfigureNamesString[] =
 		NULL, NULL, NULL
 	},
 
-	/* See main.c about why defaults for LC_foo are not all alike */
-
-	{
-		{"lc_collate", PGC_INTERNAL, PRESET_OPTIONS,
-			gettext_noop("Shows the collation order locale."),
-			NULL,
-			GUC_NOT_IN_SAMPLE | GUC_DISALLOW_IN_FILE
-		},
-		&locale_collate,
-		"C",
-		NULL, NULL, NULL
-	},
-
-	{
-		{"lc_ctype", PGC_INTERNAL, PRESET_OPTIONS,
-			gettext_noop("Shows the character classification and case conversion locale."),
-			NULL,
-			GUC_NOT_IN_SAMPLE | GUC_DISALLOW_IN_FILE
-		},
-		&locale_ctype,
-		"C",
-		NULL, NULL, NULL
-	},
-
 	{
 		{"lc_messages", PGC_SUSET, CLIENT_CONN_LOCALE,
 			gettext_noop("Sets the language in which messages are displayed."),
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..21840815c9 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1023,7 +1023,7 @@ SET client_min_messages TO WARNING;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT daticulocale FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
@@ -1031,7 +1031,7 @@ ERROR:  collation "test0" already exists
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (provider = icu, locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT daticulocale FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 RESET client_min_messages;
diff --git a/src/test/regress/expected/collate.linux.utf8.out b/src/test/regress/expected/collate.linux.utf8.out
index 6d34667ceb..01664f7c1b 100644
--- a/src/test/regress/expected/collate.linux.utf8.out
+++ b/src/test/regress/expected/collate.linux.utf8.out
@@ -1027,7 +1027,7 @@ CREATE SCHEMA test_schema;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
@@ -1039,9 +1039,9 @@ NOTICE:  collation "test0" for encoding "UTF8" already exists, skipping
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
-          quote_literal(current_setting('lc_collate')) ||
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) ||
           ', lc_ctype = ' ||
-          quote_literal(current_setting('lc_ctype')) || ');';
+          quote_literal((SELECT datctype FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
diff --git a/src/test/regress/expected/collate.windows.win1252.out b/src/test/regress/expected/collate.windows.win1252.out
index 61b421161f..b7b93959de 100644
--- a/src/test/regress/expected/collate.windows.win1252.out
+++ b/src/test/regress/expected/collate.windows.win1252.out
@@ -863,7 +863,7 @@ CREATE SCHEMA test_schema;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
@@ -875,9 +875,9 @@ NOTICE:  collation "test0" for encoding "WIN1252" already exists, skipping
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
-          quote_literal(current_setting('lc_collate')) ||
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) ||
           ', lc_ctype = ' ||
-          quote_literal(current_setting('lc_ctype')) || ');';
+          quote_literal((SELECT datctype FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..c9c2ab8fa6 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -362,14 +362,14 @@ CREATE SCHEMA test_schema;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (provider = icu, locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT daticulocale FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (provider = icu, locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT daticulocale FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 
diff --git a/src/test/regress/sql/collate.linux.utf8.sql b/src/test/regress/sql/collate.linux.utf8.sql
index 2b787507c5..132d13af0a 100644
--- a/src/test/regress/sql/collate.linux.utf8.sql
+++ b/src/test/regress/sql/collate.linux.utf8.sql
@@ -359,7 +359,7 @@ CREATE SCHEMA test_schema;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
@@ -368,9 +368,9 @@ CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
-          quote_literal(current_setting('lc_collate')) ||
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) ||
           ', lc_ctype = ' ||
-          quote_literal(current_setting('lc_ctype')) || ');';
+          quote_literal((SELECT datctype FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
diff --git a/src/test/regress/sql/collate.windows.win1252.sql b/src/test/regress/sql/collate.windows.win1252.sql
index b5c45e1810..353d769a5b 100644
--- a/src/test/regress/sql/collate.windows.win1252.sql
+++ b/src/test/regress/sql/collate.windows.win1252.sql
@@ -310,7 +310,7 @@ CREATE SCHEMA test_schema;
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test0 (locale = ' ||
-          quote_literal(current_setting('lc_collate')) || ');';
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
@@ -319,9 +319,9 @@ CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
-          quote_literal(current_setting('lc_collate')) ||
+          quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) ||
           ', lc_ctype = ' ||
-          quote_literal(current_setting('lc_ctype')) || ');';
+          quote_literal((SELECT datctype FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
-- 
2.40.0

From fc66f02976bb11b629bcf71346c2858eccbcf1a3 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 11 May 2023 10:36:04 -0700
Subject: [PATCH v5 1/7] For user-defined collations, never set
 collencoding=-1.

For new user-defined collations, always set collencoding to the
current database encoding so that it is never shadowed by a built-in
collation.

Built in collations that work with any encoding may have
collencoding=-1, and if a user defines a collation with the same name,
it will shadow the built-in collation.

Previously it was possible to create an ICU collation (which was
assigned collencoding=-1) that was shadowed by a built-in collation
and completely inaccessible.
---
 src/backend/commands/collationcmds.c          | 28 +++++++++++++------
 .../regress/expected/collate.icu.utf8.out     |  2 +-
 2 files changed, 21 insertions(+), 9 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c91fe66d9b..a53700256b 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -302,16 +302,29 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 					 errmsg("ICU rules cannot be specified unless locale provider is ICU")));
 
+		/*
+		 * The collencoding is used to hide built-in collations that are
+		 * incompatible with the current database encoding, allowing users to
+		 * define a compatible collation with the same name if
+		 * desired. Built-in collations that work with any encoding have
+		 * collencoding=-1.
+		 *
+		 * A collation that's a match to the current database encoding will
+		 * shadow a collation with the same name and collencoding=-1. We never
+		 * want a user-created collation to be shadowed by a built-in
+		 * collation, so for user-created collations, always set collencoding
+		 * to the current database encoding.
+		 */
+		collencoding = GetDatabaseEncoding();
+
 		if (collprovider == COLLPROVIDER_ICU)
 		{
 #ifdef USE_ICU
 			/*
-			 * We could create ICU collations with collencoding == database
-			 * encoding, but it seems better to use -1 so that it matches the
-			 * way initdb would create ICU collations.  However, only allow
-			 * one to be created when the current database's encoding is
-			 * supported.  Otherwise the collation is useless, plus we get
-			 * surprising behaviors like not being able to drop the collation.
+			 * Only allow an ICU collation to be created when the current
+			 * database's encoding is supported.  Otherwise the collation is
+			 * useless, plus we get surprising behaviors like not being able
+			 * to drop the collation.
 			 *
 			 * Skip this test when !USE_ICU, because the error we want to
 			 * throw for that isn't thrown till later.
@@ -321,11 +334,10 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
 						 errmsg("current database's encoding is not supported with this provider")));
 #endif
-			collencoding = -1;
 		}
 		else
 		{
-			collencoding = GetDatabaseEncoding();
+			Assert(collprovider == COLLPROVIDER_LIBC);
 			check_encoding_locale_matches(collencoding, collcollate, collctype);
 		}
 	}
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index b5a221b030..9c9e1e4f48 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1062,7 +1062,7 @@ SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
 
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
-ERROR:  collation "test11" already exists in schema "collate_tests"
+ERROR:  collation "test11" for encoding "UTF8" already exists in schema "collate_tests"
 ALTER COLLATION test1 RENAME TO test22; -- fail
 ERROR:  collation "test1" for encoding "UTF8" does not exist
 ALTER COLLATION test11 OWNER TO regress_test_role;
-- 
2.34.1

From 25824dc213272c739eecd16b17a3458fc5f81339 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 28 Apr 2023 12:22:41 -0700
Subject: [PATCH v5 2/7] ICU: fix up old libc-style locale strings.

Before transforming a locale string into a language tag, fix up old
libc-style locale strings such as 'fr_FR@euro'. Older ICU versions did
this automatically, but ICU version 64 removed that support.

Discussion: https://postgr.es/m/654a49f7ff7461bcf47be4181430678d45f93858.camel%40j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 59 ++++++++++++++++-
 src/bin/initdb/initdb.c                       | 63 ++++++++++++++++++-
 .../regress/expected/collate.icu.utf8.out     | 11 ++++
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +++
 4 files changed, 138 insertions(+), 2 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index f0b6567da1..e7b166461b 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2766,6 +2766,60 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 	pfree(lower_str);
 }
 
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = palloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pstrdup(loc_str);
+}
+
 #endif
 
 /*
@@ -2782,6 +2836,7 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
@@ -2814,7 +2869,7 @@ icu_language_tag(const char *loc_str, int elevel)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2834,6 +2889,8 @@ icu_language_tag(const char *loc_str, int elevel)
 		break;
 	}
 
+	pfree(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pfree(langtag);
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2c208ead01..2b5cc30955 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2229,6 +2229,64 @@ check_icu_locale_encoding(int user_enc)
 	return true;
 }
 
+#ifdef USE_ICU
+
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+#define ICU_VARIANT_MAP_SIZE \
+	(sizeof(icu_variant_map)/sizeof(icu_variant_map[0]))
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < ICU_VARIANT_MAP_SIZE; i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = pg_malloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pg_strdup(loc_str);
+}
+
+#endif
+
 /*
  * Convert to canonical BCP47 language tag. Must be consistent with
  * icu_language_tag().
@@ -2238,6 +2296,7 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
@@ -2268,7 +2327,7 @@ icu_language_tag(const char *loc_str)
 		int32_t		len;
 
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		len = uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/*
 		 * If the result fits in the buffer exactly (len == buflen),
@@ -2287,6 +2346,8 @@ icu_language_tag(const char *loc_str)
 		break;
 	}
 
+	pg_free(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pg_free(langtag);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 9c9e1e4f48..e0f11e3cd4 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1042,13 +1042,24 @@ ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
+ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
+WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 RESET icu_validation_level;
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-pinyin" for locale "@pinyin"
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-stroke" for locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 85e26951b6..8d5423bc17 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -378,11 +378,18 @@ RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 SET icu_validation_level = WARNING;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 RESET icu_validation_level;
 
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
 
-- 
2.34.1

From cd839f069cc09a71788bafa28730e4caf8f9d768 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 10 May 2023 10:47:16 -0700
Subject: [PATCH v5 3/7] Reduce icu_validation_level default to WARNING.

---
 doc/src/sgml/config.sgml                       | 2 +-
 src/backend/utils/adt/pg_locale.c              | 2 +-
 src/backend/utils/misc/guc_tables.c            | 2 +-
 src/backend/utils/misc/postgresql.conf.sample  | 2 +-
 src/test/regress/expected/collate.icu.utf8.out | 4 ++--
 src/test/regress/sql/collate.icu.utf8.sql      | 4 ++--
 6 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index b56f073a91..c4a9dcb9ae 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -9840,7 +9840,7 @@ SET XML OPTION { DOCUMENT | CONTENT };
        <para>
         If set to <literal>DISABLED</literal>, does not report validation
         problems at all. Otherwise reports problems at the given message
-        level. The default is <literal>ERROR</literal>.
+        level. The default is <literal>WARNING</literal>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index e7b166461b..bb4a8d84f6 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -96,7 +96,7 @@ char	   *locale_monetary;
 char	   *locale_numeric;
 char	   *locale_time;
 
-int			icu_validation_level = ERROR;
+int			icu_validation_level = WARNING;
 
 /*
  * lc_time localization cache.
diff --git a/src/backend/utils/misc/guc_tables.c b/src/backend/utils/misc/guc_tables.c
index 2f42cebaf6..8c843f4ab6 100644
--- a/src/backend/utils/misc/guc_tables.c
+++ b/src/backend/utils/misc/guc_tables.c
@@ -4689,7 +4689,7 @@ struct config_enum ConfigureNamesEnum[] =
 		 NULL
 		},
 		&icu_validation_level,
-		ERROR, icu_validation_level_options,
+		WARNING, icu_validation_level_options,
 		NULL, NULL, NULL
 	},
 
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index b70c66ca87..87bad8ecbf 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -734,7 +734,7 @@
 #lc_numeric = 'C'			# locale for number formatting
 #lc_time = 'C'				# locale for time formatting
 
-#icu_validation_level = ERROR		# report ICU locale validation
+#icu_validation_level = WARNING		# report ICU locale validation
 					# errors at the given level
 
 # default configuration for text search
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index e0f11e3cd4..12afc3b65a 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1037,6 +1037,7 @@ $$;
 RESET client_min_messages;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
+SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
@@ -1044,7 +1045,7 @@ CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-SET icu_validation_level = WARNING;
+RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
@@ -1052,7 +1053,6 @@ WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
-RESET icu_validation_level;
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
 NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 8d5423bc17..655c965f46 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -376,14 +376,14 @@ $$;
 RESET client_min_messages;
 
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
+SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
-SET icu_validation_level = WARNING;
+RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
-RESET icu_validation_level;
 
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
-- 
2.34.1

From a13a15988ab2e991e42569b8b1e0cd1d6e940baf Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 1 May 2023 15:38:29 -0700
Subject: [PATCH v5 4/7] Introduce collation provider "none".

Provides locale-unaware semantics that are implemented as fast byte
operations in Postgres, independent of the operating system or any
provider libraries.

Equivalent (in semantics and implementation) to the libc provider with
locale "C", except that LC_COLLATE and LC_CTYPE can be set
independently.

Use provider "none" for built-in collation "ucs_basic" instead of
libc.
---
 doc/src/sgml/charset.sgml              | 87 +++++++++++++++++++++-----
 doc/src/sgml/ref/create_collation.sgml |  2 +-
 doc/src/sgml/ref/create_database.sgml  |  2 +-
 doc/src/sgml/ref/createdb.sgml         |  2 +-
 doc/src/sgml/ref/initdb.sgml           |  2 +-
 src/backend/catalog/pg_collation.c     |  7 ++-
 src/backend/commands/collationcmds.c   | 84 ++++++++++++++++++++-----
 src/backend/commands/dbcommands.c      | 69 +++++++++++++++++---
 src/backend/utils/adt/pg_locale.c      | 27 +++++++-
 src/backend/utils/init/postinit.c      | 10 ++-
 src/bin/initdb/initdb.c                | 33 +++++++++-
 src/bin/initdb/t/001_initdb.pl         | 29 +++++++++
 src/bin/pg_dump/pg_dump.c              |  8 ++-
 src/bin/pg_upgrade/t/002_pg_upgrade.pl | 18 +++++-
 src/bin/psql/describe.c                |  2 +-
 src/bin/scripts/createdb.c             |  2 +-
 src/bin/scripts/t/020_createdb.pl      | 29 +++++++++
 src/include/catalog/pg_collation.dat   |  3 +-
 src/include/catalog/pg_collation.h     |  3 +
 src/test/regress/expected/collate.out  | 10 ++-
 src/test/regress/sql/collate.sql       |  6 ++
 21 files changed, 372 insertions(+), 63 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 6dd95b8966..de7c65ae35 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -342,22 +342,14 @@ initdb --locale=sv_SE
    <title>Locale Providers</title>
 
    <para>
-    <productname>PostgreSQL</productname> supports multiple <firstterm>locale
-    providers</firstterm>.  This specifies which library supplies the locale
-    data.  One standard provider name is <literal>libc</literal>, which uses
-    the locales provided by the operating system C library.  These are the
-    locales used by most tools provided by the operating system.  Another
-    provider is <literal>icu</literal>, which uses the external
-    ICU<indexterm><primary>ICU</primary></indexterm> library.  ICU locales can
-    only be used if support for ICU was configured when PostgreSQL was built.
+    A locale provider specifies which library defines the locale behavior for
+    collations and character classifications.
    </para>
 
    <para>
     The commands and tools that select the locale settings, as described
-    above, each have an option to select the locale provider.  The examples
-    shown earlier all use the <literal>libc</literal> provider, which is the
-    default.  Here is an example to initialize a database cluster using the
-    ICU provider:
+    above, each have an option to select the locale provider. Here is an
+    example to initialize a database cluster using the ICU provider:
 <programlisting>
 initdb --locale-provider=icu --icu-locale=en
 </programlisting>
@@ -370,12 +362,73 @@ initdb --locale-provider=icu --icu-locale=en
    </para>
 
    <para>
-    Which locale provider to use depends on individual requirements.  For most
-    basic uses, either provider will give adequate results.  For the libc
-    provider, it depends on what the operating system offers; some operating
-    systems are better than others.  For advanced uses, ICU offers more locale
-    variants and customization options.
+    Regardless of the locale provider, the operating system is still used to
+    provide some locale-aware behavior, such as messages (see <xref
+    linkend="guc-lc-messages"/>).
    </para>
+
+   <para>
+    The available locale providers are listed below.
+   </para>
+
+   <sect3 id="locale-provider-none">
+    <title>None</title>
+    <para>
+     The <literal>none</literal> provider uses simple built-in operations
+     which are not locale-aware.
+    </para>
+    <para>
+     The collation and character classification behavior is equivalent to
+     using the <literal>libc</literal> provider with locale
+     <literal>C</literal>, except that <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently.
+    </para>
+    <note>
+     <para>
+      When using the <literal>none</literal> locale provider, behavior may
+      depend on the database encoding.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-icu">
+    <title>ICU</title>
+    <para>
+     The <literal>icu</literal> provider uses the external
+     ICU<indexterm><primary>ICU</primary></indexterm>
+     library. <productname>PostgreSQL</productname> must have been configured
+     with support.
+    </para>
+    <para>
+     ICU provides collation and character classification behavior that is
+     independent of the operating system and database encoding, which is
+     preferable if you expect to transition to other platforms without any
+     change in results. <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
+    </para>
+    <note>
+     <para>
+      For the ICU provider, results may depend on the version of the ICU
+      library used, as it is updated to reflect changes in natural language
+      over time.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-libc">
+    <title>libc</title>
+    <para>
+     The <literal>libc</literal> provider uses the operating system's C
+     library. The collation and character classification behavior is
+     controlled by the settings <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal>, so they cannot be set independently.
+    </para>
+    <note>
+     <para>
+      The same locale name may have different behavior on different platforms
+      when using the libc provider.
+     </para>
+    </note>
+   </sect3>
+
   </sect2>
 
   <sect2 id="locale-problems">
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..5489ae7413 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -120,7 +120,7 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
      <listitem>
       <para>
        Specifies the provider to use for locale services associated with this
-       collation.  Possible values are
+       collation.  Possible values are <literal>none</literal>,
        <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
        (if the server was built with ICU support) or <literal>libc</literal>.
        <literal>libc</literal> is the default.  See <xref
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..60b9da0952 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -212,7 +212,7 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <listitem>
        <para>
         Specifies the provider to use for the default collation in this
-        database.  Possible values are
+        database.  Possible values are <literal>none</literal>,
         <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
         (if the server was built with ICU support) or <literal>libc</literal>.
         By default, the provider is the same as that of the <xref
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..326a371d34 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -168,7 +168,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry>
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>none</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         Specifies the locale provider for the database's default collation.
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..e604ab48b7 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -323,7 +323,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry id="app-initdb-option-locale-provider">
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>none</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         This option sets the locale provider for databases created in the new
diff --git a/src/backend/catalog/pg_collation.c b/src/backend/catalog/pg_collation.c
index fd022e6fc2..86b6ba2375 100644
--- a/src/backend/catalog/pg_collation.c
+++ b/src/backend/catalog/pg_collation.c
@@ -68,7 +68,12 @@ CollationCreate(const char *collname, Oid collnamespace,
 	Assert(collname);
 	Assert(collnamespace);
 	Assert(collowner);
-	Assert((collcollate && collctype) || colliculocale);
+	Assert((collprovider == COLLPROVIDER_NONE &&
+			!collcollate && !collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_LIBC &&
+			 collcollate &&  collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_ICU &&
+			!collcollate && !collctype &&  colliculocale));
 
 	/*
 	 * Make sure there is no existing collation of same name & encoding.
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index a53700256b..267a551818 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -215,7 +215,9 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 
 		if (collproviderstr)
 		{
-			if (pg_strcasecmp(collproviderstr, "icu") == 0)
+			if (pg_strcasecmp(collproviderstr, "none") == 0)
+				collprovider = COLLPROVIDER_NONE;
+			else if (pg_strcasecmp(collproviderstr, "icu") == 0)
 				collprovider = COLLPROVIDER_ICU;
 			else if (pg_strcasecmp(collproviderstr, "libc") == 0)
 				collprovider = COLLPROVIDER_LIBC;
@@ -228,6 +230,13 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		else
 			collprovider = COLLPROVIDER_LIBC;
 
+		if (collprovider == COLLPROVIDER_NONE
+			&& (localeEl || lccollateEl || lcctypeEl))
+		{
+			ereport(ERROR,
+					(errmsg("collation provider \"none\" does not support LOCALE, LC_COLLATE, or LC_CTYPE")));
+		}
+
 		if (localeEl)
 		{
 			if (collprovider == COLLPROVIDER_LIBC)
@@ -317,7 +326,15 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		 */
 		collencoding = GetDatabaseEncoding();
 
-		if (collprovider == COLLPROVIDER_ICU)
+		if (collprovider == COLLPROVIDER_NONE)
+		{
+			/*
+			 * The "none" provider works with all encodings, so no checking is
+			 * required. NB: the behavior may be different for different
+			 * encodings, though.
+			 */
+		}
+		else if (collprovider == COLLPROVIDER_ICU)
 		{
 #ifdef USE_ICU
 			/*
@@ -343,7 +360,18 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 	}
 
 	if (!collversion)
-		collversion = get_collation_actual_version(collprovider, collprovider == COLLPROVIDER_ICU ? colliculocale : collcollate);
+	{
+		char *locale;
+
+		if (collprovider == COLLPROVIDER_ICU)
+			locale = colliculocale;
+		else if (collprovider == COLLPROVIDER_LIBC)
+			locale = collcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		collversion = get_collation_actual_version(collprovider, locale);
+	}
 
 	newoid = CollationCreate(collName,
 							 collNamespace,
@@ -418,6 +446,7 @@ AlterCollation(AlterCollationStmt *stmt)
 	Form_pg_collation collForm;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 	ObjectAddress address;
@@ -442,8 +471,20 @@ AlterCollation(AlterCollationStmt *stmt)
 	datum = SysCacheGetAttr(COLLOID, tup, Anum_pg_collation_collversion, &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = SysCacheGetAttrNotNull(COLLOID, tup, collForm->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
-	newversion = get_collation_actual_version(collForm->collprovider, TextDatumGetCString(datum));
+	if (collForm->collprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_colliculocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (collForm->collprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_collcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	newversion = get_collation_actual_version(collForm->collprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -506,11 +547,18 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_database) GETSTRUCT(dbtup))->datlocprovider;
 
-		datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_daticulocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_datcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
 
 		ReleaseSysCache(dbtup);
 	}
@@ -526,11 +574,19 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_collation) GETSTRUCT(colltp))->collprovider;
 		Assert(provider != COLLPROVIDER_DEFAULT);
-		datum = SysCacheGetAttrNotNull(COLLOID, colltp,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_colliculocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_collcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
 
 		ReleaseSysCache(colltp);
 	}
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 2e242eeff2..9e73f54803 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -909,7 +909,9 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	{
 		char	   *locproviderstr = defGetString(dlocprovider);
 
-		if (pg_strcasecmp(locproviderstr, "icu") == 0)
+		if (pg_strcasecmp(locproviderstr, "none") == 0)
+			dblocprovider = COLLPROVIDER_NONE;
+		else if (pg_strcasecmp(locproviderstr, "icu") == 0)
 			dblocprovider = COLLPROVIDER_ICU;
 		else if (pg_strcasecmp(locproviderstr, "libc") == 0)
 			dblocprovider = COLLPROVIDER_LIBC;
@@ -1177,9 +1179,17 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 */
 	if (src_collversion && !dcollversion)
 	{
-		char	   *actual_versionstr;
+		char	*actual_versionstr;
+		char	*locale;
 
-		actual_versionstr = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		actual_versionstr = get_collation_actual_version(dblocprovider, locale);
 		if (!actual_versionstr)
 			ereport(ERROR,
 					(errmsg("template database \"%s\" has a collation version, but no actual collation version could be determined",
@@ -1207,7 +1217,18 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 * collation version, which is normally only the case for template0.
 	 */
 	if (dbcollversion == NULL)
-		dbcollversion = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+	{
+		char *locale;
+
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		dbcollversion = get_collation_actual_version(dblocprovider, locale);
+	}
 
 	/* Resolve default tablespace for new database */
 	if (dtablespacename && dtablespacename->arg)
@@ -2403,6 +2424,7 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	ObjectAddress address;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 
@@ -2429,10 +2451,24 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	datum = heap_getattr(tuple, Anum_pg_database_datcollversion, RelationGetDescr(rel), &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = heap_getattr(tuple, datForm->datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
-	if (isnull)
-		elog(ERROR, "unexpected null in pg_database");
-	newversion = get_collation_actual_version(datForm->datlocprovider, TextDatumGetCString(datum));
+	if (datForm->datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_daticulocale, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datForm->datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	newversion = get_collation_actual_version(datForm->datlocprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -2617,6 +2653,7 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 	HeapTuple	tp;
 	char		datlocprovider;
 	Datum		datum;
+	char	   *locale;
 	char	   *version;
 
 	tp = SearchSysCache1(DATABASEOID, ObjectIdGetDatum(dbid));
@@ -2627,8 +2664,20 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 
 	datlocprovider = ((Form_pg_database) GETSTRUCT(tp))->datlocprovider;
 
-	datum = SysCacheGetAttrNotNull(DATABASEOID, tp, datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-	version = get_collation_actual_version(datlocprovider, TextDatumGetCString(datum));
+	if (datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_daticulocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_datcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	version = get_collation_actual_version(datlocprovider, locale);
 
 	ReleaseSysCache(tp);
 
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index bb4a8d84f6..5ac5036f05 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1228,7 +1228,12 @@ lookup_collation_cache(Oid collation, bool set_flags)
 			elog(ERROR, "cache lookup failed for collation %u", collation);
 		collform = (Form_pg_collation) GETSTRUCT(tp);
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_NONE)
+		{
+			cache_entry->collate_is_c = true;
+			cache_entry->ctype_is_c = true;
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 			Datum		datum;
 			const char *collcollate;
@@ -1281,6 +1286,9 @@ lc_collate_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_NONE)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1334,6 +1342,9 @@ lc_ctype_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_NONE)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1487,8 +1498,10 @@ pg_newlocale_from_collation(Oid collid)
 	{
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return &default_locale;
-		else
+		else if (default_locale.provider == COLLPROVIDER_LIBC)
 			return (pg_locale_t) 0;
+		else
+			elog(ERROR, "cannot open collation with provider \"none\"");
 	}
 
 	cache_entry = lookup_collation_cache(collid, false);
@@ -1513,7 +1526,11 @@ pg_newlocale_from_collation(Oid collid)
 		result.provider = collform->collprovider;
 		result.deterministic = collform->collisdeterministic;
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_NONE)
+		{
+			elog(ERROR, "cannot open collation with provider \"none\"");
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 #ifdef HAVE_LOCALE_T
 			const char *collcollate;
@@ -1599,6 +1616,7 @@ pg_newlocale_from_collation(Oid collid)
 
 			collversionstr = TextDatumGetCString(datum);
 
+			Assert(collform->collprovider != COLLPROVIDER_NONE);
 			datum = SysCacheGetAttrNotNull(COLLOID, tp, collform->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
 			actual_versionstr = get_collation_actual_version(collform->collprovider,
@@ -1650,6 +1668,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (collprovider == COLLPROVIDER_NONE)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
diff --git a/src/backend/utils/init/postinit.c b/src/backend/utils/init/postinit.c
index 53420f4974..8053642fd3 100644
--- a/src/backend/utils/init/postinit.c
+++ b/src/backend/utils/init/postinit.c
@@ -461,10 +461,18 @@ CheckMyDatabase(const char *name, bool am_superuser, bool override_allow_connect
 	{
 		char	   *actual_versionstr;
 		char	   *collversionstr;
+		char	   *locale;
 
 		collversionstr = TextDatumGetCString(datum);
 
-		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, dbform->datlocprovider == COLLPROVIDER_ICU ? iculocale : collate);
+		if (dbform->datlocprovider == COLLPROVIDER_ICU)
+			locale = iculocale;
+		else if (dbform->datlocprovider == COLLPROVIDER_LIBC)
+			locale = collate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, locale);
 		if (!actual_versionstr)
 			/* should not happen */
 			elog(WARNING,
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 2b5cc30955..4cf6892bee 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2469,6 +2469,22 @@ setlocales(void)
 
 	/* set empty lc_* values to locale config if set */
 
+	if (locale_provider == COLLPROVIDER_NONE)
+	{
+		if (!lc_ctype)
+			lc_ctype = "C";
+		if (!lc_collate)
+			lc_collate = "C";
+		if (!lc_numeric)
+			lc_numeric = "C";
+		if (!lc_time)
+			lc_time = "C";
+		if (!lc_monetary)
+			lc_monetary = "C";
+		if (!lc_messages)
+			lc_messages = "C";
+	}
+
 	if (locale)
 	{
 		if (!lc_ctype)
@@ -2563,7 +2579,7 @@ usage(const char *progname)
 			 "                            set default locale in the respective category for\n"
 			 "                            new databases (default taken from environment)\n"));
 	printf(_("      --no-locale           equivalent to --locale=C\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={none|libc|icu}\n"
 			 "                            set default locale provider for new databases\n"));
 	printf(_("      --pwfile=FILE         read password for the new superuser from file\n"));
 	printf(_("  -T, --text-search-config=CFG\n"
@@ -2713,7 +2729,15 @@ setup_locale_encoding(void)
 {
 	setlocales();
 
-	if (locale_provider == COLLPROVIDER_LIBC &&
+	if (locale_provider == COLLPROVIDER_NONE &&
+		strcmp(lc_ctype, "C") == 0 &&
+		strcmp(lc_collate, "C") == 0 &&
+		strcmp(lc_time, "C") == 0 &&
+		strcmp(lc_numeric, "C") == 0 &&
+		strcmp(lc_monetary, "C") == 0 &&
+		strcmp(lc_messages, "C") == 0)
+		printf(_("The database cluster will be initialized with no locale.\n"));
+	else if (locale_provider == COLLPROVIDER_LIBC &&
 		strcmp(lc_ctype, lc_collate) == 0 &&
 		strcmp(lc_ctype, lc_time) == 0 &&
 		strcmp(lc_ctype, lc_numeric) == 0 &&
@@ -3387,7 +3411,9 @@ main(int argc, char *argv[])
 										 "-c debug_discard_caches=1");
 				break;
 			case 15:
-				if (strcmp(optarg, "icu") == 0)
+				if (strcmp(optarg, "none") == 0)
+					locale_provider = COLLPROVIDER_NONE;
+				else if (strcmp(optarg, "icu") == 0)
 					locale_provider = COLLPROVIDER_ICU;
 				else if (strcmp(optarg, "libc") == 0)
 					locale_provider = COLLPROVIDER_LIBC;
@@ -3426,6 +3452,7 @@ main(int argc, char *argv[])
 		exit(1);
 	}
 
+
 	if (icu_locale && locale_provider != COLLPROVIDER_ICU)
 		pg_fatal("%s cannot be specified unless locale provider \"%s\" is chosen",
 				 "--icu-locale", "icu");
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index 17a444d80c..fe6d224e5b 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -154,6 +154,35 @@ else
 		'locale provider ICU fails since no ICU support');
 }
 
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', "$tempdir/data6" ],
+	'locale provider none');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--locale=C',
+	  "$tempdir/data7" ],
+	'locale provider none with --locale');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--lc-collate=C',
+	  "$tempdir/data8" ],
+	'locale provider none with --lc-collate');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--lc-ctype=C',
+	  "$tempdir/data9" ],
+	'locale provider none with --lc-ctype');
+
+command_fails(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--icu-locale=en',
+	  "$tempdir/dataX" ],
+	'fails for locale provider none with ICU locale');
+
+command_fails(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--icu-rules=""',
+	  "$tempdir/dataX" ],
+	'fails for locale provider none with ICU rules');
+
 command_fails(
 	[ 'initdb', '--no-sync', '--locale-provider=xyz', "$tempdir/dataX" ],
 	'fails for invalid locale provider');
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index 41a51ec5cd..be6580ab3c 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -3070,7 +3070,9 @@ dumpDatabase(Archive *fout)
 	}
 
 	appendPQExpBufferStr(creaQry, " LOCALE_PROVIDER = ");
-	if (datlocprovider[0] == 'c')
+	if (datlocprovider[0] == 'n')
+		appendPQExpBufferStr(creaQry, "none");
+	else if (datlocprovider[0] == 'c')
 		appendPQExpBufferStr(creaQry, "libc");
 	else if (datlocprovider[0] == 'i')
 		appendPQExpBufferStr(creaQry, "icu");
@@ -13446,7 +13448,9 @@ dumpCollation(Archive *fout, const CollInfo *collinfo)
 					  fmtQualifiedDumpable(collinfo));
 
 	appendPQExpBufferStr(q, "provider = ");
-	if (collprovider[0] == 'c')
+	if (collprovider[0] == 'n')
+		appendPQExpBufferStr(q, "none");
+	else if (collprovider[0] == 'c')
 		appendPQExpBufferStr(q, "libc");
 	else if (collprovider[0] == 'i')
 		appendPQExpBufferStr(q, "icu");
diff --git a/src/bin/pg_upgrade/t/002_pg_upgrade.pl b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
index 4a7895a756..6d58f6103e 100644
--- a/src/bin/pg_upgrade/t/002_pg_upgrade.pl
+++ b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
@@ -114,12 +114,20 @@ my $original_locale = "C";
 my $original_iculocale = "";
 my $provider_field = "'c' AS datlocprovider";
 my $iculocale_field = "NULL AS daticulocale";
-if ($oldnode->pg_version >= 15 && $ENV{with_icu} eq 'yes')
+if ($oldnode->pg_version >= 15)
 {
 	$provider_field = "datlocprovider";
 	$iculocale_field = "daticulocale";
-	$original_provider = "i";
-	$original_iculocale = "fr-CA";
+
+	if ($ENV{with_icu} eq 'yes')
+	{
+		$original_provider = "i";
+		$original_iculocale = "fr-CA";
+	}
+	else
+	{
+		$original_provider = "n";
+	}
 }
 
 my @initdb_params = @custom_opts;
@@ -131,6 +139,10 @@ if ($original_provider eq "i")
 	push @initdb_params, ('--locale-provider', 'icu');
 	push @initdb_params, ('--icu-locale', 'fr-CA');
 }
+elsif ($original_provider eq "n")
+{
+	push @initdb_params, ('--locale-provider', 'none');
+}
 
 $node_params{extra} = \@initdb_params;
 $oldnode->init(%node_params);
diff --git a/src/bin/psql/describe.c b/src/bin/psql/describe.c
index 058e41e749..16e726b784 100644
--- a/src/bin/psql/describe.c
+++ b/src/bin/psql/describe.c
@@ -932,7 +932,7 @@ listAllDbs(const char *pattern, bool verbose)
 					  gettext_noop("Encoding"));
 	if (pset.sversion >= 150000)
 		appendPQExpBuffer(&buf,
-						  "  CASE d.datlocprovider WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
+						  "  CASE d.datlocprovider WHEN 'n' THEN 'none' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
 						  gettext_noop("Locale Provider"));
 	else
 		appendPQExpBuffer(&buf,
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..79367d933b 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -299,7 +299,7 @@ help(const char *progname)
 	printf(_("      --lc-ctype=LOCALE        LC_CTYPE setting for the database\n"));
 	printf(_("      --icu-locale=LOCALE      ICU locale setting for the database\n"));
 	printf(_("      --icu-rules=RULES        ICU rules setting for the database\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={none|libc|icu}\n"
 			 "                               locale provider for the database's default collation\n"));
 	printf(_("  -O, --owner=OWNER            database user to own the new database\n"));
 	printf(_("  -S, --strategy=STRATEGY      database creation strategy wal_log or file_copy\n"));
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index af3b1492e3..5aa658b671 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -83,6 +83,35 @@ else
 		'create database with ICU fails since no ICU support');
 }
 
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', 'testnone1' ],
+	'create database with provider "none"');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--locale=C',
+	  'testnone2' ],
+	'create database with provider "none" and locale "C"');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--lc-collate=C',
+	  'testnone3' ],
+	'create database with provider "none" and LC_COLLATE=C');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--lc-ctype=C',
+	  'testnone4' ],
+	'create database with provider "none" and LC_CTYPE=C');
+
+$node->command_fails(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--icu-locale=en',
+	  'testnone5' ],
+	'create database with provider "none" and ICU_LOCALE="en"');
+
+$node->command_fails(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--icu-rules=""',
+	  'testnone6' ],
+	'create database with provider "none" and ICU_RULES=""');
+
 $node->command_fails([ 'createdb', 'foobar1' ],
 	'fails if database already exists');
 
diff --git a/src/include/catalog/pg_collation.dat b/src/include/catalog/pg_collation.dat
index b6a69d1d42..40d62416ea 100644
--- a/src/include/catalog/pg_collation.dat
+++ b/src/include/catalog/pg_collation.dat
@@ -24,8 +24,7 @@
   collname => 'POSIX', collprovider => 'c', collencoding => '-1',
   collcollate => 'POSIX', collctype => 'POSIX' },
 { oid => '962', descr => 'sorts by Unicode code point',
-  collname => 'ucs_basic', collprovider => 'c', collencoding => '6',
-  collcollate => 'C', collctype => 'C' },
+  collname => 'ucs_basic', collprovider => 'n', collencoding => '6' },
 { oid => '963',
   descr => 'sorts using the Unicode Collation Algorithm with default settings',
   collname => 'unicode', collprovider => 'i', collencoding => '-1',
diff --git a/src/include/catalog/pg_collation.h b/src/include/catalog/pg_collation.h
index bfa3568451..29be3f8d94 100644
--- a/src/include/catalog/pg_collation.h
+++ b/src/include/catalog/pg_collation.h
@@ -64,6 +64,7 @@ DECLARE_UNIQUE_INDEX_PKEY(pg_collation_oid_index, 3085, CollationOidIndexId, on
 
 #ifdef EXPOSE_TO_CLIENT_CODE
 
+#define COLLPROVIDER_NONE		'n'
 #define COLLPROVIDER_DEFAULT	'd'
 #define COLLPROVIDER_ICU		'i'
 #define COLLPROVIDER_LIBC		'c'
@@ -73,6 +74,8 @@ collprovider_name(char c)
 {
 	switch (c)
 	{
+		case COLLPROVIDER_NONE:
+			return "none";
 		case COLLPROVIDER_ICU:
 			return "icu";
 		case COLLPROVIDER_LIBC:
diff --git a/src/test/regress/expected/collate.out b/src/test/regress/expected/collate.out
index 0649564485..b7603c9f6c 100644
--- a/src/test/regress/expected/collate.out
+++ b/src/test/regress/expected/collate.out
@@ -650,6 +650,13 @@ EXPLAIN (COSTS OFF)
 (3 rows)
 
 -- CREATE/DROP COLLATION
+CREATE COLLATION none ( PROVIDER = none );
+CREATE COLLATION none2 ( PROVIDER = none, LOCALE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
+CREATE COLLATION none2 ( PROVIDER = none, LC_CTYPE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
+CREATE COLLATION none2 ( PROVIDER = none, LC_COLLATE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
@@ -754,7 +761,7 @@ DETAIL:  FROM cannot be specified together with any other options.
 -- must get rid of them.
 --
 DROP SCHEMA collate_tests CASCADE;
-NOTICE:  drop cascades to 19 other objects
+NOTICE:  drop cascades to 20 other objects
 DETAIL:  drop cascades to table collate_test1
 drop cascades to table collate_test_like
 drop cascades to table collate_test2
@@ -771,6 +778,7 @@ drop cascades to function dup(anyelement)
 drop cascades to table collate_test20
 drop cascades to table collate_test21
 drop cascades to table collate_test22
+drop cascades to collation "none"
 drop cascades to collation mycoll2
 drop cascades to table collate_test23
 drop cascades to view collate_on_int
diff --git a/src/test/regress/sql/collate.sql b/src/test/regress/sql/collate.sql
index c3d40fc195..e2dceb8dff 100644
--- a/src/test/regress/sql/collate.sql
+++ b/src/test/regress/sql/collate.sql
@@ -244,6 +244,12 @@ EXPLAIN (COSTS OFF)
 
 -- CREATE/DROP COLLATION
 
+CREATE COLLATION none ( PROVIDER = none );
+
+CREATE COLLATION none2 ( PROVIDER = none, LOCALE="POSIX" ); -- fails
+CREATE COLLATION none2 ( PROVIDER = none, LC_CTYPE="POSIX" ); -- fails
+CREATE COLLATION none2 ( PROVIDER = none, LC_COLLATE="POSIX" ); -- fails
+
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
-- 
2.34.1

From 23e85920dbcfd1d3e71041f92c4adea589acd4f2 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 8 May 2023 13:48:01 -0700
Subject: [PATCH v5 5/7] ICU: for locale "C", automatically use "none" provider
 instead.

Postgres expects locale C to be optimizable to simple locale-unaware
byte operations; while ICU does not recognize the locale "C" at all,
and falls back to the root locale.

If the user specifies locale "C" when creating a new collation or a
new database with the ICU provider, automatically switch it to the
"none" provider.

If provider is libc, behavior is unchanged.
---
 doc/src/sgml/charset.sgml                     |  6 +++
 doc/src/sgml/ref/create_collation.sgml        |  6 +++
 doc/src/sgml/ref/create_database.sgml         |  5 +++
 doc/src/sgml/ref/createdb.sgml                |  5 +++
 doc/src/sgml/ref/initdb.sgml                  |  5 +++
 src/backend/commands/collationcmds.c          | 17 ++++++++
 src/backend/commands/dbcommands.c             | 21 ++++++++++
 src/bin/initdb/initdb.c                       | 10 +++++
 src/bin/initdb/t/001_initdb.pl                | 39 +++++++++++++++++++
 src/bin/scripts/createdb.c                    | 11 ++++++
 src/bin/scripts/t/020_createdb.pl             | 12 ++++++
 .../regress/expected/collate.icu.utf8.out     | 12 ++++--
 src/test/regress/sql/collate.icu.utf8.sql     |  3 ++
 13 files changed, 149 insertions(+), 3 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index de7c65ae35..5c4f713e8b 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -405,6 +405,12 @@ initdb --locale-provider=icu --icu-locale=en
      change in results. <literal>LC_COLLATE</literal> and
      <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
     </para>
+    <para>
+     The ICU provider does not accept the <literal>C</literal>
+     locale. Commands that create collations or database with the
+     <literal>icu</literal> provider and ICU locale <literal>C</literal> use
+     the provider <literal>none</literal> instead.
+    </para>
     <note>
      <para>
       For the ICU provider, results may depend on the version of the ICU
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index 5489ae7413..1ac41831d8 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -126,6 +126,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
        <literal>libc</literal> is the default.  See <xref
        linkend="locale-providers"/> for details.
       </para>
+      <para>
+       If the provider is <literal>icu</literal> and the locale is
+       <literal>C</literal> or <literal>POSIX</literal>, the provider is
+       automatically set to <literal>none</literal>; as the ICU provider
+       doesn't support an ICU locale of <literal>C</literal>.
+      </para>
      </listitem>
     </varlistentry>
 
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 60b9da0952..c730d02e15 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -190,6 +190,11 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
        <para>
         Specifies the ICU locale ID if the ICU locale provider is used.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index 326a371d34..7c573e848a 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -154,6 +154,11 @@ PostgreSQL documentation
         Specifies the ICU locale ID to be used in this database, if the
         ICU locale provider is selected.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index e604ab48b7..76993acdfe 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -250,6 +250,11 @@ PostgreSQL documentation
         Specifies the ICU locale when the ICU provider is used. Locale support
         is described in <xref linkend="locale"/>.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
        <para>
         If this option is not specified, the locale is inherited from the
         environment in which <command>initdb</command> runs. The environment's
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 267a551818..ed64e17504 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -254,6 +254,23 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		if (lcctypeEl)
 			collctype = defGetString(lcctypeEl);
 
+		/*
+		 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+		 * optimizable to byte operations (memcmp(), pg_ascii_tolower(),
+		 * etc.); transform into the "none" provider. Don't transform during
+		 * binary upgrade.
+		 */
+		if (!IsBinaryUpgrade && collprovider == COLLPROVIDER_ICU &&
+			colliculocale && (pg_strcasecmp(colliculocale, "C") == 0 ||
+							  pg_strcasecmp(colliculocale, "POSIX") == 0))
+		{
+			ereport(NOTICE,
+					(errmsg("using locale provider \"none\" for ICU locale \"%s\"",
+							colliculocale)));
+			colliculocale = NULL;
+			collprovider = COLLPROVIDER_NONE;
+		}
+
 		if (collprovider == COLLPROVIDER_LIBC)
 		{
 			if (!collcollate)
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 9e73f54803..6dc737aebb 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1043,6 +1043,27 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
 
+	/*
+	 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+	 * optimizable to byte operations (memcmp(), pg_ascii_tolower(), etc.);
+	 * transform into the "none" provider.
+	 *
+	 * Don't transform during binary upgrade or when both the provider and ICU
+	 * locale are unchanged from the template.
+	 */
+	if (!IsBinaryUpgrade && dblocprovider == COLLPROVIDER_ICU &&
+		(src_locprovider != COLLPROVIDER_ICU ||
+		 strcmp(dbiculocale, src_iculocale) != 0) &&
+		dbiculocale && (pg_strcasecmp(dbiculocale, "C") == 0 ||
+						pg_strcasecmp(dbiculocale, "POSIX") == 0))
+	{
+		ereport(NOTICE,
+				(errmsg("using locale provider \"none\" for ICU locale \"%s\"",
+						dbiculocale)));
+		dbiculocale = NULL;
+		dblocprovider = COLLPROVIDER_NONE;
+	}
+
 	if (dblocprovider == COLLPROVIDER_ICU)
 	{
 		if (!(is_encoding_supported_by_icu(encoding)))
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 4cf6892bee..ea26bf8361 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2501,6 +2501,16 @@ setlocales(void)
 			lc_messages = locale;
 	}
 
+	if (icu_locale && locale_provider == COLLPROVIDER_ICU &&
+		(pg_strcasecmp(icu_locale, "C") == 0 ||
+		 pg_strcasecmp(icu_locale, "POSIX") == 0))
+	{
+		pg_log_info("using locale provider \"none\" for ICU locale \"%s\"",
+					 icu_locale);
+		icu_locale = NULL;
+		locale_provider = COLLPROVIDER_NONE;
+	}
+
 	/*
 	 * canonicalize locale names, and obtain any missing values from our
 	 * current environment
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index fe6d224e5b..ea92b08511 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -111,6 +111,45 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'option --icu-locale');
 
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			"$tempdir/data4a"
+		],
+		'option --icu-locale=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--locale=C',
+			"$tempdir/data4b"
+		],
+		'option --icu-locale=C --locale=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-collate=C',
+			"$tempdir/data4c"
+		],
+		'option --icu-locale=C --lc-collate=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-ctype=C',
+			"$tempdir/data4d"
+		],
+		'option --icu-locale=C --lc-ctype=C');
+
 	command_fails_like(
 		[
 			'initdb',                '--no-sync',
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index 79367d933b..9caf9190cf 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -172,6 +172,17 @@ main(int argc, char *argv[])
 			lc_collate = locale;
 	}
 
+	if (locale_provider && pg_strcasecmp(locale_provider, "icu") == 0 &&
+		icu_locale &&
+		(pg_strcasecmp(icu_locale, "C") == 0 ||
+		 pg_strcasecmp(icu_locale, "POSIX") == 0))
+	{
+		pg_log_info("using locale provider \"none\" for ICU locale \"%s\"",
+					 icu_locale);
+		icu_locale = NULL;
+		locale_provider = "none";
+	}
+
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index 5aa658b671..eb3682f0fd 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -75,6 +75,18 @@ if ($ENV{with_icu} eq 'yes')
 	$node2->command_ok(
 		[ 'createdb', '-T', 'template0', '--icu-locale', 'en-US', 'foobar56' ],
 		'create database with icu locale from template database with icu provider');
+
+	# transformed into provider "none"
+	$node->command_ok(
+		[ 'createdb', '-T', 'template0', '--locale-provider=icu', '--icu-locale=C',
+		  'test_none_icu1' ],
+		'create database with provider "icu" and ICU_LOCALE="C"');
+
+	# transformed into provider "none"
+	$node->command_ok(
+		[ 'createdb', '-T', 'template0', '--locale-provider=icu', '--icu-locale=C',
+		  '--lc-ctype=C', 'test_none_icu_2' ],
+		'create database with provider "icu" and ICU_LOCALE="C" and LC_CTYPE=C');
 }
 else
 {
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 12afc3b65a..c0437231ad 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1035,6 +1035,9 @@ BEGIN
 END
 $$;
 RESET client_min_messages;
+-- uses "none" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+NOTICE:  using locale provider "none" for ICU locale "C"
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 SET icu_validation_level = ERROR;
@@ -1069,7 +1072,8 @@ SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
  test0
  test1
  test5
-(3 rows)
+ testc
+(4 rows)
 
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
@@ -1090,7 +1094,8 @@ SELECT collname, nspname, obj_description(pg_collation.oid, 'pg_collation')
  test0    | collate_tests | US English
  test11   | test_schema   | 
  test5    | collate_tests | 
-(3 rows)
+ testc    | collate_tests | 
+(4 rows)
 
 DROP COLLATION test0, test_schema.test11, test5;
 DROP COLLATION test0; -- fail
@@ -1100,7 +1105,8 @@ NOTICE:  collation "test0" does not exist, skipping
 SELECT collname FROM pg_collation WHERE collname LIKE 'test%';
  collname 
 ----------
-(0 rows)
+ testc
+(1 row)
 
 DROP SCHEMA test_schema;
 DROP ROLE regress_test_role;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 655c965f46..63c29dfe2a 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -375,6 +375,9 @@ $$;
 
 RESET client_min_messages;
 
+-- uses "none" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-- 
2.34.1

From 79732b2f94d5097b5ceebd2a22fdbb692c780156 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v5 6/7] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 ++--
 doc/src/sgml/ref/createdb.sgml                |  5 +++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++--
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 +++++++---
 src/bin/initdb/initdb.c                       | 11 ++++++--
 src/bin/scripts/createdb.c                    | 13 ++++-----
 src/bin/scripts/t/020_createdb.pl             |  4 +--
 src/test/icu/t/010_database.pl                | 23 +++++++++------
 .../regress/expected/collate.icu.utf8.out     | 28 +++++++++----------
 10 files changed, 68 insertions(+), 46 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index c730d02e15..dc57ba0c8b 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index 7c573e848a..7991153ecc 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 76993acdfe..d9ef21c422 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index ed64e17504..9a83f9f303 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -302,7 +302,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 				if (langtag && strcmp(colliculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, colliculocale)));
 
 					colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 6dc737aebb..154f20573c 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1019,7 +1019,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1033,12 +1038,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1094,7 +1101,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 			if (langtag && strcmp(dbiculocale, langtag) != 0)
 			{
 				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
+						(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 								langtag, dbiculocale)));
 
 				dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index ea26bf8361..ccb2414fed 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2157,7 +2157,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2467,7 +2471,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale_provider == COLLPROVIDER_NONE)
 	{
@@ -2499,6 +2503,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	if (icu_locale && locale_provider == COLLPROVIDER_ICU &&
@@ -3392,7 +3398,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index 9caf9190cf..51c4bb3592 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (locale_provider && pg_strcasecmp(locale_provider, "icu") == 0 &&
 		icu_locale &&
 		(pg_strcasecmp(icu_locale, "C") == 0 ||
@@ -230,6 +222,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index eb3682f0fd..81a9931c09 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -167,7 +167,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -175,7 +175,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 715b1bffd6..df4af00afe 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,16 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index c0437231ad..39f61ca281 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1058,11 +1058,11 @@ CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 -- test special variants
 CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-cu-eur" for locale "@EURO"
+NOTICE:  using standard form "und-u-cu-eur" for ICU locale "@EURO"
 CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-co-pinyin" for locale "@pinyin"
+NOTICE:  using standard form "und-u-co-pinyin" for ICU locale "@pinyin"
 CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
-NOTICE:  using standard form "und-u-co-stroke" for locale "@stroke"
+NOTICE:  using standard form "und-u-co-stroke" for ICU locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
@@ -1211,9 +1211,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1221,7 +1221,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1238,12 +1238,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1252,7 +1252,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1280,13 +1280,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1336,9 +1336,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 3ca8e0a84f6593ffff9a409bd31dc1c9ed253d3a Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 11 May 2023 12:54:31 -0700
Subject: [PATCH v5 7/7] Add default_collation_provider GUC.

Controls default collation provider for CREATE COLLATION. Does not
affect CREATE DATABASE, which gets its default from the template
database.
---
 doc/src/sgml/config.sgml                       | 17 +++++++++++++++++
 src/backend/commands/collationcmds.c           |  3 ++-
 src/backend/utils/misc/guc_tables.c            | 18 ++++++++++++++++++
 src/backend/utils/misc/postgresql.conf.sample  |  4 ++++
 src/include/commands/collationcmds.h           |  2 ++
 src/test/regress/expected/collate.icu.utf8.out | 17 +++++++++++++++++
 src/test/regress/sql/collate.icu.utf8.sql      | 10 ++++++++++
 7 files changed, 70 insertions(+), 1 deletion(-)

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index c4a9dcb9ae..038ecf9811 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -9819,6 +9819,23 @@ SET XML OPTION { DOCUMENT | CONTENT };
       </listitem>
      </varlistentry>
 
+     <varlistentry id="guc-default-collation-provider" xreflabel="default_collation_provider">
+      <term><varname>default_collation_provider</varname> (<type>enum</type>)
+      <indexterm>
+       <primary><varname>default_collation_provider</varname> configuration parameter</primary>
+      </indexterm>
+      </term>
+      <listitem>
+       <para>
+        Default collation provider for <command>CREATE
+        COLLATION</command>. Does not affect <command>CREATE
+        DATABASE</command>, which gets the default collation provider from the
+        template database. Valid values are <literal>icu</literal> and
+        <literal>libc</literal>. The default is <literal>libc</literal>.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry id="guc-icu-validation-level" xreflabel="icu_validation_level">
       <term><varname>icu_validation_level</varname> (<type>enum</type>)
       <indexterm>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 9a83f9f303..b42a660386 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -47,6 +47,7 @@ typedef struct
 	int			enc;			/* encoding */
 } CollAliasData;
 
+int		default_collation_provider = (int) COLLPROVIDER_LIBC;
 
 /*
  * CREATE COLLATION
@@ -228,7 +229,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+			collprovider = (char) default_collation_provider;
 
 		if (collprovider == COLLPROVIDER_NONE
 			&& (localeEl || lccollateEl || lcctypeEl))
diff --git a/src/backend/utils/misc/guc_tables.c b/src/backend/utils/misc/guc_tables.c
index 8c843f4ab6..d64b3a9a6f 100644
--- a/src/backend/utils/misc/guc_tables.c
+++ b/src/backend/utils/misc/guc_tables.c
@@ -35,8 +35,10 @@
 #include "access/xlogrecovery.h"
 #include "archive/archive_module.h"
 #include "catalog/namespace.h"
+#include "catalog/pg_collation.h"
 #include "catalog/storage.h"
 #include "commands/async.h"
+#include "commands/collationcmds.h"
 #include "commands/tablespace.h"
 #include "commands/trigger.h"
 #include "commands/user.h"
@@ -166,6 +168,12 @@ static const struct config_enum_entry intervalstyle_options[] = {
 	{NULL, 0, false}
 };
 
+static const struct config_enum_entry collation_provider_options[] = {
+	{"icu", (int) 'i', false},
+	{"libc", (int) 'c', false},
+	{NULL, 0, false}
+};
+
 static const struct config_enum_entry icu_validation_level_options[] = {
 	{"disabled", -1, false},
 	{"debug5", DEBUG5, false},
@@ -4683,6 +4691,16 @@ struct config_enum ConfigureNamesEnum[] =
 		NULL, NULL, NULL
 	},
 
+	{
+		{"default_collation_provider", PGC_USERSET, CLIENT_CONN_LOCALE,
+		 gettext_noop("Default collation provider for CREATE COLLATION."),
+		 NULL
+		},
+		&default_collation_provider,
+		(int) COLLPROVIDER_LIBC, collation_provider_options,
+		NULL, NULL, NULL
+	},
+
 	{
 		{"icu_validation_level", PGC_USERSET, CLIENT_CONN_LOCALE,
 		 gettext_noop("Log level for reporting invalid ICU locale strings."),
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index 87bad8ecbf..b2b015b31f 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -734,6 +734,10 @@
 #lc_numeric = 'C'			# locale for number formatting
 #lc_time = 'C'				# locale for time formatting
 
+#default_collation_provider = 'libc'	# default collation provider
+					# for CREATE COLLATION
+					# (none, icu, libc)
+
 #icu_validation_level = WARNING		# report ICU locale validation
 					# errors at the given level
 
diff --git a/src/include/commands/collationcmds.h b/src/include/commands/collationcmds.h
index b76c7b3dc3..f54389525d 100644
--- a/src/include/commands/collationcmds.h
+++ b/src/include/commands/collationcmds.h
@@ -18,6 +18,8 @@
 #include "catalog/objectaddress.h"
 #include "parser/parse_node.h"
 
+extern PGDLLIMPORT int default_collation_provider;
+
 extern ObjectAddress DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_exists);
 extern void IsThereCollationInNamespace(const char *collname, Oid nspOid);
 extern ObjectAddress AlterCollation(AlterCollationStmt *stmt);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 39f61ca281..d9da8d1310 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1038,6 +1038,23 @@ RESET client_min_messages;
 -- uses "none" provider instead
 CREATE COLLATION testc (provider = icu, locale='C');
 NOTICE:  using locale provider "none" for ICU locale "C"
+SET default_collation_provider = 'libc';
+CREATE COLLATION def_libc (LOCALE = 'C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+ collname | collprovider 
+----------+--------------
+ def_libc | c
+(1 row)
+
+SET default_collation_provider = 'icu';
+CREATE COLLATION def_icu (LOCALE = 'und');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_icu';
+ collname | collprovider 
+----------+--------------
+ def_icu  | i
+(1 row)
+
+RESET default_collation_provider;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 SET icu_validation_level = ERROR;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 63c29dfe2a..13089c7f8e 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -378,6 +378,16 @@ RESET client_min_messages;
 -- uses "none" provider instead
 CREATE COLLATION testc (provider = icu, locale='C');
 
+SET default_collation_provider = 'libc';
+CREATE COLLATION def_libc (LOCALE = 'C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+
+SET default_collation_provider = 'icu';
+CREATE COLLATION def_icu (LOCALE = 'und');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_icu';
+
+RESET default_collation_provider;
+
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-- 
2.34.1

From 9c8e9272ca807c9f75a7b32fa3190700cc600260 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 15 May 2023 13:35:07 -0700
Subject: [PATCH] ICU: check for U_STRING_NOT_TERMINATED_WARNING.

In some cases, ICU can fail to NUL-terminate a result string even if
using an appropriately-sized static buffer. The caller must either
check for len >= buflen or U_STRING_NOT_TERMINATED_WARNING.

The specific problem is related to uloc_getLanguage(), but add the
check in other cases as well.

Reported-by: Alexander Lakhin
Discussion: https://postgr.es/m/2098874d-c111-41e4-9063-30bcf135226b@gmail.com
---
 src/backend/utils/adt/pg_locale.c | 29 +++++++++++------------------
 src/bin/initdb/initdb.c           | 15 ++++-----------
 2 files changed, 15 insertions(+), 29 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index f0b6567da1..1cf93b2d20 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2468,7 +2468,7 @@ pg_ucol_open(const char *loc_str)
 
 		status = U_ZERO_ERROR;
 		uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-		if (U_FAILURE(status))
+		if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 		{
 			ereport(ERROR,
 					(errmsg("could not get language from locale \"%s\": %s",
@@ -2504,7 +2504,7 @@ pg_ucol_open(const char *loc_str)
 		 * Pretend the error came from ucol_open(), for consistent error
 		 * message across ICU versions.
 		 */
-		if (U_FAILURE(status))
+		if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 		{
 			ucol_close(collator);
 			ereport(ERROR,
@@ -2639,7 +2639,8 @@ icu_from_uchar(char **result, const UChar *buff_uchar, int32_t len_uchar)
 	status = U_ZERO_ERROR;
 	len_result = ucnv_fromUChars(icu_converter, *result, len_result + 1,
 								 buff_uchar, len_uchar, &status);
-	if (U_FAILURE(status))
+	if (U_FAILURE(status) ||
+		status == U_STRING_NOT_TERMINATED_WARNING)
 		ereport(ERROR,
 				(errmsg("%s failed: %s", "ucnv_fromUChars",
 						u_errorName(status))));
@@ -2681,7 +2682,7 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 	icu_locale_id = palloc(len + 1);
 	*status = U_ZERO_ERROR;
 	len = uloc_canonicalize(loc, icu_locale_id, len + 1, status);
-	if (U_FAILURE(*status))
+	if (U_FAILURE(*status) || *status == U_STRING_NOT_TERMINATED_WARNING)
 		return;
 
 	lower_str = asc_tolower(icu_locale_id, strlen(icu_locale_id));
@@ -2765,7 +2766,6 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 
 	pfree(lower_str);
 }
-
 #endif
 
 /*
@@ -2789,7 +2789,7 @@ icu_language_tag(const char *loc_str, int elevel)
 
 	status = U_ZERO_ERROR;
 	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
+	if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 	{
 		if (elevel > 0)
 			ereport(elevel,
@@ -2811,19 +2811,12 @@ icu_language_tag(const char *loc_str, int elevel)
 	langtag = palloc(buflen);
 	while (true)
 	{
-		int32_t		len;
-
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
 
-		/*
-		 * If the result fits in the buffer exactly (len == buflen),
-		 * uloc_toLanguageTag() will return success without nul-terminating
-		 * the result. Check for either U_BUFFER_OVERFLOW_ERROR or len >=
-		 * buflen and try again.
-		 */
+		/* try again if the buffer is not large enough */
 		if ((status == U_BUFFER_OVERFLOW_ERROR ||
-			 (U_SUCCESS(status) && len >= buflen)) &&
+			 status == U_STRING_NOT_TERMINATED_WARNING) &&
 			buflen < MaxAllocSize)
 		{
 			buflen = Min(buflen * 2, MaxAllocSize);
@@ -2878,7 +2871,7 @@ icu_validate_locale(const char *loc_str)
 	/* validate that we can extract the language */
 	status = U_ZERO_ERROR;
 	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
+	if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 	{
 		ereport(elevel,
 				(errmsg("could not get language from ICU locale \"%s\": %s",
@@ -2901,7 +2894,7 @@ icu_validate_locale(const char *loc_str)
 
 		status = U_ZERO_ERROR;
 		uloc_getLanguage(otherloc, otherlang, ULOC_LANG_CAPACITY, &status);
-		if (U_FAILURE(status))
+		if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 			continue;
 
 		if (strcmp(lang, otherlang) == 0)
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index e03d498b1e..30b576932f 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2252,7 +2252,7 @@ icu_language_tag(const char *loc_str)
 
 	status = U_ZERO_ERROR;
 	uloc_getLanguage(loc_str, lang, ULOC_LANG_CAPACITY, &status);
-	if (U_FAILURE(status))
+	if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING)
 	{
 		pg_fatal("could not get language from locale \"%s\": %s",
 				 loc_str, u_errorName(status));
@@ -2272,19 +2272,12 @@ icu_language_tag(const char *loc_str)
 	langtag = pg_malloc(buflen);
 	while (true)
 	{
-		int32_t		len;
-
 		status = U_ZERO_ERROR;
-		len = uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
 
-		/*
-		 * If the result fits in the buffer exactly (len == buflen),
-		 * uloc_toLanguageTag() will return success without nul-terminating
-		 * the result. Check for either U_BUFFER_OVERFLOW_ERROR or len >=
-		 * buflen and try again.
-		 */
+		/* try again if the buffer is not large enough */
 		if (status == U_BUFFER_OVERFLOW_ERROR ||
-			(U_SUCCESS(status) && len >= buflen))
+			status == U_STRING_NOT_TERMINATED_WARNING)
 		{
 			buflen = buflen * 2;
 			langtag = pg_realloc(langtag, buflen);
-- 
2.34.1

From 8633ec205b0b0297910cef8f931092d0c05eb3ce Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 27 Apr 2023 14:43:46 -0700
Subject: [PATCH v2] Doc improvements for language tags and custom ICU
 collations.

Separate the documentation for language tags themselves from the
available collation settings which can be included in a language tag.

Include tables of the available options, more details about the
effects of each option, and additional examples.

Also include an explanation of the "levels" of textual features and
how they relate to collation.

Discussion: https://postgr.es/m/25787ec7-4c04-9a8a-d241-4dc9be0b1ba3@postgresql.org
Reviewed-by: Jonathan S. Katz
---
 doc/src/sgml/charset.sgml | 680 +++++++++++++++++++++++++++++++-------
 1 file changed, 559 insertions(+), 121 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 6dd95b8966..ea43732ec9 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -377,7 +377,134 @@ initdb --locale-provider=icu --icu-locale=en
     variants and customization options.
    </para>
   </sect2>
+  <sect2 id="icu-locales">
+   <title>ICU Locales</title>
+   <sect3 id="icu-locale-names">
+    <title>ICU Locale Names</title>
+    <para>
+     The ICU format for the locale name is a <link
+     linkend="icu-language-tag">Language Tag</link>.
+
+<programlisting>
+CREATE COLLATION mycollation1 (PROVIDER = icu, LOCALE = 'ja-JP);
+CREATE COLLATION mycollation2 (PROVIDER = icu, LOCALE = 'fr');
+</programlisting>
+    </para>
+   </sect3>
+   <sect3 id="icu-canonicalization">
+    <title>Locale Canonicalization and Validation</title>
+    <para>
+     When defining a new ICU collation object or database with ICU as the
+     provider, the given locale name is transformed ("canonicalized") into a
+     language tag if not already in that form. For instance,
+
+<screen>
+CREATE COLLATION mycollation3 (PROVIDER = icu, LOCALE = 'en-US-u-kn-true');
+NOTICE:  using standard form "en-US-u-kn" for locale "en-US-u-kn-true"
+CREATE COLLATION mycollation4 (PROVIDER = icu, LOCALE = 'de_DE.utf8');
+NOTICE:  using standard form "de-DE" for locale "de_DE.utf8"
+</screen>
+
+     If you see this notice, ensure that the <symbol>PROVIDER</symbol> and
+     <symbol>LOCALE</symbol> are the expected result. For consistent results
+     when using the ICU provider, specify the canonical <link
+     linkend="icu-language-tag">language tag</link> instead of relying on the
+     transformation.
+    </para>
+    <para>
+     A locale with no language name, or the special language name
+     <literal>root</literal>, is transformed to have the language
+     <literal>und</literal> ("undefined").
+    </para>
+    <para>
+     ICU can transform most libc locale names, as well as some other formats,
+     into language tags for easier transition to ICU. If a libc locale name is
+     used in ICU, it may not have precisely the same behavior as in libc.
+    </para>
+    <para>
+     If there is a problem interpreting the locale name, or if the locale name
+     represents a language or region that ICU does not recognize, you will see
+     the following error:
+
+<screen>
+CREATE COLLATION nonsense (PROVIDER = icu, LOCALE = 'nonsense');
+ERROR:  ICU locale "nonsense" has unknown language "nonsense"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+</screen>
+
+     <xref
+     linkend="guc-icu-validation-level"/> controls how the message is
+     reported. If set below <literal>ERROR</literal>, the collation will still
+     be created, but the behavior may not be what the user intended.
+    </para>
+   </sect3>
+   <sect3 id="icu-language-tag">
+    <title>Language Tag</title>
+    <para>
+     A language tag, defined in BCP 47, is a standardized identifier used to
+     identify languages, regions, and other information about a locale.
+    </para>
+    <para>
+     Basic language tags are simply
+     <replaceable>language</replaceable><literal>-</literal><replaceable>region</replaceable>;
+     or even just <replaceable>language</replaceable>. The
+     <replaceable>language</replaceable> is a language code
+     (e.g. <literal>fr</literal> for French), and
+     <replaceable>region</replaceable> is a region code
+     (e.g. <literal>CA</literal> for Canada). Examples:
+     <literal>ja-JP</literal>, <literal>de</literal>, or
+     <literal>fr-CA</literal>.
+    </para>
+    <para>
+     Collation settings may be included in the language tag to customize
+     collation behavior. ICU allows extensive customization, such as
+     sensitivity (or insensitivity) to accents, case, and punctuation;
+     treatment of digits within text; and many other options to satisfy a
+     variety of uses.
+    </para>
+    <para>
+     To include this additional collation information in a language tag,
+     append <literal>-u</literal>, which indicates there are additional
+     collation settings, followed by one or more
+     <literal>-</literal><replaceable>key</replaceable><literal>-</literal><replaceable>value</replaceable>
+     pairs. The <replaceable>key</replaceable> is the key for a <link
+     linkend="icu-collation-settings">collation setting</link> and
+     <replaceable>value</replaceable> is a valid value for that setting. For
+     boolean settings, the <literal>-</literal><replaceable>key</replaceable>
+     may be specified without a corresponding
+     <literal>-</literal><replaceable>value</replaceable>, which implies a
+     value of <literal>true</literal>.
+    </para>
+    <para>
+     For example, the language tag <literal>en-US-u-kn-ks-level2</literal>
+     means the locale with the English language in the US region, with
+     collation settings <literal>kn</literal> set to <literal>true</literal>
+     and <literal>ks</literal> set to <literal>level2</literal>. Those
+     settings mean the collation will be case-insensitive and treat a sequence
+     of digits as a single number:
 
+<screen>
+CREATE COLLATION mycollation5 (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'en-US-u-kn-ks-level2');
+SELECT 'aB' = 'Ab' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+
+SELECT 'N-45' &lt; 'N-123' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+</screen>
+    </para>
+    <para>
+     See <xref linkend="icu-custom-collations"/> for details and additional
+     examples of using language tags with custom collation information for the
+     locale.
+    </para>
+   </sect3>
+  </sect2>
   <sect2 id="locale-problems">
    <title>Problems</title>
 
@@ -658,6 +785,13 @@ SELECT * FROM test1 ORDER BY a || b COLLATE "fr_FR";
     code byte values.
    </para>
 
+   <note>
+    <para>
+     The <literal>C</literal> and <literal>POSIX</literal> locales may behave
+     differently depending on the database encoding.
+    </para>
+   </note>
+
    <para>
     Additionally, two SQL standard collation names are available:
 
@@ -869,132 +1003,24 @@ CREATE COLLATION german (provider = libc, locale = 'de_DE');
    <sect4 id="collation-managing-create-icu">
     <title>ICU Collations</title>
 
-   <para>
-    ICU allows collations to be customized beyond the basic language+country
-    set that is preloaded by <command>initdb</command>.  Users are encouraged
-    to define their own collation objects that make use of these facilities to
-    suit the sorting behavior to their requirements.
-    See <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
-    and <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink> for
-    information on ICU locale naming.  The set of acceptable names and
-    attributes depends on the particular ICU version.
-   </para>
-
-   <para>
-    Here are some examples:
-
-    <variablelist>
-     <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de@collation=phonebook');</literal></term>
-      <listitem>
-       <para>German collation with phone book collation type</para>
-       <para>
-        The first example selects the ICU locale using a <quote>language
-        tag</quote> per BCP 47.  The second example uses the traditional
-        ICU-specific locale syntax.  The first style is preferred going
-        forward, and is used internally to store locales.
-       </para>
-       <para>
-        Note that you can name the collation objects in the SQL environment
-        anything you want.  In this example, we follow the naming style that
-        the predefined collations use, which in turn also follow BCP 47, but
-        that is not required for user-defined collations.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = '@collation=emoji');</literal></term>
-      <listitem>
-       <para>
-        Root collation with Emoji collation type, per Unicode Technical Standard #51
-       </para>
-       <para>
-        Observe how in the traditional ICU locale naming system, the root
-        locale is selected by an empty string.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en@colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en@colCaseFirst=upper');</literal></term>
-      <listitem>
-       <para>
-        Sort upper-case letters before lower-case letters.  (The default is
-        lower-case letters first.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-    <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en@colCaseFirst=upper;colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Combines both of the above options.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kn-true">
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en-u-kn-true');</literal></term>
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en@colNumeric=yes');</literal></term>
-      <listitem>
-       <para>
-        Numeric ordering, sorts sequences of digits by their numeric value,
-        for example: <literal>A-21</literal> &lt; <literal>A-123</literal>
-        (also known as natural sort).
-       </para>
-      </listitem>
-     </varlistentry>
-    </variablelist>
-
-    See <ulink url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
-    Technical Standard #35</ulink>
-    and <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink> for
-    details.  The list of possible collation types (<literal>co</literal>
-    subtag) can be found in
-    the <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
-    repository</ulink>.
-   </para>
+    <para>
+     ICU collations can be created like:
 
-   <para>
-    Note that while this system allows creating collations that <quote>ignore
-    case</quote> or <quote>ignore accents</quote> or similar (using the
-    <literal>ks</literal> key), in order for such collations to act in a
-    truly case- or accent-insensitive manner, they also need to be declared as not
-    <firstterm>deterministic</firstterm> in <command>CREATE COLLATION</command>;
-    see <xref linkend="collation-nondeterministic"/>.
-    Otherwise, any strings that compare equal according to the collation but
-    are not byte-wise equal will be sorted according to their byte values.
-   </para>
+<programlisting>
+CREATE COLLATION german (provider = icu, locale = 'de-DE');
+</programlisting>
 
-   <note>
+     ICU locales are specified as a BCP 47 <link
+     linkend="icu-language-tag">Language Tag</link>, but can also accept most
+     libc-style locale names. If possible, libc-style locale names are
+     transformed into language tags.
+    </para>
     <para>
-     By design, ICU will accept almost any string as a locale name and match
-     it to the closest locale it can provide, using the fallback procedure
-     described in its documentation.  Thus, there will be no direct feedback
-     if a collation specification is composed using features that the given
-     ICU installation does not actually support.  It is therefore recommended
-     to create application-level test cases to check that the collation
-     definitions satisfy one's requirements.
+     New ICU collations can customize collation behavior extensively by
+     including collation attributes in the langugage tag. See <xref
+     linkend="icu-custom-collations"/> for details and examples.
     </para>
-   </note>
    </sect4>
-
    <sect4 id="collation-copy">
    <title>Copying Collations</title>
 
@@ -1072,6 +1098,418 @@ CREATE COLLATION ignore_accents (provider = icu, locale = 'und-u-ks-level1-kc-tr
     </tip>
    </sect3>
   </sect2>
+  <sect2 id="icu-custom-collations">
+   <title>ICU Custom Collations</title>
+
+   <para>
+    ICU allows extensive control over collation behavior by defining new
+    collations with collation settings as a part of the language tag. These
+    settings can modify the collation order to suit a variety of needs. For
+    instance:
+
+<programlisting>
+-- ignore differences in accents and case
+CREATE COLLATION ignore_accent_case (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ks-level1');
+SELECT 'Å' = 'A' COLLATE ignore_accent_case; -- true
+SELECT 'z' = 'Z' COLLATE ignore_accent_case; -- true
+
+-- upper case letters sort before lower case.
+CREATE COLLATION upper_first (PROVIDER=icu, LOCALE = 'und-u-kf-upper');
+SELECT 'B' &lt; 'b' COLLATE upper_first; -- true
+
+-- treat digits numerically and ignore punctuation
+CREATE COLLATION num_ignore_punct (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ka-shifted-kn');
+SELECT 'id-45' &lt; 'id-123' COLLATE num_ignore_punct; -- true
+SELECT 'w;x*y-z' = 'wxyz' COLLATE num_ignore_punct; -- true
+</programlisting>
+
+    Many of the available options are described in <xref
+    linkend="icu-collation-settings"/>, or see <xref
+    linkend="icu-external-references"/> for more details.
+   </para>
+   <sect3 id="icu-collation-comparison-levels">
+    <title>ICU Comparison Levels</title>
+    <para>
+     Comparison of two strings (collation) in ICU is determined by a
+     multi-level process, where textual features are grouped into
+     "levels". Treatment of each level is controlled by the <link
+     linkend="icu-collation-settings-table">collation settings</link>. Higher
+     levels correspond to finer textual features.
+    </para>
+    <para>
+     <table id="icu-collation-levels">
+      <title>ICU Collation Levels</title>
+      <tgroup cols="3">
+       <thead>
+        <row>
+         <entry>Level</entry>
+         <entry>Description</entry>
+         <entry><literal>'f' = 'f'</literal></entry>
+         <entry><literal>'ab' = U&amp;'a\2063b'</literal></entry>
+         <entry><literal>'x-y' = 'x_y'</literal></entry>
+         <entry><literal>'g' = 'G'</literal></entry>
+         <entry><literal>'n' = 'ñ'</literal></entry>
+         <entry><literal>'y' = 'z'</literal></entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry>level1</entry>
+         <entry>Base Character</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level2</entry>
+         <entry>Accents</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level3</entry>
+         <entry>Case/Variants</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level4</entry>
+         <entry>Punctuation</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>identic</entry>
+         <entry>All</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+
+     The above table shows which textual feature differences are
+     considered significant when determining equality at the given level. The
+     unicode character <literal>U+2063</literal> is an invisible separator,
+     and as seen in the table, is ignored for at all levels of comparison less
+     than <literal>identic</literal>.
+    </para>
+    <para>
+     At every level, even with full normalization off, basic normalization is
+     performed. For example, <literal>'á'</literal> may be composed of the
+     code points <literal>U&amp;'\0061\0301'</literal> or the single code
+     point <literal>U&amp;'\00E1'</literal>, and those sequences will be
+     considered equal even at the <literal>identic</literal> level. To treat
+     any difference in code point representation as distinct, use a collation
+     created with <symbol>DETERMINISTIC</symbol> set to
+     <literal>false</literal>.
+    </para>
+    <para>
+     Examples:
+
+<programlisting>
+CREATE COLLATION level3 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level3');
+CREATE COLLATION level4 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level4');
+CREATE COLLATION identic (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-identic');
+
+-- invisible separator ignored at all levels except identic
+SELECT 'ab' = U&amp;'a\2063b' COLLATE level4; -- true
+SELECT 'ab' = U&amp;'a\2063b' COLLATE identic; -- false
+
+-- punctuation ignored at level3 but not at level 4
+SELECT 'x-y' = 'x_y' COLLATE level3; -- true
+SELECT 'x-y' = 'x_y' COLLATE level4; -- false
+</programlisting>
+
+    </para>
+    <note>
+     <para>
+      For many collation settings, you must create the collation with
+      <option>DETERMINISTIC</option> set to <literal>false</literal> for the
+      setting to have the desired effect. Additionally, some settings only
+      take effect when the key <literal>ka</literal> is set to
+      <literal>shifted</literal> (see <xref
+      linkend="icu-collation-settings-table"/>).
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="icu-collation-settings">
+    <title>Collation Settings for an ICU Locale</title>
+    <para>
+     <table id="icu-collation-settings-table">
+      <title>ICU Collation Settings</title>
+      <tgroup cols="4">
+       <thead>
+        <row>
+         <entry>Key</entry>
+         <entry>Values</entry>
+         <entry>Default</entry>
+         <entry>Description</entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry><literal>co</literal></entry>
+         <entry><literal>emoji</literal>, <literal>phonebk</literal>, <literal>standard</literal>, <replaceable>...</replaceable></entry>
+         <entry><literal>standard</literal></entry>
+         <entry>
+          Collation type. See <xref linkend="icu-external-references"/> for additional options and details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ks</literal></entry>
+         <entry><literal>level1</literal>, <literal>level2</literal>, <literal>level3</literal>, <literal>level4</literal>, <literal>identic</literal></entry>
+         <entry><literal>level3</literal></entry>
+         <entry>
+          Sensitivity (or "strength") when determining equality, with
+          <literal>level1</literal> the least sensitive to differences and
+          <literal>identic</literal> the most sensitive to differences. See
+          <xref linkend="icu-collation-levels"/> for details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ka</literal></entry>
+         <entry><literal>noignore</literal>, <literal>shifted</literal></entry>
+         <entry><literal>noignore</literal></entry>
+         <entry>
+          If set to <literal>shifted</literal>, causes some characters
+          (e.g. punctuation or space) to be ignored in comparison. Key
+          <literal>ks</literal> must be set to <literal>level3</literal> or
+          lower to take effect. Set key <literal>kv</literal> to control which
+          character classes are ignored.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kb</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          Backwards comparison for the level 2 differences. For example,
+          locale <literal>und-u-kb</literal> sorts <literal>'àe'</literal>
+          before <literal>'aé'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kk</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Enable full normalization; may affect performance. Basic
+           normalization is performed even when set to
+           <literal>false</literal>. Locales for languages that require full
+           normalization typically enable it by default.
+          </para>
+          <para>
+           Full normalization is important in some cases, such as when
+           multiple accents are applied to a single character. For instance,
+           <literal>'ệ'</literal> can be composed of code points
+           <literal>U&amp;'\0065\0323\0302'</literal> or
+           <literal>U&amp;'\0065\0302\0323'</literal>. With full normalization
+           on, these code point sequences are treated as equal; otherwise they
+           are unequal.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kc</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Separates case into a "level 2.5" that falls between accents and
+           other level 3 features.
+          </para>
+          <para>
+           If set to <literal>true</literal> and <literal>ks</literal> is set
+           to <literal>level1</literal>, will ignore accents but take case
+           into account.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kf</literal></entry>
+         <entry>
+          <literal>upper</literal>, <literal>lower</literal>,
+          <literal>false</literal>
+         </entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>upper</literal>, upper case sorts before lower
+          case. If set to <literal>lower</literal>, lower case sorts before
+          upper case. If set to <literal>false</literal>, the sort depends on
+          the rules of the locale.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kn</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>true</literal>, numbers within a string are
+          treated as a single numeric value rather than a sequence of
+          digits. For example, <literal>'id-45'</literal> sorts before
+          <literal>'id-123'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kr</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>,
+          <literal>digit</literal>, <replaceable>script-id</replaceable>
+         </entry>
+         <entry></entry>
+         <entry>
+          <para>
+           Set to one or more of the valid values, or any BCP 47
+           <replaceable>script-id</replaceable>, e.g. <literal>latn</literal>
+           ("Latin") or <literal>grek</literal> ("Greek"). Multiple values are
+           separated by "<literal>-</literal>".
+          </para>
+          <para>
+           Redefines the ordering of classes of characters; those characters
+           belonging to a class earlier in the list sort before characters
+           belonging to a class later in the list. For instance, the value
+           <literal>digit-currency-space</literal> (as part of a language tag
+           like <literal>und-u-kr-digit-currency-space</literal>) sorts
+           punctuation before digits and spaces.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kv</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>
+         </entry>
+         <entry><literal>punct</literal></entry>
+         <entry>
+          Classes of characters ignored during comparison at level 3. Setting
+          to a later value includes earlier values;
+          e.g. <literal>symbol</literal> also includes
+          <literal>punct</literal> and <literal>space</literal> in the
+          characters to be ignored. Key <literal>ka</literal> must be set to
+          <literal>shifted</literal> and key <literal>ks</literal> must be set
+          to <literal>level3</literal> or lower to take effect.
+         </entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+      Defaults may depend on locale. The above table is not meant to be
+      complete. See <xref linkend="icu-external-references"/> for additional
+      options and details.
+    </para>
+   </sect3>
+   <sect3 id="icu-locale-examples">
+    <title>Examples</title>
+    <para>
+     <variablelist>
+      <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
+       <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
+       <listitem>
+        <para>German collation with phone book collation type</para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
+       <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
+       <listitem>
+        <para>
+         Root collation with Emoji collation type, per Unicode Technical Standard #51
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
+       <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
+       <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
+       <listitem>
+        <para>
+         Sort upper-case letters before lower-case letters.  (The default is
+         lower-case letters first.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
+       <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Combines both of the above options.
+        </para>
+       </listitem>
+      </varlistentry>
+     </variablelist>
+    </para>
+   </sect3>
+   <sect3 id="icu-external-references">
+    <title>External References for ICU</title>
+    <para>
+     This section (<xref linkend="icu-custom-collations"/>) is only a brief
+     overview of ICU behavior and language tags. Refer to the following
+     documents for technical details, additional options, and new behavior:
+    </para>
+    <itemizedlist>
+     <listitem>
+      <para>
+       <ulink
+           url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
+       Technical Standard #35</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
+       repository</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink>
+      </para>
+     </listitem>
+    </itemizedlist>
+   </sect3>
+  </sect2>
  </sect1>
 
  <sect1 id="multibyte">
-- 
2.34.1

From d0d2375fa55618b60f361f6bb64b2c49490125b9 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 27 Apr 2023 14:43:46 -0700
Subject: [PATCH] Doc improvements for language tags and custom ICU collations.

Separate the documentation for language tags themselves from the
available collation settings which can be included in a language tag.

Include tables of the available options, more details about the
effects of each option, and additional examples.

Also include an explanation of the "levels" of textual features and
how they relate to collation.

Discussion: https://postgr.es/m/25787ec7-4c04-9a8a-d241-4dc9be0b1ba3@postgresql.org
Reviewed-by: Jonathan S. Katz
---
 doc/src/sgml/charset.sgml | 683 +++++++++++++++++++++++++++++++-------
 1 file changed, 562 insertions(+), 121 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 6dd95b8966..6b9c323edd 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -377,7 +377,134 @@ initdb --locale-provider=icu --icu-locale=en
     variants and customization options.
    </para>
   </sect2>
+  <sect2 id="icu-locales">
+   <title>ICU Locales</title>
+   <sect3 id="icu-locale-names">
+    <title>ICU Locale Names</title>
+    <para>
+     The ICU format for the locale name is a <link
+     linkend="icu-language-tag">Language Tag</link>.
+
+<programlisting>
+CREATE COLLATION mycollation1 (PROVIDER = icu, LOCALE = 'ja-JP);
+CREATE COLLATION mycollation2 (PROVIDER = icu, LOCALE = 'fr');
+</programlisting>
+    </para>
+   </sect3>
+   <sect3 id="icu-canonicalization">
+    <title>Locale Canonicalization and Validation</title>
+    <para>
+     When defining a new ICU collation object or database with ICU as the
+     provider, the given locale name is transformed ("canonicalized") into a
+     language tag if not already in that form. For instance,
+
+<screen>
+CREATE COLLATION mycollation3 (PROVIDER = icu, LOCALE = 'en-US-u-kn-true');
+NOTICE:  using standard form "en-US-u-kn" for locale "en-US-u-kn-true"
+CREATE COLLATION mycollation4 (PROVIDER = icu, LOCALE = 'de_DE.utf8');
+NOTICE:  using standard form "de-DE" for locale "de_DE.utf8"
+</screen>
+
+     If you see this notice, ensure that the <symbol>PROVIDER</symbol> and
+     <symbol>LOCALE</symbol> are the expected result. For consistent results
+     when using the ICU provider, specify the canonical <link
+     linkend="icu-language-tag">language tag</link> instead of relying on the
+     transformation.
+    </para>
+    <para>
+     A locale with no language name, or the special language name
+     <literal>root</literal>, is transformed to have the language
+     <literal>und</literal> ("undefined").
+    </para>
+    <para>
+     ICU can transform most libc locale names, as well as some other formats,
+     into language tags for easier transition to ICU. If a libc locale name is
+     used in ICU, it may not have precisely the same behavior as in libc.
+    </para>
+    <para>
+     If there is a problem interpreting the locale name, or if the locale name
+     represents a language or region that ICU does not recognize, you will see
+     the following warning:
+
+<screen>
+CREATE COLLATION nonsense (PROVIDER = icu, LOCALE = 'nonsense');
+WARNING:  ICU locale "nonsense" has unknown language "nonsense"
+HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION
+</screen>
+
+     <xref linkend="guc-icu-validation-level"/> controls how the message is
+     reported. Unless set to <literal>ERROR</literal>, the collation will
+     still be created, but the behavior may not be what the user intended.
+    </para>
+   </sect3>
+   <sect3 id="icu-language-tag">
+    <title>Language Tag</title>
+    <para>
+     A language tag, defined in BCP 47, is a standardized identifier used to
+     identify languages, regions, and other information about a locale.
+    </para>
+    <para>
+     Basic language tags are simply
+     <replaceable>language</replaceable><literal>-</literal><replaceable>region</replaceable>;
+     or even just <replaceable>language</replaceable>. The
+     <replaceable>language</replaceable> is a language code
+     (e.g. <literal>fr</literal> for French), and
+     <replaceable>region</replaceable> is a region code
+     (e.g. <literal>CA</literal> for Canada). Examples:
+     <literal>ja-JP</literal>, <literal>de</literal>, or
+     <literal>fr-CA</literal>.
+    </para>
+    <para>
+     Collation settings may be included in the language tag to customize
+     collation behavior. ICU allows extensive customization, such as
+     sensitivity (or insensitivity) to accents, case, and punctuation;
+     treatment of digits within text; and many other options to satisfy a
+     variety of uses.
+    </para>
+    <para>
+     To include this additional collation information in a language tag,
+     append <literal>-u</literal>, which indicates there are additional
+     collation settings, followed by one or more
+     <literal>-</literal><replaceable>key</replaceable><literal>-</literal><replaceable>value</replaceable>
+     pairs. The <replaceable>key</replaceable> is the key for a <link
+     linkend="icu-collation-settings">collation setting</link> and
+     <replaceable>value</replaceable> is a valid value for that setting. For
+     boolean settings, the <literal>-</literal><replaceable>key</replaceable>
+     may be specified without a corresponding
+     <literal>-</literal><replaceable>value</replaceable>, which implies a
+     value of <literal>true</literal>.
+    </para>
+    <para>
+     For example, the language tag <literal>en-US-u-kn-ks-level2</literal>
+     means the locale with the English language in the US region, with
+     collation settings <literal>kn</literal> set to <literal>true</literal>
+     and <literal>ks</literal> set to <literal>level2</literal>. Those
+     settings mean the collation will be case-insensitive and treat a sequence
+     of digits as a single number:
 
+<screen>
+CREATE COLLATION mycollation5 (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'en-US-u-kn-ks-level2');
+SELECT 'aB' = 'Ab' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+
+SELECT 'N-45' &lt; 'N-123' COLLATE mycollation5 as result;
+ result
+--------
+ t
+(1 row)
+</screen>
+    </para>
+    <para>
+     See <xref linkend="icu-custom-collations"/> for details and additional
+     examples of using language tags with custom collation information for the
+     locale.
+    </para>
+   </sect3>
+  </sect2>
   <sect2 id="locale-problems">
    <title>Problems</title>
 
@@ -658,6 +785,13 @@ SELECT * FROM test1 ORDER BY a || b COLLATE "fr_FR";
     code byte values.
    </para>
 
+   <note>
+    <para>
+     The <literal>C</literal> and <literal>POSIX</literal> locales may behave
+     differently depending on the database encoding.
+    </para>
+   </note>
+
    <para>
     Additionally, two SQL standard collation names are available:
 
@@ -869,132 +1003,24 @@ CREATE COLLATION german (provider = libc, locale = 'de_DE');
    <sect4 id="collation-managing-create-icu">
     <title>ICU Collations</title>
 
-   <para>
-    ICU allows collations to be customized beyond the basic language+country
-    set that is preloaded by <command>initdb</command>.  Users are encouraged
-    to define their own collation objects that make use of these facilities to
-    suit the sorting behavior to their requirements.
-    See <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
-    and <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink> for
-    information on ICU locale naming.  The set of acceptable names and
-    attributes depends on the particular ICU version.
-   </para>
-
-   <para>
-    Here are some examples:
-
-    <variablelist>
-     <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
-      <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de@collation=phonebook');</literal></term>
-      <listitem>
-       <para>German collation with phone book collation type</para>
-       <para>
-        The first example selects the ICU locale using a <quote>language
-        tag</quote> per BCP 47.  The second example uses the traditional
-        ICU-specific locale syntax.  The first style is preferred going
-        forward, and is used internally to store locales.
-       </para>
-       <para>
-        Note that you can name the collation objects in the SQL environment
-        anything you want.  In this example, we follow the naming style that
-        the predefined collations use, which in turn also follow BCP 47, but
-        that is not required for user-defined collations.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
-      <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = '@collation=emoji');</literal></term>
-      <listitem>
-       <para>
-        Root collation with Emoji collation type, per Unicode Technical Standard #51
-       </para>
-       <para>
-        Observe how in the traditional ICU locale naming system, the root
-        locale is selected by an empty string.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en@colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
-      <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en@colCaseFirst=upper');</literal></term>
-      <listitem>
-       <para>
-        Sort upper-case letters before lower-case letters.  (The default is
-        lower-case letters first.)
-       </para>
-      </listitem>
-     </varlistentry>
-
-    <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
-      <term><literal>CREATE COLLATION special (provider = icu, locale = 'en@colCaseFirst=upper;colReorder=grek-latn');</literal></term>
-      <listitem>
-       <para>
-        Combines both of the above options.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="collation-managing-create-icu-en-u-kn-true">
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en-u-kn-true');</literal></term>
-      <term><literal>CREATE COLLATION numeric (provider = icu, locale = 'en@colNumeric=yes');</literal></term>
-      <listitem>
-       <para>
-        Numeric ordering, sorts sequences of digits by their numeric value,
-        for example: <literal>A-21</literal> &lt; <literal>A-123</literal>
-        (also known as natural sort).
-       </para>
-      </listitem>
-     </varlistentry>
-    </variablelist>
-
-    See <ulink url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
-    Technical Standard #35</ulink>
-    and <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink> for
-    details.  The list of possible collation types (<literal>co</literal>
-    subtag) can be found in
-    the <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
-    repository</ulink>.
-   </para>
+    <para>
+     ICU collations can be created like:
 
-   <para>
-    Note that while this system allows creating collations that <quote>ignore
-    case</quote> or <quote>ignore accents</quote> or similar (using the
-    <literal>ks</literal> key), in order for such collations to act in a
-    truly case- or accent-insensitive manner, they also need to be declared as not
-    <firstterm>deterministic</firstterm> in <command>CREATE COLLATION</command>;
-    see <xref linkend="collation-nondeterministic"/>.
-    Otherwise, any strings that compare equal according to the collation but
-    are not byte-wise equal will be sorted according to their byte values.
-   </para>
+<programlisting>
+CREATE COLLATION german (provider = icu, locale = 'de-DE');
+</programlisting>
 
-   <note>
+     ICU locales are specified as a BCP 47 <link
+     linkend="icu-language-tag">Language Tag</link>, but can also accept most
+     libc-style locale names. If possible, libc-style locale names are
+     transformed into language tags.
+    </para>
     <para>
-     By design, ICU will accept almost any string as a locale name and match
-     it to the closest locale it can provide, using the fallback procedure
-     described in its documentation.  Thus, there will be no direct feedback
-     if a collation specification is composed using features that the given
-     ICU installation does not actually support.  It is therefore recommended
-     to create application-level test cases to check that the collation
-     definitions satisfy one's requirements.
+     New ICU collations can customize collation behavior extensively by
+     including collation attributes in the langugage tag. See <xref
+     linkend="icu-custom-collations"/> for details and examples.
     </para>
-   </note>
    </sect4>
-
    <sect4 id="collation-copy">
    <title>Copying Collations</title>
 
@@ -1072,6 +1098,421 @@ CREATE COLLATION ignore_accents (provider = icu, locale = 'und-u-ks-level1-kc-tr
     </tip>
    </sect3>
   </sect2>
+  <sect2 id="icu-custom-collations">
+   <title>ICU Custom Collations</title>
+
+   <para>
+    ICU allows extensive control over collation behavior by defining new
+    collations with collation settings as a part of the language tag. These
+    settings can modify the collation order to suit a variety of needs. For
+    instance:
+
+<programlisting>
+-- ignore differences in accents and case
+CREATE COLLATION ignore_accent_case (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ks-level1');
+SELECT 'Å' = 'A' COLLATE ignore_accent_case; -- true
+SELECT 'z' = 'Z' COLLATE ignore_accent_case; -- true
+
+-- upper case letters sort before lower case.
+CREATE COLLATION upper_first (PROVIDER=icu, LOCALE = 'und-u-kf-upper');
+SELECT 'B' &lt; 'b' COLLATE upper_first; -- true
+
+-- treat digits numerically and ignore punctuation
+CREATE COLLATION num_ignore_punct (PROVIDER = icu, DETERMINISTIC = false, LOCALE = 'und-u-ka-shifted-kn');
+SELECT 'id-45' &lt; 'id-123' COLLATE num_ignore_punct; -- true
+SELECT 'w;x*y-z' = 'wxyz' COLLATE num_ignore_punct; -- true
+</programlisting>
+
+    Many of the available options are described in <xref
+    linkend="icu-collation-settings"/>, or see <xref
+    linkend="icu-external-references"/> for more details.
+   </para>
+   <sect3 id="icu-collation-comparison-levels">
+    <title>ICU Comparison Levels</title>
+    <para>
+     Comparison of two strings (collation) in ICU is determined by a
+     multi-level process, where textual features are grouped into
+     "levels". Treatment of each level is controlled by the <link
+     linkend="icu-collation-settings-table">collation settings</link>. Higher
+     levels correspond to finer textual features.
+    </para>
+    <para>
+     <table id="icu-collation-levels">
+      <title>ICU Collation Levels</title>
+      <tgroup cols="3">
+       <thead>
+        <row>
+         <entry>Level</entry>
+         <entry>Description</entry>
+         <entry><literal>'f' = 'f'</literal></entry>
+         <entry><literal>'ab' = U&amp;'a\2063b'</literal></entry>
+         <entry><literal>'x-y' = 'x_y'</literal></entry>
+         <entry><literal>'g' = 'G'</literal></entry>
+         <entry><literal>'n' = 'ñ'</literal></entry>
+         <entry><literal>'y' = 'z'</literal></entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry>level1</entry>
+         <entry>Base Character</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level2</entry>
+         <entry>Accents</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level3</entry>
+         <entry>Case/Variants</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>level4</entry>
+         <entry>Punctuation</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+        <row>
+         <entry>identic</entry>
+         <entry>All</entry>
+         <entry><literal>true</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+
+     The above table shows which textual feature differences are
+     considered significant when determining equality at the given level. The
+     unicode character <literal>U+2063</literal> is an invisible separator,
+     and as seen in the table, is ignored for at all levels of comparison less
+     than <literal>identic</literal>.
+    </para>
+    <para>
+     At every level, even with full normalization off, basic normalization is
+     performed. For example, <literal>'á'</literal> may be composed of the
+     code points <literal>U&amp;'\0061\0301'</literal> or the single code
+     point <literal>U&amp;'\00E1'</literal>, and those sequences will be
+     considered equal even at the <literal>identic</literal> level. To treat
+     any difference in code point representation as distinct, use a collation
+     created with <symbol>DETERMINISTIC</symbol> set to
+     <literal>true</literal>.
+    </para>
+    <sect4 id="icu-collation-level-examples">
+     <title>Collation Level Examples</title>
+     <para>
+
+<programlisting>
+CREATE COLLATION level3 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level3');
+CREATE COLLATION level4 (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-level4');
+CREATE COLLATION identic (PROVIDER=icu, DETERMINISTIC=false, LOCALE='und-u-ka-shifted-ks-identic');
+
+-- invisible separator ignored at all levels except identic
+SELECT 'ab' = U&amp;'a\2063b' COLLATE level4; -- true
+SELECT 'ab' = U&amp;'a\2063b' COLLATE identic; -- false
+
+-- punctuation ignored at level3 but not at level 4
+SELECT 'x-y' = 'x_y' COLLATE level3; -- true
+SELECT 'x-y' = 'x_y' COLLATE level4; -- false
+</programlisting>
+
+     </para>
+    </sect4>
+   </sect3>
+   <sect3 id="icu-collation-settings">
+    <title>Collation Settings for an ICU Locale</title>
+    <para>
+     <table id="icu-collation-settings-table">
+      <title>ICU Collation Settings</title>
+      <tgroup cols="4">
+       <thead>
+        <row>
+         <entry>Key</entry>
+         <entry>Values</entry>
+         <entry>Default</entry>
+         <entry>Description</entry>
+        </row>
+       </thead>
+       <tbody>
+        <row>
+         <entry><literal>co</literal></entry>
+         <entry><literal>emoji</literal>, <literal>phonebk</literal>, <literal>standard</literal>, <replaceable>...</replaceable></entry>
+         <entry><literal>standard</literal></entry>
+         <entry>
+          Collation type. See <xref linkend="icu-external-references"/> for additional options and details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ks</literal></entry>
+         <entry><literal>level1</literal>, <literal>level2</literal>, <literal>level3</literal>, <literal>level4</literal>, <literal>identic</literal></entry>
+         <entry><literal>level3</literal></entry>
+         <entry>
+          Sensitivity (or "strength") when determining equality, with
+          <literal>level1</literal> the least sensitive to differences and
+          <literal>identic</literal> the most sensitive to differences. See
+          <xref linkend="icu-collation-levels"/> for details.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>ka</literal></entry>
+         <entry><literal>noignore</literal>, <literal>shifted</literal></entry>
+         <entry><literal>noignore</literal></entry>
+         <entry>
+          If set to <literal>shifted</literal>, causes some characters
+          (e.g. punctuation or space) to be ignored in comparison. Key
+          <literal>ks</literal> must be set to <literal>level3</literal> or
+          lower to take effect. Set key <literal>kv</literal> to control which
+          character classes are ignored.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kb</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          Backwards comparison for the level 2 differences. For example,
+          locale <literal>und-u-kb</literal> sorts <literal>'àe'</literal>
+          before <literal>'aé'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kk</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Enable full normalization; may affect performance. Basic
+           normalization is performed even when set to
+           <literal>false</literal>. Locales for languages that require full
+           normalization typically enable it by default.
+          </para>
+          <para>
+           Full normalization is important in some cases, such as when
+           multiple accents are applied to a single character. For instance,
+           <literal>'ệ'</literal> can be composed of code points
+           <literal>U&amp;'\0065\0323\0302'</literal> or
+           <literal>U&amp;'\0065\0302\0323'</literal>. With full normalization
+           on, these code point sequences are treated as equal; otherwise they
+           are unequal.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kc</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          <para>
+           Separates case into a "level 2.5" that falls between accents and
+           other level 3 features.
+          </para>
+          <para>
+           If set to <literal>true</literal> and <literal>ks</literal> is set
+           to <literal>level1</literal>, will ignore accents but take case
+           into account.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kf</literal></entry>
+         <entry>
+          <literal>upper</literal>, <literal>lower</literal>,
+          <literal>false</literal>
+         </entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>upper</literal>, upper case sorts before lower
+          case. If set to <literal>lower</literal>, lower case sorts before
+          upper case. If set to <literal>false</literal>, the sort depends on
+          the rules of the locale.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kn</literal></entry>
+         <entry><literal>true</literal>, <literal>false</literal></entry>
+         <entry><literal>false</literal></entry>
+         <entry>
+          If set to <literal>true</literal>, numbers within a string are
+          treated as a single numeric value rather than a sequence of
+          digits. For example, <literal>'id-45'</literal> sorts before
+          <literal>'id-123'</literal>.
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kr</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>,
+          <literal>digit</literal>, <replaceable>script-id</replaceable>
+         </entry>
+         <entry></entry>
+         <entry>
+          <para>
+           Set to one or more of the valid values, or any BCP 47
+           <replaceable>script-id</replaceable>, e.g. <literal>latn</literal>
+           ("Latin") or <literal>grek</literal> ("Greek"). Multiple values are
+           separated by "<literal>-</literal>".
+          </para>
+          <para>
+           Redefines the ordering of classes of characters; those characters
+           belonging to a class earlier in the list sort before characters
+           belonging to a class later in the list. For instance, the value
+           <literal>digit-currency-space</literal> (as part of a language tag
+           like <literal>und-u-kr-digit-currency-space</literal>) sorts
+           punctuation before digits and spaces.
+          </para>
+         </entry>
+        </row>
+        <row>
+         <entry><literal>kv</literal></entry>
+         <entry>
+          <literal>space</literal>, <literal>punct</literal>,
+          <literal>symbol</literal>, <literal>currency</literal>
+         </entry>
+         <entry><literal>punct</literal></entry>
+         <entry>
+          Classes of characters ignored during comparison at level 3. Setting
+          to a later value includes earlier values;
+          e.g. <literal>symbol</literal> also includes
+          <literal>punct</literal> and <literal>space</literal> in the
+          characters to be ignored. Key <literal>ka</literal> must be set to
+          <literal>shifted</literal> and key <literal>ks</literal> must be set
+          to <literal>level3</literal> or lower to take effect.
+         </entry>
+        </row>
+       </tbody>
+      </tgroup>
+     </table>
+      Defaults may depend on locale. The above table is not meant to be
+      complete. See <xref linkend="icu-external-references"/> for additional
+      options and details.
+    </para>
+    <note>
+     <para>
+      For many collation settings, you must create the collation with
+      <option>DETERMINISTIC</option> set to <literal>false</literal> for the
+      setting to have the desired effect (see <xref
+      linkend="collation-nondeterministic"/>). Additionally, some settings
+      only take effect when the key <literal>ka</literal> is set to
+      <literal>shifted</literal> (see <xref
+      linkend="icu-collation-settings-table"/>).
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="icu-locale-examples">
+    <title>Examples</title>
+    <para>
+     <variablelist>
+      <varlistentry id="collation-managing-create-icu-de-u-co-phonebk-x-icu">
+       <term><literal>CREATE COLLATION "de-u-co-phonebk-x-icu" (provider = icu, locale = 'de-u-co-phonebk');</literal></term>
+       <listitem>
+        <para>German collation with phone book collation type</para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-und-u-co-emoji-x-icu">
+       <term><literal>CREATE COLLATION "und-u-co-emoji-x-icu" (provider = icu, locale = 'und-u-co-emoji');</literal></term>
+       <listitem>
+        <para>
+         Root collation with Emoji collation type, per Unicode Technical Standard #51
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kr-grek-latn">
+       <term><literal>CREATE COLLATION latinlast (provider = icu, locale = 'en-u-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Sort Greek letters before Latin ones.  (The default is Latin before Greek.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper">
+       <term><literal>CREATE COLLATION upperfirst (provider = icu, locale = 'en-u-kf-upper');</literal></term>
+       <listitem>
+        <para>
+         Sort upper-case letters before lower-case letters.  (The default is
+         lower-case letters first.)
+        </para>
+       </listitem>
+      </varlistentry>
+
+      <varlistentry id="collation-managing-create-icu-en-u-kf-upper-kr-grek-latn">
+       <term><literal>CREATE COLLATION special (provider = icu, locale = 'en-u-kf-upper-kr-grek-latn');</literal></term>
+       <listitem>
+        <para>
+         Combines both of the above options.
+        </para>
+       </listitem>
+      </varlistentry>
+     </variablelist>
+    </para>
+   </sect3>
+   <sect3 id="icu-external-references">
+    <title>External References for ICU</title>
+    <para>
+     This section (<xref linkend="icu-custom-collations"/>) is only a brief
+     overview of ICU behavior and language tags. Refer to the following
+     documents for technical details, additional options, and new behavior:
+    </para>
+    <itemizedlist>
+     <listitem>
+      <para>
+       <ulink
+           url="https://www.unicode.org/reports/tr35/tr35-collation.html">Unicode
+       Technical Standard #35</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://tools.ietf.org/html/bcp47">BCP 47</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://github.com/unicode-org/cldr/blob/master/common/bcp47/collation.xml">CLDR
+       repository</ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/locale/"></ulink>
+      </para>
+     </listitem>
+     <listitem>
+      <para>
+       <ulink url="https://unicode-org.github.io/icu/userguide/collation/api.html"></ulink>
+      </para>
+     </listitem>
+    </itemizedlist>
+   </sect3>
+  </sect2>
  </sect1>
 
  <sect1 id="multibyte">
-- 
2.34.1

From dc7200153a9ac65c2518b32b789d1a9dc4454850 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 8 May 2023 13:48:01 -0700
Subject: [PATCH v6 2/5] ICU: for locale "C", automatically use "none" provider
 instead.

Postgres expects locale C to be optimizable to simple locale-unaware
byte operations; while ICU does not recognize the locale "C" at all,
and falls back to the root locale.

If the user specifies locale "C" when creating a new collation or a
new database with the ICU provider, automatically switch it to the
"none" provider.

If provider is libc, behavior is unchanged.
---
 doc/src/sgml/charset.sgml                     |  6 +++
 doc/src/sgml/ref/create_collation.sgml        |  6 +++
 doc/src/sgml/ref/create_database.sgml         |  5 +++
 doc/src/sgml/ref/createdb.sgml                |  5 +++
 doc/src/sgml/ref/initdb.sgml                  |  5 +++
 src/backend/commands/collationcmds.c          | 17 ++++++++
 src/backend/commands/dbcommands.c             | 21 ++++++++++
 src/bin/initdb/initdb.c                       | 10 +++++
 src/bin/initdb/t/001_initdb.pl                | 39 +++++++++++++++++++
 src/bin/scripts/createdb.c                    | 11 ++++++
 src/bin/scripts/t/020_createdb.pl             | 12 ++++++
 .../regress/expected/collate.icu.utf8.out     | 14 +++++--
 src/test/regress/sql/collate.icu.utf8.sql     |  6 +++
 13 files changed, 154 insertions(+), 3 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 7a791a2b7c..68bad646e9 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -405,6 +405,12 @@ initdb --locale-provider=icu --icu-locale=en
      change in results. <literal>LC_COLLATE</literal> and
      <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
     </para>
+    <para>
+     The ICU provider does not accept the <literal>C</literal>
+     locale. Commands that create collations or database with the
+     <literal>icu</literal> provider and ICU locale <literal>C</literal> use
+     the provider <literal>none</literal> instead.
+    </para>
     <note>
      <para>
       For the ICU provider, results may depend on the version of the ICU
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index 5489ae7413..1ac41831d8 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -126,6 +126,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
        <literal>libc</literal> is the default.  See <xref
        linkend="locale-providers"/> for details.
       </para>
+      <para>
+       If the provider is <literal>icu</literal> and the locale is
+       <literal>C</literal> or <literal>POSIX</literal>, the provider is
+       automatically set to <literal>none</literal>; as the ICU provider
+       doesn't support an ICU locale of <literal>C</literal>.
+      </para>
      </listitem>
     </varlistentry>
 
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 60b9da0952..c730d02e15 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -190,6 +190,11 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
        <para>
         Specifies the ICU locale ID if the ICU locale provider is used.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index 326a371d34..7c573e848a 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -154,6 +154,11 @@ PostgreSQL documentation
         Specifies the ICU locale ID to be used in this database, if the
         ICU locale provider is selected.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index e604ab48b7..76993acdfe 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -250,6 +250,11 @@ PostgreSQL documentation
         Specifies the ICU locale when the ICU provider is used. Locale support
         is described in <xref linkend="locale"/>.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>none</literal>, as the ICU
+        provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
        <para>
         If this option is not specified, the locale is inherited from the
         environment in which <command>initdb</command> runs. The environment's
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index aeaf6c419e..8bc6f8347d 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -254,6 +254,23 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		if (lcctypeEl)
 			collctype = defGetString(lcctypeEl);
 
+		/*
+		 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+		 * optimizable to byte operations (memcmp(), pg_ascii_tolower(),
+		 * etc.); transform into the "none" provider. Don't transform during
+		 * binary upgrade.
+		 */
+		if (!IsBinaryUpgrade && collprovider == COLLPROVIDER_ICU &&
+			colliculocale && (pg_strcasecmp(colliculocale, "C") == 0 ||
+							  pg_strcasecmp(colliculocale, "POSIX") == 0))
+		{
+			ereport(NOTICE,
+					(errmsg("using locale provider \"none\" for ICU locale \"%s\"",
+							colliculocale)));
+			colliculocale = NULL;
+			collprovider = COLLPROVIDER_NONE;
+		}
+
 		if (collprovider == COLLPROVIDER_LIBC)
 		{
 			if (!collcollate)
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 9e73f54803..6dc737aebb 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1043,6 +1043,27 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
 
+	/*
+	 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+	 * optimizable to byte operations (memcmp(), pg_ascii_tolower(), etc.);
+	 * transform into the "none" provider.
+	 *
+	 * Don't transform during binary upgrade or when both the provider and ICU
+	 * locale are unchanged from the template.
+	 */
+	if (!IsBinaryUpgrade && dblocprovider == COLLPROVIDER_ICU &&
+		(src_locprovider != COLLPROVIDER_ICU ||
+		 strcmp(dbiculocale, src_iculocale) != 0) &&
+		dbiculocale && (pg_strcasecmp(dbiculocale, "C") == 0 ||
+						pg_strcasecmp(dbiculocale, "POSIX") == 0))
+	{
+		ereport(NOTICE,
+				(errmsg("using locale provider \"none\" for ICU locale \"%s\"",
+						dbiculocale)));
+		dbiculocale = NULL;
+		dblocprovider = COLLPROVIDER_NONE;
+	}
+
 	if (dblocprovider == COLLPROVIDER_ICU)
 	{
 		if (!(is_encoding_supported_by_icu(encoding)))
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 4a6cad3cb9..e5ec2a243e 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2440,6 +2440,16 @@ setlocales(void)
 			lc_messages = locale;
 	}
 
+	if (icu_locale && locale_provider == COLLPROVIDER_ICU &&
+		(pg_strcasecmp(icu_locale, "C") == 0 ||
+		 pg_strcasecmp(icu_locale, "POSIX") == 0))
+	{
+		pg_log_info("using locale provider \"none\" for ICU locale \"%s\"",
+					 icu_locale);
+		icu_locale = NULL;
+		locale_provider = COLLPROVIDER_NONE;
+	}
+
 	/*
 	 * canonicalize locale names, and obtain any missing values from our
 	 * current environment
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index fe6d224e5b..ea92b08511 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -111,6 +111,45 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'option --icu-locale');
 
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			"$tempdir/data4a"
+		],
+		'option --icu-locale=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--locale=C',
+			"$tempdir/data4b"
+		],
+		'option --icu-locale=C --locale=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-collate=C',
+			"$tempdir/data4c"
+		],
+		'option --icu-locale=C --lc-collate=C');
+
+	# transformed to provider=none
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-ctype=C',
+			"$tempdir/data4d"
+		],
+		'option --icu-locale=C --lc-ctype=C');
+
 	command_fails_like(
 		[
 			'initdb',                '--no-sync',
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index 79367d933b..9caf9190cf 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -172,6 +172,17 @@ main(int argc, char *argv[])
 			lc_collate = locale;
 	}
 
+	if (locale_provider && pg_strcasecmp(locale_provider, "icu") == 0 &&
+		icu_locale &&
+		(pg_strcasecmp(icu_locale, "C") == 0 ||
+		 pg_strcasecmp(icu_locale, "POSIX") == 0))
+	{
+		pg_log_info("using locale provider \"none\" for ICU locale \"%s\"",
+					 icu_locale);
+		icu_locale = NULL;
+		locale_provider = "none";
+	}
+
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index 5aa658b671..eb3682f0fd 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -75,6 +75,18 @@ if ($ENV{with_icu} eq 'yes')
 	$node2->command_ok(
 		[ 'createdb', '-T', 'template0', '--icu-locale', 'en-US', 'foobar56' ],
 		'create database with icu locale from template database with icu provider');
+
+	# transformed into provider "none"
+	$node->command_ok(
+		[ 'createdb', '-T', 'template0', '--locale-provider=icu', '--icu-locale=C',
+		  'test_none_icu1' ],
+		'create database with provider "icu" and ICU_LOCALE="C"');
+
+	# transformed into provider "none"
+	$node->command_ok(
+		[ 'createdb', '-T', 'template0', '--locale-provider=icu', '--icu-locale=C',
+		  '--lc-ctype=C', 'test_none_icu_2' ],
+		'create database with provider "icu" and ICU_LOCALE="C" and LC_CTYPE=C');
 }
 else
 {
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index c658ee1404..7c186e9f69 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1035,6 +1035,9 @@ BEGIN
 END
 $$;
 RESET client_min_messages;
+-- uses "none" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+NOTICE:  using locale provider "none" for ICU locale "C"
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 SET icu_validation_level = ERROR;
@@ -1058,8 +1061,11 @@ SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
  test0
  test1
  test5
-(3 rows)
+ testc
+(4 rows)
 
+DROP COLLATION test1;
+CREATE COLLATION test1 (provider = icu, locale = 'und');
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
 ERROR:  collation "test11" already exists in schema "collate_tests"
@@ -1079,7 +1085,8 @@ SELECT collname, nspname, obj_description(pg_collation.oid, 'pg_collation')
  test0    | collate_tests | US English
  test11   | test_schema   | 
  test5    | collate_tests | 
-(3 rows)
+ testc    | collate_tests | 
+(4 rows)
 
 DROP COLLATION test0, test_schema.test11, test5;
 DROP COLLATION test0; -- fail
@@ -1089,7 +1096,8 @@ NOTICE:  collation "test0" does not exist, skipping
 SELECT collname FROM pg_collation WHERE collname LIKE 'test%';
  collname 
 ----------
-(0 rows)
+ testc
+(1 row)
 
 DROP SCHEMA test_schema;
 DROP ROLE regress_test_role;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 7bd0901281..e59200df9a 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -375,6 +375,9 @@ $$;
 
 RESET client_min_messages;
 
+-- uses "none" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
@@ -388,6 +391,9 @@ CREATE COLLATION test5 FROM test0;
 
 SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
 
+DROP COLLATION test1;
+CREATE COLLATION test1 (provider = icu, locale = 'und');
+
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
 ALTER COLLATION test1 RENAME TO test22; -- fail
-- 
2.34.1

From c04053021eaa6db480143393a7de83525a8f4f7e Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v6 3/5] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 +++--
 doc/src/sgml/ref/createdb.sgml                |  5 +++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++---
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++++----
 src/bin/initdb/initdb.c                       | 11 ++++++---
 src/bin/scripts/createdb.c                    | 13 ++++-------
 src/bin/scripts/t/020_createdb.pl             |  4 ++--
 src/test/icu/t/010_database.pl                | 23 ++++++++++++-------
 .../regress/expected/collate.icu.utf8.out     | 22 +++++++++---------
 10 files changed, 65 insertions(+), 43 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index c730d02e15..dc57ba0c8b 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index 7c573e848a..7991153ecc 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 76993acdfe..d9ef21c422 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 8bc6f8347d..21615746f9 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -302,7 +302,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 				if (langtag && strcmp(colliculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, colliculocale)));
 
 					colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 6dc737aebb..154f20573c 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1019,7 +1019,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1033,12 +1038,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1094,7 +1101,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 			if (langtag && strcmp(dbiculocale, langtag) != 0)
 			{
 				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
+						(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 								langtag, dbiculocale)));
 
 				dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index e5ec2a243e..f0827154cd 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2164,7 +2164,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2406,7 +2410,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale_provider == COLLPROVIDER_NONE)
 	{
@@ -2438,6 +2442,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	if (icu_locale && locale_provider == COLLPROVIDER_ICU &&
@@ -3331,7 +3337,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index 9caf9190cf..51c4bb3592 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (locale_provider && pg_strcasecmp(locale_provider, "icu") == 0 &&
 		icu_locale &&
 		(pg_strcasecmp(icu_locale, "C") == 0 ||
@@ -230,6 +222,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index eb3682f0fd..81a9931c09 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -167,7 +167,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -175,7 +175,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 715b1bffd6..df4af00afe 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,16 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8});
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 7c186e9f69..cf1852c89d 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1202,9 +1202,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1212,7 +1212,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1229,12 +1229,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1243,7 +1243,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1271,13 +1271,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1327,9 +1327,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 2a857a2cb080dbc015c59b89acbb195ae7991a99 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Thu, 11 May 2023 12:54:31 -0700
Subject: [PATCH v6 4/5] Add default_collation_provider GUC.

Controls default collation provider for CREATE COLLATION. Does not
affect CREATE DATABASE, which gets its default from the template
database.
---
 doc/src/sgml/config.sgml                      | 17 +++++++++++++
 doc/src/sgml/ref/create_collation.sgml        | 15 ++++++++---
 src/backend/commands/collationcmds.c          |  8 +++++-
 src/backend/utils/misc/guc_tables.c           | 18 +++++++++++++
 src/backend/utils/misc/postgresql.conf.sample |  4 +++
 src/include/commands/collationcmds.h          |  2 ++
 .../regress/expected/collate.icu.utf8.out     | 25 +++++++++++++++++++
 src/test/regress/sql/collate.icu.utf8.sql     | 13 ++++++++++
 8 files changed, 97 insertions(+), 5 deletions(-)

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index 18ce06729b..58a1046340 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -9820,6 +9820,23 @@ SET XML OPTION { DOCUMENT | CONTENT };
       </listitem>
      </varlistentry>
 
+     <varlistentry id="guc-default-collation-provider" xreflabel="default_collation_provider">
+      <term><varname>default_collation_provider</varname> (<type>enum</type>)
+      <indexterm>
+       <primary><varname>default_collation_provider</varname> configuration parameter</primary>
+      </indexterm>
+      </term>
+      <listitem>
+       <para>
+        Default collation provider for <command>CREATE
+        COLLATION</command>. Does not affect <command>CREATE
+        DATABASE</command>, which gets the default collation provider from the
+        template database. Valid values are <literal>icu</literal> and
+        <literal>libc</literal>. The default is <literal>libc</literal>.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry id="guc-icu-validation-level" xreflabel="icu_validation_level">
       <term><varname>icu_validation_level</varname> (<type>enum</type>)
       <indexterm>
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index 1ac41831d8..c9b3e6e218 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -121,10 +121,17 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
       <para>
        Specifies the provider to use for locale services associated with this
        collation.  Possible values are <literal>none</literal>,
-       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
-       (if the server was built with ICU support) or <literal>libc</literal>.
-       <literal>libc</literal> is the default.  See <xref
-       linkend="locale-providers"/> for details.
+       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
+       the server was built with ICU support) or <literal>libc</literal>.  See
+       <xref linkend="locale-providers"/> for details.
+      </para>
+      <para>
+       If <replaceable>provider</replaceable> is not specified, and
+       <replaceable>lc_collate</replaceable> or
+       <replaceable>lc_ctype</replaceable> is specified, the
+       <literal>libc</literal> provider is used. Otherwise, the default
+       provider is controlled by <xref
+       linkend="guc-default-collation-provider"/>.
       </para>
       <para>
        If the provider is <literal>icu</literal> and the locale is
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 21615746f9..25e8d32fd9 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -47,6 +47,7 @@ typedef struct
 	int			enc;			/* encoding */
 } CollAliasData;
 
+int		default_collation_provider = (int) COLLPROVIDER_LIBC;
 
 /*
  * CREATE COLLATION
@@ -228,7 +229,12 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+		{
+			if (lccollateEl || lcctypeEl)
+				collprovider = COLLPROVIDER_LIBC;
+			else
+				collprovider = (char) default_collation_provider;
+		}
 
 		if (collprovider == COLLPROVIDER_NONE
 			&& (localeEl || lccollateEl || lcctypeEl))
diff --git a/src/backend/utils/misc/guc_tables.c b/src/backend/utils/misc/guc_tables.c
index 844781a7f5..901cfda819 100644
--- a/src/backend/utils/misc/guc_tables.c
+++ b/src/backend/utils/misc/guc_tables.c
@@ -35,8 +35,10 @@
 #include "access/xlogrecovery.h"
 #include "archive/archive_module.h"
 #include "catalog/namespace.h"
+#include "catalog/pg_collation.h"
 #include "catalog/storage.h"
 #include "commands/async.h"
+#include "commands/collationcmds.h"
 #include "commands/tablespace.h"
 #include "commands/trigger.h"
 #include "commands/user.h"
@@ -166,6 +168,12 @@ static const struct config_enum_entry intervalstyle_options[] = {
 	{NULL, 0, false}
 };
 
+static const struct config_enum_entry collation_provider_options[] = {
+	{"icu", (int) 'i', false},
+	{"libc", (int) 'c', false},
+	{NULL, 0, false}
+};
+
 static const struct config_enum_entry icu_validation_level_options[] = {
 	{"disabled", -1, false},
 	{"debug5", DEBUG5, false},
@@ -4683,6 +4691,16 @@ struct config_enum ConfigureNamesEnum[] =
 		NULL, NULL, NULL
 	},
 
+	{
+		{"default_collation_provider", PGC_USERSET, CLIENT_CONN_LOCALE,
+		 gettext_noop("Default collation provider for CREATE COLLATION."),
+		 NULL
+		},
+		&default_collation_provider,
+		(int) COLLPROVIDER_LIBC, collation_provider_options,
+		NULL, NULL, NULL
+	},
+
 	{
 		{"icu_validation_level", PGC_USERSET, CLIENT_CONN_LOCALE,
 		 gettext_noop("Log level for reporting invalid ICU locale strings."),
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index c8018da04a..c1f247378d 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -734,6 +734,10 @@
 #lc_numeric = 'C'			# locale for number formatting
 #lc_time = 'C'				# locale for time formatting
 
+#default_collation_provider = 'libc'	# default collation provider
+					# for CREATE COLLATION
+					# (none, icu, libc)
+
 #icu_validation_level = WARNING		# report ICU locale validation
 					# errors at the given level
 
diff --git a/src/include/commands/collationcmds.h b/src/include/commands/collationcmds.h
index b76c7b3dc3..f54389525d 100644
--- a/src/include/commands/collationcmds.h
+++ b/src/include/commands/collationcmds.h
@@ -18,6 +18,8 @@
 #include "catalog/objectaddress.h"
 #include "parser/parse_node.h"
 
+extern PGDLLIMPORT int default_collation_provider;
+
 extern ObjectAddress DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_exists);
 extern void IsThereCollationInNamespace(const char *collname, Oid nspOid);
 extern ObjectAddress AlterCollation(AlterCollationStmt *stmt);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index cf1852c89d..ea96e27f45 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1038,6 +1038,31 @@ RESET client_min_messages;
 -- uses "none" provider instead
 CREATE COLLATION testc (provider = icu, locale='C');
 NOTICE:  using locale provider "none" for ICU locale "C"
+SET default_collation_provider = 'libc';
+CREATE COLLATION def_libc (LOCALE = 'C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+ collname | collprovider 
+----------+--------------
+ def_libc | c
+(1 row)
+
+DROP COLLATION def_libc;
+SET default_collation_provider = 'icu';
+CREATE COLLATION def_icu (LOCALE = 'und');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_icu';
+ collname | collprovider 
+----------+--------------
+ def_icu  | i
+(1 row)
+
+CREATE COLLATION def_libc (LC_COLLATE = 'C', LC_CTYPE='C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+ collname | collprovider 
+----------+--------------
+ def_libc | c
+(1 row)
+
+RESET default_collation_provider;
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 SET icu_validation_level = ERROR;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index e59200df9a..ee607ca3a5 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -378,6 +378,19 @@ RESET client_min_messages;
 -- uses "none" provider instead
 CREATE COLLATION testc (provider = icu, locale='C');
 
+SET default_collation_provider = 'libc';
+CREATE COLLATION def_libc (LOCALE = 'C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+DROP COLLATION def_libc;
+
+SET default_collation_provider = 'icu';
+CREATE COLLATION def_icu (LOCALE = 'und');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_icu';
+CREATE COLLATION def_libc (LC_COLLATE = 'C', LC_CTYPE='C');
+SELECT collname, collprovider FROM pg_collation WHERE collname='def_libc';
+
+RESET default_collation_provider;
+
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
-- 
2.34.1

From de37bfb02dcc41c2e932a788ba10a05e5a539870 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 1 May 2023 15:38:29 -0700
Subject: [PATCH v6 1/5] Introduce collation provider "none".

Provides locale-unaware semantics that are implemented as fast byte
operations in Postgres, independent of the operating system or any
provider libraries.

Equivalent (in semantics and implementation) to the libc provider with
locale "C", except that LC_COLLATE and LC_CTYPE can be set
independently.

Use provider "none" for built-in collation "ucs_basic" instead of
libc.
---
 doc/src/sgml/charset.sgml              | 87 +++++++++++++++++++++-----
 doc/src/sgml/ref/create_collation.sgml |  2 +-
 doc/src/sgml/ref/create_database.sgml  |  2 +-
 doc/src/sgml/ref/createdb.sgml         |  2 +-
 doc/src/sgml/ref/initdb.sgml           |  2 +-
 src/backend/catalog/pg_collation.c     |  7 ++-
 src/backend/commands/collationcmds.c   | 84 +++++++++++++++++++++----
 src/backend/commands/dbcommands.c      | 69 +++++++++++++++++---
 src/backend/utils/adt/pg_locale.c      | 27 +++++++-
 src/backend/utils/init/postinit.c      | 10 ++-
 src/bin/initdb/initdb.c                | 33 +++++++++-
 src/bin/initdb/t/001_initdb.pl         | 29 +++++++++
 src/bin/pg_dump/pg_dump.c              |  8 ++-
 src/bin/pg_upgrade/t/002_pg_upgrade.pl | 18 +++++-
 src/bin/psql/describe.c                |  2 +-
 src/bin/scripts/createdb.c             |  2 +-
 src/bin/scripts/t/020_createdb.pl      | 29 +++++++++
 src/include/catalog/pg_collation.dat   |  3 +-
 src/include/catalog/pg_collation.h     |  3 +
 src/test/regress/expected/collate.out  | 10 ++-
 src/test/regress/sql/collate.sql       |  6 ++
 21 files changed, 373 insertions(+), 62 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 9db14649aa..7a791a2b7c 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -342,22 +342,14 @@ initdb --locale=sv_SE
    <title>Locale Providers</title>
 
    <para>
-    <productname>PostgreSQL</productname> supports multiple <firstterm>locale
-    providers</firstterm>.  This specifies which library supplies the locale
-    data.  One standard provider name is <literal>libc</literal>, which uses
-    the locales provided by the operating system C library.  These are the
-    locales used by most tools provided by the operating system.  Another
-    provider is <literal>icu</literal>, which uses the external
-    ICU<indexterm><primary>ICU</primary></indexterm> library.  ICU locales can
-    only be used if support for ICU was configured when PostgreSQL was built.
+    A locale provider specifies which library defines the locale behavior for
+    collations and character classifications.
    </para>
 
    <para>
     The commands and tools that select the locale settings, as described
-    above, each have an option to select the locale provider.  The examples
-    shown earlier all use the <literal>libc</literal> provider, which is the
-    default.  Here is an example to initialize a database cluster using the
-    ICU provider:
+    above, each have an option to select the locale provider. Here is an
+    example to initialize a database cluster using the ICU provider:
 <programlisting>
 initdb --locale-provider=icu --icu-locale=en
 </programlisting>
@@ -370,12 +362,73 @@ initdb --locale-provider=icu --icu-locale=en
    </para>
 
    <para>
-    Which locale provider to use depends on individual requirements.  For most
-    basic uses, either provider will give adequate results.  For the libc
-    provider, it depends on what the operating system offers; some operating
-    systems are better than others.  For advanced uses, ICU offers more locale
-    variants and customization options.
+    Regardless of the locale provider, the operating system is still used to
+    provide some locale-aware behavior, such as messages (see <xref
+    linkend="guc-lc-messages"/>).
    </para>
+
+   <para>
+    The available locale providers are listed below.
+   </para>
+
+   <sect3 id="locale-provider-none">
+    <title>None</title>
+    <para>
+     The <literal>none</literal> provider uses simple built-in operations
+     which are not locale-aware.
+    </para>
+    <para>
+     The collation and character classification behavior is equivalent to
+     using the <literal>libc</literal> provider with locale
+     <literal>C</literal>, except that <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently.
+    </para>
+    <note>
+     <para>
+      When using the <literal>none</literal> locale provider, behavior may
+      depend on the database encoding.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-icu">
+    <title>ICU</title>
+    <para>
+     The <literal>icu</literal> provider uses the external
+     ICU<indexterm><primary>ICU</primary></indexterm>
+     library. <productname>PostgreSQL</productname> must have been configured
+     with support.
+    </para>
+    <para>
+     ICU provides collation and character classification behavior that is
+     independent of the operating system and database encoding, which is
+     preferable if you expect to transition to other platforms without any
+     change in results. <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
+    </para>
+    <note>
+     <para>
+      For the ICU provider, results may depend on the version of the ICU
+      library used, as it is updated to reflect changes in natural language
+      over time.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-libc">
+    <title>libc</title>
+    <para>
+     The <literal>libc</literal> provider uses the operating system's C
+     library. The collation and character classification behavior is
+     controlled by the settings <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal>, so they cannot be set independently.
+    </para>
+    <note>
+     <para>
+      The same locale name may have different behavior on different platforms
+      when using the libc provider.
+     </para>
+    </note>
+   </sect3>
+
   </sect2>
   <sect2 id="icu-locales">
    <title>ICU Locales</title>
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..5489ae7413 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -120,7 +120,7 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
      <listitem>
       <para>
        Specifies the provider to use for locale services associated with this
-       collation.  Possible values are
+       collation.  Possible values are <literal>none</literal>,
        <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
        (if the server was built with ICU support) or <literal>libc</literal>.
        <literal>libc</literal> is the default.  See <xref
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..60b9da0952 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -212,7 +212,7 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <listitem>
        <para>
         Specifies the provider to use for the default collation in this
-        database.  Possible values are
+        database.  Possible values are <literal>none</literal>,
         <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
         (if the server was built with ICU support) or <literal>libc</literal>.
         By default, the provider is the same as that of the <xref
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..326a371d34 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -168,7 +168,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry>
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>none</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         Specifies the locale provider for the database's default collation.
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..e604ab48b7 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -323,7 +323,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry id="app-initdb-option-locale-provider">
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>none</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         This option sets the locale provider for databases created in the new
diff --git a/src/backend/catalog/pg_collation.c b/src/backend/catalog/pg_collation.c
index fd022e6fc2..86b6ba2375 100644
--- a/src/backend/catalog/pg_collation.c
+++ b/src/backend/catalog/pg_collation.c
@@ -68,7 +68,12 @@ CollationCreate(const char *collname, Oid collnamespace,
 	Assert(collname);
 	Assert(collnamespace);
 	Assert(collowner);
-	Assert((collcollate && collctype) || colliculocale);
+	Assert((collprovider == COLLPROVIDER_NONE &&
+			!collcollate && !collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_LIBC &&
+			 collcollate &&  collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_ICU &&
+			!collcollate && !collctype &&  colliculocale));
 
 	/*
 	 * Make sure there is no existing collation of same name & encoding.
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c91fe66d9b..aeaf6c419e 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -215,7 +215,9 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 
 		if (collproviderstr)
 		{
-			if (pg_strcasecmp(collproviderstr, "icu") == 0)
+			if (pg_strcasecmp(collproviderstr, "none") == 0)
+				collprovider = COLLPROVIDER_NONE;
+			else if (pg_strcasecmp(collproviderstr, "icu") == 0)
 				collprovider = COLLPROVIDER_ICU;
 			else if (pg_strcasecmp(collproviderstr, "libc") == 0)
 				collprovider = COLLPROVIDER_LIBC;
@@ -228,6 +230,13 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		else
 			collprovider = COLLPROVIDER_LIBC;
 
+		if (collprovider == COLLPROVIDER_NONE
+			&& (localeEl || lccollateEl || lcctypeEl))
+		{
+			ereport(ERROR,
+					(errmsg("collation provider \"none\" does not support LOCALE, LC_COLLATE, or LC_CTYPE")));
+		}
+
 		if (localeEl)
 		{
 			if (collprovider == COLLPROVIDER_LIBC)
@@ -302,6 +311,16 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 					 errmsg("ICU rules cannot be specified unless locale provider is ICU")));
 
+		if (collprovider == COLLPROVIDER_NONE)
+		{
+			/*
+			 * Behavior may be different in different encodings, so set
+			 * collencoding to the current database encoding. No validation is
+			 * required, because the "none" provider is compatible with any
+			 * encoding.
+			 */
+			collencoding = GetDatabaseEncoding();
+		}
 		if (collprovider == COLLPROVIDER_ICU)
 		{
 #ifdef USE_ICU
@@ -331,7 +350,18 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 	}
 
 	if (!collversion)
-		collversion = get_collation_actual_version(collprovider, collprovider == COLLPROVIDER_ICU ? colliculocale : collcollate);
+	{
+		char *locale;
+
+		if (collprovider == COLLPROVIDER_ICU)
+			locale = colliculocale;
+		else if (collprovider == COLLPROVIDER_LIBC)
+			locale = collcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		collversion = get_collation_actual_version(collprovider, locale);
+	}
 
 	newoid = CollationCreate(collName,
 							 collNamespace,
@@ -406,6 +436,7 @@ AlterCollation(AlterCollationStmt *stmt)
 	Form_pg_collation collForm;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 	ObjectAddress address;
@@ -430,8 +461,20 @@ AlterCollation(AlterCollationStmt *stmt)
 	datum = SysCacheGetAttr(COLLOID, tup, Anum_pg_collation_collversion, &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = SysCacheGetAttrNotNull(COLLOID, tup, collForm->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
-	newversion = get_collation_actual_version(collForm->collprovider, TextDatumGetCString(datum));
+	if (collForm->collprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_colliculocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (collForm->collprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_collcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	newversion = get_collation_actual_version(collForm->collprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -494,11 +537,18 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_database) GETSTRUCT(dbtup))->datlocprovider;
 
-		datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_daticulocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_datcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
 
 		ReleaseSysCache(dbtup);
 	}
@@ -514,11 +564,19 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_collation) GETSTRUCT(colltp))->collprovider;
 		Assert(provider != COLLPROVIDER_DEFAULT);
-		datum = SysCacheGetAttrNotNull(COLLOID, colltp,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_colliculocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_collcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
 
 		ReleaseSysCache(colltp);
 	}
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 2e242eeff2..9e73f54803 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -909,7 +909,9 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	{
 		char	   *locproviderstr = defGetString(dlocprovider);
 
-		if (pg_strcasecmp(locproviderstr, "icu") == 0)
+		if (pg_strcasecmp(locproviderstr, "none") == 0)
+			dblocprovider = COLLPROVIDER_NONE;
+		else if (pg_strcasecmp(locproviderstr, "icu") == 0)
 			dblocprovider = COLLPROVIDER_ICU;
 		else if (pg_strcasecmp(locproviderstr, "libc") == 0)
 			dblocprovider = COLLPROVIDER_LIBC;
@@ -1177,9 +1179,17 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 */
 	if (src_collversion && !dcollversion)
 	{
-		char	   *actual_versionstr;
+		char	*actual_versionstr;
+		char	*locale;
 
-		actual_versionstr = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		actual_versionstr = get_collation_actual_version(dblocprovider, locale);
 		if (!actual_versionstr)
 			ereport(ERROR,
 					(errmsg("template database \"%s\" has a collation version, but no actual collation version could be determined",
@@ -1207,7 +1217,18 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 * collation version, which is normally only the case for template0.
 	 */
 	if (dbcollversion == NULL)
-		dbcollversion = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+	{
+		char *locale;
+
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		dbcollversion = get_collation_actual_version(dblocprovider, locale);
+	}
 
 	/* Resolve default tablespace for new database */
 	if (dtablespacename && dtablespacename->arg)
@@ -2403,6 +2424,7 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	ObjectAddress address;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 
@@ -2429,10 +2451,24 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	datum = heap_getattr(tuple, Anum_pg_database_datcollversion, RelationGetDescr(rel), &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = heap_getattr(tuple, datForm->datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
-	if (isnull)
-		elog(ERROR, "unexpected null in pg_database");
-	newversion = get_collation_actual_version(datForm->datlocprovider, TextDatumGetCString(datum));
+	if (datForm->datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_daticulocale, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datForm->datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	newversion = get_collation_actual_version(datForm->datlocprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -2617,6 +2653,7 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 	HeapTuple	tp;
 	char		datlocprovider;
 	Datum		datum;
+	char	   *locale;
 	char	   *version;
 
 	tp = SearchSysCache1(DATABASEOID, ObjectIdGetDatum(dbid));
@@ -2627,8 +2664,20 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 
 	datlocprovider = ((Form_pg_database) GETSTRUCT(tp))->datlocprovider;
 
-	datum = SysCacheGetAttrNotNull(DATABASEOID, tp, datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-	version = get_collation_actual_version(datlocprovider, TextDatumGetCString(datum));
+	if (datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_daticulocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_datcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_NONE */
+
+	version = get_collation_actual_version(datlocprovider, locale);
 
 	ReleaseSysCache(tp);
 
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index eea1d1ae0f..95eb5cf464 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1228,7 +1228,12 @@ lookup_collation_cache(Oid collation, bool set_flags)
 			elog(ERROR, "cache lookup failed for collation %u", collation);
 		collform = (Form_pg_collation) GETSTRUCT(tp);
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_NONE)
+		{
+			cache_entry->collate_is_c = true;
+			cache_entry->ctype_is_c = true;
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 			Datum		datum;
 			const char *collcollate;
@@ -1281,6 +1286,9 @@ lc_collate_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_NONE)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1334,6 +1342,9 @@ lc_ctype_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_NONE)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1487,8 +1498,10 @@ pg_newlocale_from_collation(Oid collid)
 	{
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return &default_locale;
-		else
+		else if (default_locale.provider == COLLPROVIDER_LIBC)
 			return (pg_locale_t) 0;
+		else
+			elog(ERROR, "cannot open collation with provider \"none\"");
 	}
 
 	cache_entry = lookup_collation_cache(collid, false);
@@ -1513,7 +1526,11 @@ pg_newlocale_from_collation(Oid collid)
 		result.provider = collform->collprovider;
 		result.deterministic = collform->collisdeterministic;
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_NONE)
+		{
+			elog(ERROR, "cannot open collation with provider \"none\"");
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 #ifdef HAVE_LOCALE_T
 			const char *collcollate;
@@ -1599,6 +1616,7 @@ pg_newlocale_from_collation(Oid collid)
 
 			collversionstr = TextDatumGetCString(datum);
 
+			Assert(collform->collprovider != COLLPROVIDER_NONE);
 			datum = SysCacheGetAttrNotNull(COLLOID, tp, collform->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
 			actual_versionstr = get_collation_actual_version(collform->collprovider,
@@ -1650,6 +1668,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (collprovider == COLLPROVIDER_NONE)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
diff --git a/src/backend/utils/init/postinit.c b/src/backend/utils/init/postinit.c
index 53420f4974..8053642fd3 100644
--- a/src/backend/utils/init/postinit.c
+++ b/src/backend/utils/init/postinit.c
@@ -461,10 +461,18 @@ CheckMyDatabase(const char *name, bool am_superuser, bool override_allow_connect
 	{
 		char	   *actual_versionstr;
 		char	   *collversionstr;
+		char	   *locale;
 
 		collversionstr = TextDatumGetCString(datum);
 
-		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, dbform->datlocprovider == COLLPROVIDER_ICU ? iculocale : collate);
+		if (dbform->datlocprovider == COLLPROVIDER_ICU)
+			locale = iculocale;
+		else if (dbform->datlocprovider == COLLPROVIDER_LIBC)
+			locale = collate;
+		else
+			locale = NULL; /* COLLPROVIDER_NONE */
+
+		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, locale);
 		if (!actual_versionstr)
 			/* should not happen */
 			elog(WARNING,
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 30b576932f..4a6cad3cb9 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2408,6 +2408,22 @@ setlocales(void)
 
 	/* set empty lc_* values to locale config if set */
 
+	if (locale_provider == COLLPROVIDER_NONE)
+	{
+		if (!lc_ctype)
+			lc_ctype = "C";
+		if (!lc_collate)
+			lc_collate = "C";
+		if (!lc_numeric)
+			lc_numeric = "C";
+		if (!lc_time)
+			lc_time = "C";
+		if (!lc_monetary)
+			lc_monetary = "C";
+		if (!lc_messages)
+			lc_messages = "C";
+	}
+
 	if (locale)
 	{
 		if (!lc_ctype)
@@ -2502,7 +2518,7 @@ usage(const char *progname)
 			 "                            set default locale in the respective category for\n"
 			 "                            new databases (default taken from environment)\n"));
 	printf(_("      --no-locale           equivalent to --locale=C\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={none|libc|icu}\n"
 			 "                            set default locale provider for new databases\n"));
 	printf(_("      --pwfile=FILE         read password for the new superuser from file\n"));
 	printf(_("  -T, --text-search-config=CFG\n"
@@ -2652,7 +2668,15 @@ setup_locale_encoding(void)
 {
 	setlocales();
 
-	if (locale_provider == COLLPROVIDER_LIBC &&
+	if (locale_provider == COLLPROVIDER_NONE &&
+		strcmp(lc_ctype, "C") == 0 &&
+		strcmp(lc_collate, "C") == 0 &&
+		strcmp(lc_time, "C") == 0 &&
+		strcmp(lc_numeric, "C") == 0 &&
+		strcmp(lc_monetary, "C") == 0 &&
+		strcmp(lc_messages, "C") == 0)
+		printf(_("The database cluster will be initialized with no locale.\n"));
+	else if (locale_provider == COLLPROVIDER_LIBC &&
 		strcmp(lc_ctype, lc_collate) == 0 &&
 		strcmp(lc_ctype, lc_time) == 0 &&
 		strcmp(lc_ctype, lc_numeric) == 0 &&
@@ -3326,7 +3350,9 @@ main(int argc, char *argv[])
 										 "-c debug_discard_caches=1");
 				break;
 			case 15:
-				if (strcmp(optarg, "icu") == 0)
+				if (strcmp(optarg, "none") == 0)
+					locale_provider = COLLPROVIDER_NONE;
+				else if (strcmp(optarg, "icu") == 0)
 					locale_provider = COLLPROVIDER_ICU;
 				else if (strcmp(optarg, "libc") == 0)
 					locale_provider = COLLPROVIDER_LIBC;
@@ -3365,6 +3391,7 @@ main(int argc, char *argv[])
 		exit(1);
 	}
 
+
 	if (icu_locale && locale_provider != COLLPROVIDER_ICU)
 		pg_fatal("%s cannot be specified unless locale provider \"%s\" is chosen",
 				 "--icu-locale", "icu");
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index 17a444d80c..fe6d224e5b 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -154,6 +154,35 @@ else
 		'locale provider ICU fails since no ICU support');
 }
 
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', "$tempdir/data6" ],
+	'locale provider none');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--locale=C',
+	  "$tempdir/data7" ],
+	'locale provider none with --locale');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--lc-collate=C',
+	  "$tempdir/data8" ],
+	'locale provider none with --lc-collate');
+
+command_ok(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--lc-ctype=C',
+	  "$tempdir/data9" ],
+	'locale provider none with --lc-ctype');
+
+command_fails(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--icu-locale=en',
+	  "$tempdir/dataX" ],
+	'fails for locale provider none with ICU locale');
+
+command_fails(
+	[ 'initdb', '--no-sync', '--locale-provider=none', '--icu-rules=""',
+	  "$tempdir/dataX" ],
+	'fails for locale provider none with ICU rules');
+
 command_fails(
 	[ 'initdb', '--no-sync', '--locale-provider=xyz', "$tempdir/dataX" ],
 	'fails for invalid locale provider');
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index f9cbeb65ab..ddc8a5f71f 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -3070,7 +3070,9 @@ dumpDatabase(Archive *fout)
 	}
 
 	appendPQExpBufferStr(creaQry, " LOCALE_PROVIDER = ");
-	if (datlocprovider[0] == 'c')
+	if (datlocprovider[0] == 'n')
+		appendPQExpBufferStr(creaQry, "none");
+	else if (datlocprovider[0] == 'c')
 		appendPQExpBufferStr(creaQry, "libc");
 	else if (datlocprovider[0] == 'i')
 		appendPQExpBufferStr(creaQry, "icu");
@@ -13429,7 +13431,9 @@ dumpCollation(Archive *fout, const CollInfo *collinfo)
 					  fmtQualifiedDumpable(collinfo));
 
 	appendPQExpBufferStr(q, "provider = ");
-	if (collprovider[0] == 'c')
+	if (collprovider[0] == 'n')
+		appendPQExpBufferStr(q, "none");
+	else if (collprovider[0] == 'c')
 		appendPQExpBufferStr(q, "libc");
 	else if (collprovider[0] == 'i')
 		appendPQExpBufferStr(q, "icu");
diff --git a/src/bin/pg_upgrade/t/002_pg_upgrade.pl b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
index 4a7895a756..6d58f6103e 100644
--- a/src/bin/pg_upgrade/t/002_pg_upgrade.pl
+++ b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
@@ -114,12 +114,20 @@ my $original_locale = "C";
 my $original_iculocale = "";
 my $provider_field = "'c' AS datlocprovider";
 my $iculocale_field = "NULL AS daticulocale";
-if ($oldnode->pg_version >= 15 && $ENV{with_icu} eq 'yes')
+if ($oldnode->pg_version >= 15)
 {
 	$provider_field = "datlocprovider";
 	$iculocale_field = "daticulocale";
-	$original_provider = "i";
-	$original_iculocale = "fr-CA";
+
+	if ($ENV{with_icu} eq 'yes')
+	{
+		$original_provider = "i";
+		$original_iculocale = "fr-CA";
+	}
+	else
+	{
+		$original_provider = "n";
+	}
 }
 
 my @initdb_params = @custom_opts;
@@ -131,6 +139,10 @@ if ($original_provider eq "i")
 	push @initdb_params, ('--locale-provider', 'icu');
 	push @initdb_params, ('--icu-locale', 'fr-CA');
 }
+elsif ($original_provider eq "n")
+{
+	push @initdb_params, ('--locale-provider', 'none');
+}
 
 $node_params{extra} = \@initdb_params;
 $oldnode->init(%node_params);
diff --git a/src/bin/psql/describe.c b/src/bin/psql/describe.c
index ab4279ed58..c842a62ae9 100644
--- a/src/bin/psql/describe.c
+++ b/src/bin/psql/describe.c
@@ -932,7 +932,7 @@ listAllDbs(const char *pattern, bool verbose)
 					  gettext_noop("Encoding"));
 	if (pset.sversion >= 150000)
 		appendPQExpBuffer(&buf,
-						  "  CASE d.datlocprovider WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
+						  "  CASE d.datlocprovider WHEN 'n' THEN 'none' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
 						  gettext_noop("Locale Provider"));
 	else
 		appendPQExpBuffer(&buf,
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..79367d933b 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -299,7 +299,7 @@ help(const char *progname)
 	printf(_("      --lc-ctype=LOCALE        LC_CTYPE setting for the database\n"));
 	printf(_("      --icu-locale=LOCALE      ICU locale setting for the database\n"));
 	printf(_("      --icu-rules=RULES        ICU rules setting for the database\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={none|libc|icu}\n"
 			 "                               locale provider for the database's default collation\n"));
 	printf(_("  -O, --owner=OWNER            database user to own the new database\n"));
 	printf(_("  -S, --strategy=STRATEGY      database creation strategy wal_log or file_copy\n"));
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index af3b1492e3..5aa658b671 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -83,6 +83,35 @@ else
 		'create database with ICU fails since no ICU support');
 }
 
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', 'testnone1' ],
+	'create database with provider "none"');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--locale=C',
+	  'testnone2' ],
+	'create database with provider "none" and locale "C"');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--lc-collate=C',
+	  'testnone3' ],
+	'create database with provider "none" and LC_COLLATE=C');
+
+$node->command_ok(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--lc-ctype=C',
+	  'testnone4' ],
+	'create database with provider "none" and LC_CTYPE=C');
+
+$node->command_fails(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--icu-locale=en',
+	  'testnone5' ],
+	'create database with provider "none" and ICU_LOCALE="en"');
+
+$node->command_fails(
+	[ 'createdb', '-T', 'template0', '--locale-provider=none', '--icu-rules=""',
+	  'testnone6' ],
+	'create database with provider "none" and ICU_RULES=""');
+
 $node->command_fails([ 'createdb', 'foobar1' ],
 	'fails if database already exists');
 
diff --git a/src/include/catalog/pg_collation.dat b/src/include/catalog/pg_collation.dat
index b6a69d1d42..40d62416ea 100644
--- a/src/include/catalog/pg_collation.dat
+++ b/src/include/catalog/pg_collation.dat
@@ -24,8 +24,7 @@
   collname => 'POSIX', collprovider => 'c', collencoding => '-1',
   collcollate => 'POSIX', collctype => 'POSIX' },
 { oid => '962', descr => 'sorts by Unicode code point',
-  collname => 'ucs_basic', collprovider => 'c', collencoding => '6',
-  collcollate => 'C', collctype => 'C' },
+  collname => 'ucs_basic', collprovider => 'n', collencoding => '6' },
 { oid => '963',
   descr => 'sorts using the Unicode Collation Algorithm with default settings',
   collname => 'unicode', collprovider => 'i', collencoding => '-1',
diff --git a/src/include/catalog/pg_collation.h b/src/include/catalog/pg_collation.h
index bfa3568451..29be3f8d94 100644
--- a/src/include/catalog/pg_collation.h
+++ b/src/include/catalog/pg_collation.h
@@ -64,6 +64,7 @@ DECLARE_UNIQUE_INDEX_PKEY(pg_collation_oid_index, 3085, CollationOidIndexId, on
 
 #ifdef EXPOSE_TO_CLIENT_CODE
 
+#define COLLPROVIDER_NONE		'n'
 #define COLLPROVIDER_DEFAULT	'd'
 #define COLLPROVIDER_ICU		'i'
 #define COLLPROVIDER_LIBC		'c'
@@ -73,6 +74,8 @@ collprovider_name(char c)
 {
 	switch (c)
 	{
+		case COLLPROVIDER_NONE:
+			return "none";
 		case COLLPROVIDER_ICU:
 			return "icu";
 		case COLLPROVIDER_LIBC:
diff --git a/src/test/regress/expected/collate.out b/src/test/regress/expected/collate.out
index 0649564485..b7603c9f6c 100644
--- a/src/test/regress/expected/collate.out
+++ b/src/test/regress/expected/collate.out
@@ -650,6 +650,13 @@ EXPLAIN (COSTS OFF)
 (3 rows)
 
 -- CREATE/DROP COLLATION
+CREATE COLLATION none ( PROVIDER = none );
+CREATE COLLATION none2 ( PROVIDER = none, LOCALE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
+CREATE COLLATION none2 ( PROVIDER = none, LC_CTYPE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
+CREATE COLLATION none2 ( PROVIDER = none, LC_COLLATE="POSIX" ); -- fails
+ERROR:  collation provider "none" does not support LOCALE, LC_COLLATE, or LC_CTYPE
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
@@ -754,7 +761,7 @@ DETAIL:  FROM cannot be specified together with any other options.
 -- must get rid of them.
 --
 DROP SCHEMA collate_tests CASCADE;
-NOTICE:  drop cascades to 19 other objects
+NOTICE:  drop cascades to 20 other objects
 DETAIL:  drop cascades to table collate_test1
 drop cascades to table collate_test_like
 drop cascades to table collate_test2
@@ -771,6 +778,7 @@ drop cascades to function dup(anyelement)
 drop cascades to table collate_test20
 drop cascades to table collate_test21
 drop cascades to table collate_test22
+drop cascades to collation "none"
 drop cascades to collation mycoll2
 drop cascades to table collate_test23
 drop cascades to view collate_on_int
diff --git a/src/test/regress/sql/collate.sql b/src/test/regress/sql/collate.sql
index c3d40fc195..e2dceb8dff 100644
--- a/src/test/regress/sql/collate.sql
+++ b/src/test/regress/sql/collate.sql
@@ -244,6 +244,12 @@ EXPLAIN (COSTS OFF)
 
 -- CREATE/DROP COLLATION
 
+CREATE COLLATION none ( PROVIDER = none );
+
+CREATE COLLATION none2 ( PROVIDER = none, LOCALE="POSIX" ); -- fails
+CREATE COLLATION none2 ( PROVIDER = none, LC_CTYPE="POSIX" ); -- fails
+CREATE COLLATION none2 ( PROVIDER = none, LC_COLLATE="POSIX" ); -- fails
+
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
-- 
2.34.1

From 274a887f8970647b2c932ee55c4783095719985d Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Fri, 28 Apr 2023 12:22:41 -0700
Subject: [PATCH v6 5/5] ICU: fix up old libc-style locale strings.

Before transforming a locale string into a language tag, fix up old
libc-style locale strings such as 'fr_FR@euro'. Older ICU versions did
this automatically, but ICU version 64 removed that support.

Discussion: https://postgr.es/m/654a49f7ff7461bcf47be4181430678d45f93858.camel%40j-davis.com
---
 src/backend/utils/adt/pg_locale.c             | 57 ++++++++++++++++-
 src/bin/initdb/initdb.c                       | 61 ++++++++++++++++++-
 .../regress/expected/collate.icu.utf8.out     | 11 ++++
 src/test/regress/sql/collate.icu.utf8.sql     |  7 +++
 4 files changed, 134 insertions(+), 2 deletions(-)

diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 95eb5cf464..2ee81e9804 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -2787,6 +2787,58 @@ icu_set_collation_attributes(UCollator *collator, const char *loc,
 
 	pfree(lower_str);
 }
+
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < lengthof(icu_variant_map); i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = palloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pstrdup(loc_str);
+}
+
 #endif
 
 /*
@@ -2803,6 +2855,7 @@ icu_language_tag(const char *loc_str, int elevel)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
@@ -2833,7 +2886,7 @@ icu_language_tag(const char *loc_str, int elevel)
 	while (true)
 	{
 		status = U_ZERO_ERROR;
-		uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+		uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/* try again if the buffer is not large enough */
 		if ((status == U_BUFFER_OVERFLOW_ERROR ||
@@ -2848,6 +2901,8 @@ icu_language_tag(const char *loc_str, int elevel)
 		break;
 	}
 
+	pfree(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pfree(langtag);
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index f0827154cd..1304a235ce 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2240,6 +2240,61 @@ check_icu_locale_encoding(int user_enc)
 	return true;
 }
 
+#ifdef USE_ICU
+
+static const char *icu_variant_map[][2] = {
+	{ "@EURO",   "@currency=EUR" },
+	{ "@PINYIN", "@collation=pinyin" },
+	{ "@STROKE", "@collation=stroke" },
+};
+
+/*
+ * ICU version 64 removed the ability to transform locale strings of the form
+ * '...@VARIANT' into proper language tags. Perform the transformation from
+ * within Postgres so that ICU supports any libc locale name consistently,
+ * regardless of the ICU version.
+ */
+static char *
+icu_fix_variants(const char *loc_str)
+{
+	const char *old_variant = strrchr(loc_str, '@');
+
+	/*
+	 * Extract a variant of the form '...@VARIANT', and replace with
+	 * the appropriate '...@keyword=value' if found in the map.
+	 */
+	if (old_variant)
+	{
+		size_t prefix_len = old_variant - loc_str; /* bytes before the '@' */
+
+		for (int i = 0; i < lengthof(icu_variant_map); i++)
+		{
+			const char *map_variant = icu_variant_map[i][0];
+			const char *map_replacement = icu_variant_map[i][1];
+
+			if (pg_strcasecmp(old_variant, map_variant) == 0)
+			{
+				size_t	 replacement_len = strlen(map_replacement);
+				size_t	 result_len;
+				char	*result;
+
+				result_len = prefix_len + replacement_len + 1;
+				result = pg_malloc(result_len);
+
+				memcpy(result, loc_str, prefix_len);
+				memcpy(result + prefix_len, map_replacement, replacement_len);
+				result[prefix_len + replacement_len] = '\0';
+
+				return result;
+			}
+		}
+	}
+
+	return pg_strdup(loc_str);
+}
+
+#endif
+
 /*
  * Convert to canonical BCP47 language tag. Must be consistent with
  * icu_language_tag().
@@ -2249,6 +2304,7 @@ icu_language_tag(const char *loc_str)
 {
 #ifdef USE_ICU
 	UErrorCode	 status;
+	char		*fixed_loc_str = icu_fix_variants(loc_str);
 	char		 lang[ULOC_LANG_CAPACITY];
 	char		*langtag;
 	size_t		 buflen = 32;	/* arbitrary starting buffer size */
@@ -2277,7 +2333,8 @@ icu_language_tag(const char *loc_str)
 	while (true)
 	{
 		status = U_ZERO_ERROR;
-		uloc_toLanguageTag(loc_str, langtag, buflen, strict, &status);
+
+		uloc_toLanguageTag(fixed_loc_str, langtag, buflen, strict, &status);
 
 		/* try again if the buffer is not large enough */
 		if (status == U_BUFFER_OVERFLOW_ERROR ||
@@ -2291,6 +2348,8 @@ icu_language_tag(const char *loc_str)
 		break;
 	}
 
+	pg_free(fixed_loc_str);
+
 	if (U_FAILURE(status))
 	{
 		pg_free(langtag);
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index ea96e27f45..692e8cdf18 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1071,12 +1071,23 @@ ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
+ERROR:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
+WARNING:  could not convert locale name "@ASDF" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-cu-eur" for ICU locale "@EURO"
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-pinyin" for ICU locale "@pinyin"
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
+NOTICE:  using standard form "und-u-co-stroke" for ICU locale "@stroke"
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index ee607ca3a5..0b90e2a5b9 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -395,9 +395,16 @@ CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, nee
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); -- fails
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@ASDF'); DROP COLLATION testx;
+
+-- test special variants
+CREATE COLLATION testx (provider = icu, locale = '@EURO'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@pinyin'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = '@stroke'); DROP COLLATION testx;
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
-- 
2.34.1

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 9db14649aa..96a23bf530 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -1140,6 +1140,14 @@ SELECT 'w;x*y-z' = 'wxyz' COLLATE num_ignore_punct; -- true
      <table id="icu-collation-levels">
       <title>ICU Collation Levels</title>
       <tgroup cols="8">
+       <colspec colname="col1" colwidth="1*"/>
+       <colspec colname="col2" colwidth="1.25*"/>
+       <colspec colname="col3" colwidth="1*"/>
+       <colspec colname="col4" colwidth="1*"/>
+       <colspec colname="col5" colwidth="1*"/>
+       <colspec colname="col6" colwidth="1*"/>
+       <colspec colname="col7" colwidth="1*"/>
+       <colspec colname="col8" colwidth="1*"/>
        <thead>
         <row>
          <entry>Level</entry>

�PNG


IHDR
�
�"3-�@iCCPICC ProfileH��WXS��[��@h�K	�	"5��Z�ATB �A��.*�v��
]Q�bA�����bAEYv�M
���|o�o����3�9sfn�NpD�<T�|a�8.$�>6%�Nz
@�{�����11������u�H�+R�������p
@b ��p�!>^��
 Jy�)�")�h�a�/��,9���9�+�I�cA���
�#�@���E�,������'�F��7?�t�m��b�>#����ifkr8Y�X>YQ
��8���t����'�a�J�84N:g�������X�>aFT4���d���lIh��5��`��:���	���`a^T�����!�;�*(d'@��B~AP��f�xR����)f1�Y�X�W���$7�����g+�1����d�)[	�� V��� 7>\a3�8�5d#��I���8�/	��cE���8�}Y~��|���v��/�N��k�rd���`��Bf���`l��\x�� ���g|ab�B���0 N>���b��?/D��A�ZP��'�
)��3E�1	�8��NX�<|�,�@k�r�������{�
�A�34"Y�#��xP���
���z���_�Y��
d�z�d#r���A8����(���$�2�x�������*������aB&B�H�<���,�A�@b(1�h����7
���:��sh��	O]���k�n�����OQF�n���E���������@u�����
w�~����Y�"niV�?i�m?�����DF��d���#U�T��U���1?�X3������?����`��%�;��c'�s�Q����	���I���z,�]C��d��B�?�
��4�N�N�N_�}����w4`MM���L�E���B��H����������&V��@t:�s����epp��w.��}��?���a�O�2gs%�"9�K/��P�O�>0�����o��@�	 L��g�}.S�0��r����&��{�~������.�k��==����gAH�!��	b��#��E��$IA��,D�H��<�Y��G� 5�>�0r9�t!��H/���b�
���V�(��2�p4
�f���bt>�]�V����$z
��v�/�
`��f�9`��Ec�X&&�faeXV��a�p��`�X�'�4��;��'�\|2>_���w�
x~����T�!���E`��S��
�v�!�i�,���D�5�>�)��t�b�b=�����8@"��I�$R4�C*$����v�ZH�I=�J�J&J�J�J�JB��
�]J��.+=U�LV'[����dyy)y��|��C�L��XS|(	��\�ZJ�4�.��������r��@y��Z���g�(T�T�Sa���HT���P9�rK�
�J���SS���%��)�}�U���*[��:[�R�A���K5���Sm�Z�Z��
��j}�du+u�:G}�z��a��4�����5vi��x�I�����i����yJ�
���X4.mm�4�G��e�����*�������������=U�R��v��c�����Y��_���']#]�._w�n��e��z#����zez�z��>�����s���7��3�
�b
�l48m�7Bk������G�6D
���n5�0026
1�3:e�g�c�o�c����q�	���D`�����9]����������������-��������J�����S������[��-L,"-fX�Z��$[2,�-�X�[����J�Z`�h��Z��m]l]k}��j�g3�����-��a�k����j�f�mWiw��w��o��I�9R8�z�
�C�C��G���F���,F��Z>�}�7'7�<�mNwFk�]2�y�kg;g�s��U�K��l�&�W���|���7�hn�n�Z���{�����{=,<�=�<n0�1���������G=?z�{z������;�{���1�c�c��y�c�������K�M�����g�����{�o�������i��a�f�p

x��b�d��C�;�4����6�
�
�q�r"��<����e����<�f��������a!�h�D#�"WF����F5F�hv���{1�1�c��ccb+c������O���+�]B@���;�6����$����������+����;s���AJS*)5)u{����q��������]o=~��s&�M86Qm"g��tBzr���/�hN5g ��Q���eq�p_��y�x�|�
��L�����|�Vf�f�eWd�	X���W9�9�r��F����K���W�O�?,��
�&O�:�Kd/*uO���zr�8\��)_�T��;$6�_$�|�*�>LI�r`��T���iv�M{Z\��t|:wz��sg<����e2+cV�l���g��	��s.en���K�JV����<�y���9���Km�j�������;�,Z��[��|�SyE���������_�d.�\��t�2�2�������\���x����+V�W��z�z��s���P�H�t��X���b��u_�g��VPY_eX��������7�m2�T���f���[B�4T[UWl%n-��d[������l7�^�������q;�j<jjv�ZZ��Jj{w����'pOS�C��z����`�d��}������z�q������C�Ce
H�������������a�[���q<������c�����|���e���D�����Z'��95������������	>s����r����s^��g�o��~�������n��t�l��q�������1]�/�]>y%�������E]���x�����7y7������v���w��%�-��~������?l���v�>� �A����wq�x\��K��'�'OM��<s~v�7����q�{^�^|�+�S����6/���WG����W�W�����������������}~_�A���������?=�<������_����;�?8(��9�_
V43��;��@��3�8��OV��U����3���P��c����
�n��/���@�O��������\)-Dx��5#?��"?s���-������
Md|�8���eXIfMM*

>

F
(
�i
N�
�
��x�

��

�ASCIIScreenshot]~Z	pHYs%%
IR$�
�iTXtXML:com.adobe.xmp<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="XMP Core 6.0.0">
   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
      <rdf:Description rdf:about=""
            xmlns:exif="http://ns.adobe.com/exif/1.0/">
         <exif:PixelYDimension>494</exif:PixelYDimension>
         <exif:PixelXDimension>470</exif:PixelXDimension>
         <exif:UserComment>Screenshot</exif:UserComment>
      </rdf:Description>
   </rdf:RDF>
</x:xmpmeta>
ax�iDOT�(��o>� �@IDATx
���TE��{u
k�������YsF��0`Q	�U��(�
D�����"�y�OE1
fX0�b����������o=�xf���3���;����;���������S��]��B���%�����s���w��sN���`H�����?����7�I�b�&
�V� 
V��
K�����,�3`���I���~	&��0`-,#�a�j���0`���I��X��r��o`�X��`(,
��2���X(
V�)�
K�����,�
��J���~
&��0`-,#�a�j���0`���I��X��r��o`�X��`(,
��2���X(
V�)�
K�����,�
��J���~
&��0`-,#�a�j���0`���I��X��r��o`�X��`(,
��2���X(
V�)�
K�����,�
��J���~
&��0`-,#�a�j���0`���I��X��r��o`�X��`(,
��2���X(
V�)�
K�����,�
��J ���7��s�=�d��N;��w��`����=�����p��?�����(�����|P�������*��2����g���{�9w�}����[�Y��u6K�JT�
k�Y�ld32	�������^zi�$H?~�_z������������������O�����{�������f�M�����L��U�V������7�p��}��5�}��I�k.=_��Xc
���k���%�l����"�~���p2�,
W������@.`
�
O���#�I�'�|�?���2�
�|���J�Y�f������9s����_H��z��7��s�=�A�r����*L0`��Ak�&����+��|���[7���:*g�_z�%��7�|��x�G4��_�.��%���sg�6$������
kn�Xj�~�m�����%�k�5������t�n���!Y�Y�~���?��O��,�����?�����O?u���?4m�Eu{���^��k����[oub
v�����S~�
7tb
v��M���?�f����r��{��G���m���3�}������_v����[���m����2�,X��=��S��t�<�������h�n�Eq?����E����K7}�t���k��6�,�!���g��u��Q:�+?��������6"1�Gy����&��^{m����^��#�6�x��oR��<��S�������o����m�l�ls��{��7�����b�-�e�A������ZK�(�������VZ)��j�m��j��y.�HF&��k�A,�4�u�]������w���v�j�2�j�7�x��r�)���2�v��X�U@����������0`��������K2��������g�}�I'��WYe/����5��/��=�P��Kxy��|����o�'�p�_~����+������)S����3gzd@{H[a��P�q���A��+����K��v�m�����(O��*/)�-��<y�����[����3&�s���4���eC���S^|�v���8���E]������L����1�ai�����k��C���I
#/NZ������Z�#�<�7Sp����kIc
���v�i�a]�c�X!���>+i���m��
Xqy����I�v|��'^<g5m�W�Z�3f(�D�Vg��_���;���G���/���x��~���~����o����:uj�]�t��{�f}p�
�E�����|��}���g����V�t���
����k�����l`��[o����j+��0�b�-"^_|���408AAi���2���_�+�N20��A4�(�X1�GXw`��e�]�O�<Y�9�6j���l5_�V��5_�
X�O�]S`eRG��3t�P�0i`s�����Nb�V���Q���4<������I������%�h��}�]<�^3���	$�TM���[B�������@c���rq����������
X�w�C8�m<N���������:��)S�h�0�	�
2���Ch��><�z��7j��@��0����
k�e\5�
�\�K�n�i�+��f'L��������~�K.���cB�&]�a�d
V�7�d�x���|���\�i�N%���/�����&��W_]��M�I�L�
X�����{�U>�����)%���c��
�e�y��k=���)�~��������+������ab�2`��Q,2g���g5T���@���t��4�4�*�H:�t���o��
G�����U-�1�:d��QX
����N����/���6lX��b���-:��#frQ��b����O�����#^�^w��I�!��U�`L���~*��k\|����.��92N����T.`�/��6m�h�U
`=��c���&���Bu`�q)��5V���f�l:��C�Yx�0`������X�u����XqPar��f��X[n�e��������Q6�+�I�6�l��Dc�X���Q�������B�
V�`�@>��`�s���;�X��i��T���Y���&`��#F��i���t�M���v<��������� �\�*�Vj]f
�����Xk|��v/���L����Gl�D9�����^�&P����/��2�;N�0!b&�u�Y'J�u�i~�E�`���e���
t88q$cp�!��l��������^�r� '^�Z-^���6�`%��,�<wq����_�U�
���%�;x�N�4I��W�;���&`�g�Y/��*M���6�FM�^�z��3�d�IZp�����^01��q�e����Z�r9J�<��mk��4Z-�V���}E��X�|�I]d�9�!����ipj�4(#�+�F�7)&L��8�e,�u����-'�4��h
yw�m7��"�.Dm��*)Z4�?�����r���C#��rh�����Z�������.�S�����J��ES��A�#���-�z�q�y9PA���������( �~�\��C(yq��j����8_x��dJ}���������<���4�t�Mmp����8��G��}�%	m��`�@�`��3/��P���]�d�Z�#�<���������|'/+�������(��Fl��)i~�r����!��ZR���qZ�h�N�E���W�G��*���i��}�N�	���G����K�F��Zi��b^%����*�n�
k@�T_J`��>Y;L���
k�%Z��Xks\���
k�"�u(
�:�Ft���B��"��8���RK�����M~��9�E����h�Y�:��
kz#�l����b�ZU�S�%`�Zj��&?����{e�Z���@J����]6`m��j��
k-�����0`-�Dk��
km�k��2`-ZdV�%`�Z����.�6Bh�X���G��Tj	��Z������6���^�V9��IDw�}�99��h�����s����H<w�	'�d�n�J5&������������
kjQ�v�|�*�;9KV�U�o� ����:�4���w'�7��tz\d)��dY�X��W��6`�����w>`E��H1Ni�*U��D�Q�w�wtLP�F]�������/.[�%^��j���,�|��jL��a�J����L,��X��.S�
k]
$G��t��8�^b��x��$���$D��2e�~7��q(�XK-���g�Z��Zt��
�ZrO?��[{����y�xS��{�uz�m��F)EB�E�s]Hx:'����{�=�x�N�{:	-��5}�t5uJ��(��Y�TkxZ�n����������5�LA�:��sO��ZkE������-��I��(Ox���mS���<��k��]y��n���ZFB�i����#q`��ssj�I5}����&���>��������VZI�Ix;��C��g��Q�F9	m��kR�\cB;
�5*�E��L#?	
���6m��9s�8��H��(���uS�NUyl����h9q�x��������T
�V�HTx;�':��R��������0����<�H��q��iSn�S&G?q���y�����5�6�I���?��c/NL����\�8�!Vk��]���3�������_I���X���~�_u�U�m={���M������;j�h�����X�)����H����5����=qN)/��(pz��m����k���|�s�=��A�i\Y�����z	��<	�������L����?�|�{�������ZG�g�	�H����y���g�1��m����_�~�~�����h%�/���I��q�7����#��o~/����x��2��t^���l����&�������U4	
t}�y�E���$�����7��|����Es��TL����K�T���z�yxK�R��o_M;��3����h4�&���'�|��w��i�
F�)����GM��<�HMM���	�����r�N���z���tq���#�@8S^�hM�)C���.�y�����k�h��h����(iL��'�w��L#?�3��@k����5 :i��Yg��E#��g|H�����I�' =�����-�hZK�1`m����
X�c����\�*=���u���6�F(��^L��M�i4B�\��4l�����t@#�����i�[�n�zhH��e��n���(
^��5���o��f��hAq���S��/���d+��
6,�5����?#��s��Q���c�|0�f�c���Q�o�Q��[`1)G��.���|i���Yh<���u�]������j���ib:����s��c-�t�M��	�O���Q��nX[J��U�
ku�W�Z�X�u��cV���_���
ZJ.`��GM��gb&����%�m��������4L�q�����_z��j����
�Y�:�i����hT����;2�P����p#M�_}����%�\���h��|�A�w�����h^x��(-�"��`�OSO.����xZ�!G�^x��Q5�0�un���Q���+����r��?���*�X+a*�
��?F���b�3n�*��i��N���C��I���
�������c���G�
���S'}F}P�����uE��h��]F��5���4���|�Y��r
+r]t�E�+}f�xa�d2`�������V�X�hK�
VL������3!��"����9���3��6���9R�q���C9D�/�8.���o�x�z�Z�\C�V��U$~�Rc-4��������:�!d��o6�y��^����������E^��V������+�'x��q��3&�������
V�\�>iM��
{L�����={v��*�����[����L�v�g��w��c~&m�v����i|�����S�/n
����l!J�����]h<��_�����L�X�����	�7L��@��0
�����fia>`
&�I�&i[��Y3�[+�����a�'�@��1�~
�@l�`�
&Z�e����%
k��1h4��R6��'���:��PT����
^��a�hy�����W�^�5g��`����W+��G&4i����X^}���w��'�0O�r�S�?�"{Lx���l^i�3���RD�K �{I�/�@{�����=]��zO^��Yg�=��-	�J �J��o�
k��Q��0���������V'<�Q
-&���_^?lUA�^}����bN\g�u���}����8�H����.^N�hqK,����j�J����H+&/�B,�
XY�d���6�n���O?)�����/�F�.�:� �3�
P<o���� ���y�M7){�gk	��C0�"��+[F(�^�#Fx^P��K��WbKI���$�����z�y��B�����_���_�`�^��R
�l�z��W������>������gx	�Ys��S��
X�%���k'/���s��#f<w�9���8N��=
I&����
2(a
a'�g�hxN�����V'u�Zdk���$�I=�W0g��DN@�N"�.��2N@Q�����#�[��$�s1��,��������/M�E�u�z*��gNS�����K�&���������t�/XKWk�9��U4��34&�<0`�#{I��5E}_T+���N�pD�!���I��0`-�tk��
k��e�zRm���R�9
�e-�m���N�:t����P�&	�
�b0`�	�.%`��(��zPm�*NEN�s3I�a���������nL���
k#�V�EX�p�su���5W,�$Pn	��[������6����0`m��AH����]4`-�k��
k-�����0`-��k��
km�c�{a��d�:��
kr	�h�Z!��ZE�C�%`�Zn	����&�����"�`��Tr��{����F�����nj�/M��
kS%X�
X�c��������������~���i��9	��$|]Q�+)s-���r5`m����
k}�s�^�Q�����6�l��V:^K})z cXc���D	�&�����~��z�-
P�+����4e�
X�"��)k�Z?c�������1O����{z|�&�l���'�|�$�����n��vr�+#uH�M'a��Fm�Qb$]��rs��q����c�q���y��sQ�GF���C�>}�{�����pn�����7OKI:M���O�?��M#�������O.�`��;��i�:��v�>I7�d���\�f���$�\��� 	��m��I7�{��6�B}�Pt�{a}VB��6m�Dup�\�}�YG���?��=���NB��o+)2Q��1`m!�W[���id����Z�E?���=��3��_��G7�`/��4��H��������5�������[#v������K�7��*a�2����^��w�����@�����]jQ�<�}���=��4��[n��[o=-!`�;�0�*Z�����>]��p����7�T^����!0�����Q���K.��7�{����@�1"�,�����'�+4c�/a�<���b�"�ZcU$�7qW���
�#/.�X��?r0`���B��J&��Z��S9m�@��3-��\��i�h.����������3����a�$(6�i�����h-T�����x�~��
�$��� *�>�s��l
Z���`��Q�Fi~�����g��u�]7*E�q�
����5�\���.�����^��=�)e�%�0v�\sM����*��W_i���i�r�]wi�s^� �L�w^��-��{�1��:O>�d/Z�����4�`��L��<:��6���mI�����^��~��Q������_|�EM{��g��[o��N�����a����:IK�nj2�3��<y���4i�N�C��&L��D��4
`+�C���1������]u�UQI1��8��M/�i��h�v��s!�V�PO�B2�\����-�W�[�
E��qM�}�6v�X�<���}s��OG�yqA�Tv�>����A.�+�X+ud*�]��5-��\��9r�%���2��	�����J5�(fF&S�N=������^�z�3�b����!C��o��V����)��)��"k�^��dI�i���g�d�Y��?~<9�u>��KN����[h���@X�����C�5��E�U�&��q�>�����]v��?�#=h�+����B���a�Z��\���o���Nt����-#G���Z�Rs"��k����`zP�.NK^���q��5�����I�-;�9��w��Q�4��V.i�(�8�E!�R$__r��$����i���k��/0���O�h�k������4}y���G�n�B����9�b�5$
���t%������w������=04q�D�P
�@q`E�O_}4e�ub�`M;���<��l���<
��.�t��+�(`��(�������n�"i�5�\��QR;��r���������j�+���?��s|���u
{�UV��-���`�jy��O����q`�/���w���7`������XkmD��\�
+���!��1Y�\�:%����k��f
�,������sL�qs��_��*�/����E�1�
�����������y�2�
<��3�3����Z��g-�v�-H�����@
J#<���m
f��t�*�
��r�����t��h^& @6�e�H��K�4Hy�?x�c����U��h�f
R��j��
k��\��������v��7���n��T�>�)�52�����>�EK9�����_5C9;7j5`����?����+���j�lq�=����<�3�s��(�b�����C�d���	|
s�`��G��e�'*x_v�e���F.lo�L���B����Gy$$5�.$S�[�/
��f���m@�pJ��h����/x��7>{���g�\���X�de��&�#g�

����.G(jZ%��5�J��k�
k��I��(	Xi�w�������x
��>�(j'&���������/�IQ�D2i�9��#�
�;u���h`�d��L���t���
 �Z���>x�=}C|F�)��s�1��]v�E�2�0�,
���EoW��G���`�|r������@M���!m�u�I�B2}���S�%���k4�'�x"#��M��;<(�/��
����O~��B��������n���!��A�m��o�J��n���$��F�w��������!������5�#?s'����gb���<��������drr��1������6�pC'���QZ���~F�������'/�S�����o"��������v����a>�d0-�&�La��/I�"+N���x!g�a�Y1�b�pb�w�_��z{1E+:���T��S1�3`���h��:��e[��v9����N4���,�I �XS
������ t��
Y��8q�r���f�&��I���d��iF�5=��;W������$�8	�6Nn�V����F<���	��d�@L�1a�e�XES_X�k������
k��Vo�X�m��k�� K6	�$`��]&J��5Q4��������z�8	�6Nn�V����F<���	��d�@L�1a�e�XES_X�k������
k��Vo�X�m������i?r��{��8b��u�]	���d�#��+ITw�	'��(
����z��B��*�Q�Y��$�W�y�c�d"32	h\�K/��E$����?,�I=��P#B4�����P���D�9�����`��\#�@1�����V���C��w�3z�����e��5kVN`���s�=7�V��~��z�}���7|��U���K�)�W��X�z�Sw�L�Au����4#�y������?m��h4�t�����K'�����^t���q����u�]���*��C�U)�"m��"V��
X�t����XEC��I@��>}�F}��j�85r
k��[��t	I�Qm�!��6�/����;���d���SO=��=�I(:=X����N���/:	����!2��{���8�^#��4�N��X����s�i�%��k��M(������O?�$����rNB����#
���C�g{��g������9s�h���w�qG'!���I�r��:H��	m�~������5�|rE��/��`�n�Wt�/�_����H�����e�i	�2Pyq�8� ���-�y*�g����XYd5�JH9���o���W���Kj�8
i�M7��n>|x$Sbr�O6Zc%i�v�4����p��������=k��-��P���{%�*m n�����b}
�Guq���m
�k����9��q�WD�r]�s�=�>4H��������MZ���gyA�x����i����L���u�a��{�yy1��G����0`@�.��m~�\y(a�����1�4i��r�-�W�~�k����u������'����XY�$�u.�C�JO/16u�f��$������y�����`�
k�^>`%�h��/����|����Ub���'j|��mOc�AH�'�X�Y
��x*{������jZ��_~�E�K|��������8hv�a
P����#�;2�<>��#������_T��
��-�=�X$��O>���'�����4�<�r����Q�Fy���w�q&����0`�Uv�,
�d����\��d���H���v�-���=��(ML��/�hO�K�hgqz����|V��������F��z4d��x����\��������	&D@����f_HPu��s�(y���Z���|�I���#Nh���w�&���I[xim���4��|�����M:c`���\��h�z��q�Z���
k��Rv��5��j=[>`����n\a'�(��w��|���o�V
`���c������k=q`e+
�!��y���kM��(Xe]U�`fE�����"Y���P�����+�^��6�Di����K�"���
2��������]v�����A��-W���^�ye}���wFEupc�Z�\�.��@����X`e2g��t`el�Xb	���)X_x��K�n�B�����v�m��������a��)����+��i8�y�<� ���"��(4X�+�K~�5�T>����wRuu�n�Z��^t�
X�Ym�d`�q)N�h�_}���Z[�l`e�`�zm6�y��jf�NGK�b�-��P6��s�9��O�>����M;M9eJ�`j�&x��k�\_y��'2c�<���v�m�U���
k�}Q7`-J\���X`�C0*��b%N4�����l�`�d;���T�8��
S������E6�� hix��	G)4�\�G0�w�a��1k���
��=�h�R�E��>�(U��L�i ����e�N"���!mC0'�ss�d}���j�3�8#TW���u�H%�Tb��LI��	=a�k��N9��t1	b+ �����p�m���s
���zkO�����&�NCQ�>Y3 ���T
k�8
k��>��S�����za\
@�P����^��]�7����6���e�o���L���%�uU��p���E��=���I�
N~Z|����1c�_�<?���_{�����h��
�������;�mL���������k�����$�"�}��G=����1��/���������u�"�U�Ys
�o���������/���?���u���V�Z���?����^o�|� ,�m+�L�^��u�]�����@��k���A��x���qM������#=�K���7+�J�_}��m[m���vp'u�����_���0����;�8�����#�/j8����G����r��q��e�M��d�jh�����]{e�lx��n������FX���J�J?x1�/��e�]�#��0`��B	��K2�9����R��S�D�s8N�#���R�~����x�j�����)I��D
[�)9bIEr��f�d�����JN�������iT��;�B����#�[3��o�[�d���xn��A���������-W9 B�, ��R�8��(Sv�R�<�.�Xs���R�
��hX9���$P	��C���������Q=�v`�l`9B�L9��Q2�B&�B0`-$!{�X�w��f`��`w��:qp�����5M=|���$PJ	��R������v����U3���u�8��j��4�6	4U�M�`}�7`��q.��j����&�I���D��q6�5>�i�g��VR���%`�Z�������eU�9
Xkzx�s%��
k�Y�lXk|��v��5��,_=K����G?}�
X����s����Z�J$
�	��������^����$��{��8���u�]��$������(�$���\H�'���Nu�������
$`�Z�T
M4`��A��&�;�J������M�:��o��M�6-	 ���}W9�?J/�b���N�EV>��~�1bD�,,K�����J�7`���*w3�X�����g
g���s�����$�x��Ab�:	[��������`/�;9x?o>{�|0`m>YWsM��<z%l�
�s����������bvw�}�;��C���/�<��Dr;����2e�~'f��*
�fw�Vf�Z�CW��V��H�T���?�D/!�2 !����sx����}E��j���3�q�/��;��$�YF�\7�{�������	%�[z����<g�R��9	�$~�#2���sk��V�����5W�!�P�����[�_�����\m��4jL�n��{w�^zi�+~����u��Ae8j�(�)Rh�!�
p4�-!�T�
��|r]�\e,�W	��*��#�4��,O�K�x�2�7�(�>��k/��f���o��t�A~�E���z����X��1%V(�Z�����<����v[/@�1@��J>9(�A]��{��G��4Hc�������_���f�5JOu����K���^
/�aQ������+�X�,��E������#�h:��KP��o���^���k��QX#^��{��������y���
��|�M�F��eK\��N:I�W����f�q_h32�M�,k�h�AL�<&�z�L�1c������O�;�����
��'������{o/?
�����D�5`�&�/����	X<x��
U�l.�D���F��3gz��|���5M������Z����l`�Q
������_Gy�
�K�Tm3v�uWm
���~�����9�6c�
�~�)�Di��j��Zp��}Qh���}n	�����fJ��5Su{��b��[o��oY�
�"A���^#YI�m
	d�%Xv�>i�$M:t��0a�~�
V�����>��}q�1����;G�c���rb���=�XM��(
�L���~���kz�|���}���/k�~��WA��8]w�u�+-��<*����hg��&�5!���cRYK�U������%`��,��z��A���K86��0��!������&��$�������~M���2�S������
s��
��;�8M�l�>}�hz0m�����&������_�;l��x�����wM0`@Fz�M�X_x�}��`�%�������3��=Z�1	�|�X��=����~i$`��FJu�'��9��j�JM��q���5�@I�*�St��?~����5����N�\u�l=Q^q�JL�P���������k�~a�
h�q
�:p��xr��l`�0��w�y'#�l�g��1_'Q�qL*g��J���WY�U�X�eSWO��U5�I�I9PZ`
��7�X#�>}zv����	���ii���g��m����_�f�K�i�u�Ygiv�}��>�8�����x��!I�Yg%'�$J3�Ie-�W	��*�J��
k�l��I���	:iO�<Y��������SP����� 
��-������2�1�;��C���GZ\��hlf����6_Z`M�/�}i/�[����j�
�|�'0��|��(����i��i��q�~�����3���cF
��)
��b��,	�f	�^o��5�=�2s�
7�6����D��e����%�n���?z����V^ye��gz<Ze/�����E��"��`����54��
��l�S��*�����'�t��:����w0������i��������0���_o��4m�u��b�u�N�:y^4.��2/�V�2�,�^���!^$h��x6�4��]��J����L,��X��.S��oR@�B����;��^��rH���o#/������d�z �Vy��u@<b���G��b�-�k�x��2�
���{�
�	p�����g�M��4M>��������{9��H��t9�!����_l�a{
�^��3��	'���]vYO����Dr �:�Qo`�����y������E�y������8����
kaY���%�����w�����/���������"&�����l�����L� T���_~�DCs���Y�����r�-��F�����q�Y���2
��I�/d%{j����7z�T�C�j�d��
��I���[=u�H9i�����Z�c'/�����o���eUs�����8`���f������J���~
i$`��FJu����(1D�g"'���QJ�/��=0	,��
���H��5��� O5
+&��n�I#�����~���8F9���C��p�����I��XI��#
V���	X���r�-#'�N����X_M�/�����H��5�p�Q$
�H�}QM�Z�#e�oI	������n����������5�5"
��2w��������2R�����
kKJ�z�6`���*kK
X�*^c^#0`���,s7X�,�jao�Z-#e�lI	������n������)�*���s�=����>��r����C�I���p�����J^G�������`'g�:9�������v�P����S��d��vrr4c���������bg�Z1C��
i
�r���������6���#�����M�:��o��M�6��u���p����1��>�����v�P�f8N���Ow���9�������K@'g��{�I`w'���^v�U?���
��Lz�;�aCs����-�)����Y�� {
��E��/�h`m��*,gG�	���P��{��cs��G�����O>�`���|c��W��TP�,�M
FK6%)�MK�)^wR��x�R^u�Q
���;w�\KP�n�TV�,���K�8��@��0��������0P���3���M�>�IP��&��W\1c�����I�U'������I�5�_k����-s������NB�5�k��g�V�D�	|BF������L������v����i���&���{�==������S���Ss&���v��I���M��+�������s{���[z��3� ���M�nw�e�\ 	/�k��a-R��iz�OR[��8w�G��vJ�;��uk�zCyx�E�I�\��w��:q�����t��{w'��X�������8GN��	��-��M4"��%`x��,��������>��|����9#s3��)��]�U����j�@����w�y
4�#J��-��R�u�'	k�e=�@����#�hW/��"/!�4���!����h�sqF�D"��C�����_��E�G|U
�(�y��*�����v���+���']x���3������`�
��i�"Ac8p����M|������Fl	BNl�A�i��������y�qb)Kpw	��|��JL�����o��f��I�O�[o�5��E��g�]�v��p�
�C~����C������%6��H��|m�)���X{��������g�}�I'���dJ�_|�c�_b�%���H������
���mk�%\��\���^$
��Q�t������(���x���g��*�G�~���i��<y�'���{��i�4f�M�bM����?q���\�*��
j`���B��$b��(K��=���Ub��'�xBy���y���Tb��=+�����k�h�~��6�4�&g����W^�C����O?��}��#�i�h��������_H�k��B��Jq����N;M?����@}
Xg���cs�)�D\	��\$���;w�g�y���O<��9R;a���-pa��B��*
X�p����$`���t�����g�������&��T�`�!������7�x�����;J������92Js����93J��bv�b6�&L�>A;F�L��w�������bn������1A
Xy)�h�pu���@c�����_d1l���M��6��zk/��z����^��Q�4m
��V�u�'n$���R����k���>i��n������@�]�t����e,���kW���I�x����������F�r�b�Z�c��-KVY�TbB�5H/kf
�&�8����:9���,|0�
���)��Y�|�.��l`�uU���-1����0�����i>
�7?>���Z'��&�a�i�%�\e9���4��;����h��G��\���mkV^R���K���([6C�����Y�-�����e/>+/eK.���I��3�a�Gf��!��=��J#�J��l�
ke�K��*	Xi&C����^\��y�X��Yd�Et��?����I=\1��V&k���B�\�o������[s�K��'Y�!���[��N��f�V�Z���uYL�?�����5X�����{QZ��E�
\����xU���:�x�y��w28����cE`�
X3�c7U(
�*�r49	X_y���=@��.�IDAT�����L\5*&��n�-jF�X�0�Y����w�a������
��\H�l
�rw�y��;��k��P�P3��B>��P`0���
��A��\��5m[���(�x1Mg�xg�����x��vgL�qb��t�� ��t��%`�Z��V�6'
+&E99'���k��Io�@i�U�	��qS0�P8�� ������A�f=7N���- ���kL�L��rH(�L�Z2�a����Yo%-hfxs��@ �I�;Px�6��q
R���P����S'����c�H��������jRR�B~���������k��y��n{��������mC�n��!��B	�V�����������Aa���'���<�q�	��G�,������W_�iq'�>}�h�2[sXkE��s�#g)P^g�u"~x�(|�8f=Q���������k���x��u�w�����7�<e����k���B��*�`���_�I�V�m18����`�e���n�-(8����8M[�z3�x�2�����K����
o\���2�Hj�����d���C��H���e�YF��1�C�~����0f��S/V�`	���9��ksJ�z�2`���+i��
+k��<����
�L�u�p`m��p��V�����
��p1b�t�<I�9U��;��Y��w0k�lo��%<����4�$���,�������2��4��BG�n��&5_��<���w�j����	2�
x&��J����:�cv^����/�P�C?���|m���PB�b-k@�
r�E��$������l��l��m�m@?�o��7��aOj���������r�����ojv�P��O���C�b�"i�-"����7�X�!��\I'/�v'�4�����K*)N
S�������S�P�l�p��FO[�W�����`=�I�(�em�L����UODj�0e�*�Ds�ZB���An��?N��������m�^j��m�x%78�Yrr'We�p�]G�=<��"U_�����/�r�'D;f�������Tn	����&�"	X��8"�t��I��pT`6qT�f�d��Gvoh	������R����4�%�U��N��N��j�1��E[/W'�?'�6N<���a�jh�X!�:,b�Z�����-
��o���;��r�~ ��`y�<v�����$��0`m�!���V�0���-
��W��C#�����+[LP�p���$`�ZQ�Q��1`���i���$�6oO�6�@�%`��x��SI�z�<}5`�#{dX(
V�)���
k)�A�:d�b�%`��d������4`-,#�a0`��@	���R�1`��A�.6Y�Ma]00`��a.��|��!
�-�#��r�-W���P	p��������N�?v���a��
pr�;�����u��Ms�����������
N4`�������?��I@cfr�o."���f=����K�3y9��q�q9��u�����A�X���d�YL��hv�*���
������!��1-���C�i��9s�h@ek#�]f��U6�-�� ��+�����h�
���K�b����I���t�J��j�
ke�G��&���Q�%"��O?]�jZ��l�_���hd��N:I�i��:r�U�����r�?g#+<�����B�gk��d�o���[c����s5(����?���{����~������$���w�}5�L�+����{���h��4S$������F^5j���an����"�|��'N��:����N�����O?��=���N��i4��8oXb�68���<����{�����D��;��XG���N����4��}��NL�*�|��Sv����$D��pt�~����C�tw{Ue$������H@'N�H;N��?�2�I���r�zok��:�%�W��5��%�Kc�C�5n���y�m�E]�%����<������������7N�kb�%��8�x���y���#>;���?��4.'k{{�����6^�C��U��r���������m�~>��A�4��a
���b��&&��4A�	6. �qMY%���p�M���q�'o4�8��o	
�����GE���6^�k�o��FQ��H���p���!o=|��Z���>�)��2�	��g���
����o?M#x��S��]����[�D�T�����p�y�E2/Y�
|��Qc��Q>�U��0�I�i�c�=�y��S���` �n
����>$��={j��/����� ��_��9��h�M/�����6�x�����O<���	�3c�mm�P{��!����.]���h�<��>�)6��Z�
Xka������*j����G}�N�Ac%}��A�6|�����x����&
I�r]�[s�5#���s�*�B
X������y����[��S�j��L�?	0��"h~��Cpb}1/�+���&MR>C��T]?��H�;1��be�s���G6�p�	�l�=��~��;��={������J�$*6�zO7`��_@�����NF��[c%�8�y�NL�*�.��
<X�M�y���FU
Jw���;�5�G�NK�����?91#�VX!�����D�r����j�@.���7~�x'[D�K/��Dk��G���[�����X?
�Z#k�b�u�����ub�uO>�d���-/
n������f<#d]�>}��K/��^�1y�*��Gy��/qb���wX���v*/��6��?�]u�U�V���:�(G�Y��w�m�����������=](�t�k�j]�4�n���v�XY/����$�#i�!�hx����<�%/NK����b�����q�_~�����?S�q�,����):N�������o�������;v��y�BR�}���Z7�m��u��:2b;
�?[w��X
|��<Z9������������i��ed9����`@�V����V<e����*��(+f�@87
�k��+�h@�-(�AXE���Y����c�F��B4F?}��p[�o^0eS?{������_8W}��G��u�9�$/6"#{�Np���I �X�I����Xek��D�X�����kI�'�E$�Bxs/��(���o�i8EA$�L��4�m��F���B��F�9
g�@8H����s������������/!�����Zq�����b!&s=9)��ua��V
��:r���S�
�Ql�xY�n(
�X��RJ����L�2%���G�@
���������jM�uSMX����6���{�S�t��I=^/��2���u� �g�}�����,���;�h9L�� ����P�<M��o�fj����\JL�\�|��W��^�O9��6��p���"��KE��1�R?f_x�a�B������O�I�5Um�m��MlN�����
���H��5��� O.`e�
����Y�<a<h��;b�����������W�5���s�j�h\��%�S���h����y��^�W��C"�m
��n���_�u��{(�����/���q���H:@�cB
H�o��|�//�����\����M��\N���ih���\dH�t�*?�y���H�?�����X���F�,3�Q~����G~����������dR��a+����8�DUQ�`9�"J�3V�:�/�BT�����<��HO��������s�xe�� s�?�x�QI����r�}�N��9#	��'&u7d�'�YOk<�9�*�?K�U���,�*Y�����'����� �t`�[������$`�Zc���6Fj5X�������'�`'k��3[����
kMk�;e�Zr�V'C��7ku�J���y�]���V������%���I	������S�%iu24`��q�V7�X�W��Z�
k��\��m�Zb�����
kMk�;e�Zr�V'C��7ku�J���y�]���V������%���I	������S�%iu24`��q�V7�X�W��Z�
k��\��]��*���O<���'g�:N2j9	p"�D�q ��}����pr�e�5��k6`mf�Wiu�U:p�nv�+��	B.����?��~������+B~������k�u�H���3YD�����"���a�Z�C�<
�T`��S�Nur��
k��R�"
�>p��,1eS���L��0����
k�e\5T2�J�q��N;�V�/I�
9�
���RKUP�������V����H6�����o�q�=��{������o�$�x�?��]����%t[�����z�������-�I`q������VZ)���"X%\�#m�-�tL��DT�*���>wm������^��9	5�x��s�����o��f���$����vyS����Kj��G�������id��,��"�,�:"���5�I\'��4��:�����k/G��8I�um/���_}�E�#�cMY�/����l�������������]�*��N_��X�e���O��32	h�Ob����~����3���P�l���Po����2���6��K3�Ai`�[o�U�_t�E^��i��s�=7��*?]
(������6J#>+yw�e��v��5�|z����b���^B��#�8B�i�W\��'��@���%h�8G����k�����*�J_;�0���x��c�9�K7�7q���}�������������Q����3fh����?�!��I�Y������o����?���T����
��w�y	��O8���+ �	��4�"X<1h����ri�k�]�_[<����t���D����`���U�'����>�d��=�d�1c��}0�'��@A�h`�g�e���Q�E��{�����5J�%�	��D.�L��_��^��n�M��u�]^b��?�P�
l�@���I`8
$|NY;��D��b���Fl�w�����7�x�w��E��{��������������:�~���^4uM#h�h�`@J0��>��j!�<i L��@b���4~�����\E��?�����P�F^�'���g���i��E����5~a�Z�\����H���&	Xw�}w�X��3��[o��
W���zk/�M���_}��=ZZ�A��d�F��o���������h�qX��hzb~�&L�>A�8p`�ht�	��}�Fi\���w��M�2E���IL��Ny1������N�x
$fvMC��Sx�s�&�����O����M7���o�������z/�C��<��X ��������^�����/��t� �6	�*
�_eQ�W��������K%4D���&U4*��@���N�W^yeHRm��������r]`���G�c4.����j�r�������7����"�R�����X����3`��y��{�����o�^x��Q^�y`��S���}�YM�5g-+k��l���S�O=L�h�q��K>@��6I��U�w���Lz!yaa@s��'�K�����S���������O���iUr���t��g��ul����U�V^���Z�k�����X��G�����z��[�Pe��`�}�����~���;�)��S'}6d�-�Xy�I���Jy^p�/��&�8�S����9$�K~�5����@y�b�������<�xm�Z��Z�>��^�U�1��,����$4��$�
�@�
������=
�X�+C;���w��1Qg�9���<��������G��_NQ�t�!�����U`e
6h���:2�����:+�
��}���g��c�s������q��2���d���]�2%`��)�����"+7����%1]��$@��x��I�O.�U�)�r�5���_�������8���f�����m/q�����;��>���	G)�A��Z2e��T�R>^(��|�-�h���?k�C=T��=���J�������z�����+��-�%�s���4�b�����9Na�Z����k��4����t���$b�~&I���$+�gA<h�^{m&��;�s��n������2c��U����}���#a]}�������Gi�.0gRv�-��l���p�=h�����2���q���e����(QV��j���������<��{�QG�$��(il'
�v"��@�4ip�t�%����+�\���i6��'������'��i�~��q�+m��������>����4�B�����~>s�LC����G��j[���
k=�r��h��t��$`�shu����N���F��I8h0h�L�8��/]���1��#F��}�a	NN���o����uN�28&_��8P���a�&~�g=1�w5h���D��NCh�����'�����]vY��/����1����jh�X�1�v��Y5z�11O�4)Vd$>x�����
M�����Q/-�	/r�E�
`�<���e0��Sy
����l#b{`��X��yq�2`���nZ?��%�-��+t�2g=p]&�"����������^�h�x��r�x*�u^z���������d��Q���qR� G�q���p����y���TOk�D'���lkr��#@���1!�l�q�}�"��A
e�iO������K4b=A���8i	9�<��N���jzxK�9������
���a��>�AhTy0`��1���V���@�X[@�YUr6�������F�'
���Jl�
k%�J������R�Ct�.��XS�y*��NNG�������
kK�@u�o�Z�T�V��]�y+ �
��q�#�`�iv��0`mY�WK���2Ren�
k�l�kB�51�e��
k�E\�V�8Y+[V�-+�j����ZF���4`-���}MH���&����0`-������'ke�J���e�_-��V�H����e���		���0���equTP*`�d"96�=��Cz:Q���TkV!p����s<���	�3����B���;���N�Etr�n�����$`�ZA�Q�Mi���V�V$����b�H~�,�?���z�l��z���&/�l8OX�����4e�3�xh��U�������j���R���zK
�.`�8��+�?9�������%��k����GP��o�=
5�/o�gr��������W���o�p7S���F��/$'�j�?�p'`�$zK��e}~��w;�!�$j���3e����w���}���N��5w�u_�����'�J���T��
������%���f�]x��Ab�u�]w����N��9������^}�U�+a���G�2��%�F���e�Otgg���^|�E��K/��V[M��t����s�9��K/������n��9�sy�$w�qG'�����.(��-d$��ch��|kP$M{)O{%nlT��=D�9�������p����u�O�(qa����U�\s�k����&���v��3�\�� ��I���t��iN��z[?�&�|�`���L��vz�I'y	
�%|�����������]�z	K��W��r�y�%6������j������}���5�*�=�V�;�0����SN�������E�����M�������n��`��4��Wq��W'^,��r�}�'��8�!m��=�RK-�&p9�	����'O��U��
�k�y����w�Y���t��
Q���qo��k���8 
�N�Xb�h������8q���?�����K'�Z�dk��<�E�-	Xg����"�6��	P.�!Y�i&t��������zD�����w_2���:p��@9�����>I~�]w�|����4�:D	��kx
�I��x���O�@?s�L/�H}h����i/��s��J�S��c�i�'�|��q�}�i��=��s�w|���t=�"�
 kTZ	��V������VG��~%
+��������1h�b���P�'�|Ryf;��p�2e��9��#"�\���>b��b�U@�jF>1Ik���&M���C��	&�G�_5m�E��<����c�=V����?���j��G�����&W~
XI����E��������;�lq�h�b"�b��d^x���J+
����V�������\��7��K.�;ZQ�d�O�1	J����hiIt��k����?~��~�6lXF������{�������S�#�l���� vs�q�i�{��'��}�>}4]�p}����9m�+Z+i�.�lFa��.��c�����A�a=fi��K����2�E���8���S.`]�`�N�L��j�4{�l}&�?P��k
q�r��`���=z4x��S'} �!z�&���d�~�Z����}�M;m{��������%�����0`-�<k��
k��l�����Xa�t�{S[&m&p��V�(���M8��9r�H��)4�d�>c��Yg�������-X�;6#7�7}���!!	X�8����L�^x�Xq\
�YZ����4c�|Y6*�XK+�Z�f�Z�#[d�����C�I�u�8������(��M7��r~�q�Z�uA�[}����b�@#���:����E�W_}u]S�<��'y0Y�v�(_p�MP8��� 
�w�v�m�e�OHj��X��
0L�8`�m/��
��8L��EG�����/�[84g*����x��l�%��
k	�X,X�`��t1	X�����[�� @���W^9�p ���PR������w�%���7�eX3fLT,�6�:��(M�yj9�+�i8��m�V��������_=�i�����y	�����k���h��g��[xd_���F���J��b]�qG�4��lX'�=��*�#�v�aZ�!�����k���?�i�(�X����[<k�h�l����t_-l�,
�&��.���0�d�R����n�fa�<����b/!(c4J��
L��-���������@
Z)e�8�<��L���b�E[�C��u]�����5���+��+��a���K>�+�_@��L��(+���V�m6���4������9d���o�����n����{Di�?���L>�t��-���G��Y�������=K��
k�l������d�2Jw�����"z��k�Q�qj���R���bJ�1��s
F'&�\�Qm��X��%
d':4���F����)7�W�����dO��`'fc���S������7W��io������,�������mCz2�L�����J/
;y��2�E���8���S�#
��f�������W��;������i�p^����C*4��8@i������%`�Z�#\����F�U�����bb�5L��#k�z�8=9YN4[7�b)�$
�Z������|��*��
�K�r�x�f< �����W32�M�I������,2`-�X���
k������%`���2��
X�q���f�2�X��XknH��!�������V��Y��_��/�j����G�m6`-�P�e�I������,2`-�X��i5
+0���A�@���e���Jn�
k%�N3��R���G�=��;wv:t�)
9�P�u�x�
=}(g�J��V�`�X��%��$-9���tU����_O7��z�@����S.�
�sp�=���j	'�t��o��<���9�}��~�M6�o��Q���E4�+���w���o��� ��
q"��y���a���D ���������g/[�4j�>���	��D�'�-g�� 
�����%�|�3��N���>�l�l�9�P �R������i��4pC>������f��9��I�+��#�wM������_�cW���V��p�;��a��_���^w�uz��Q9���;�?���Q�4G���(���{����i�x�D�
���i�����1#g�b�����^�vQ��+����#�!����J�~�i��|�-����o��F�����Px�]W_}uvs<�'ByZ�n�����?�A�b����b����E�B{E�DX��~��=���4��\'�����rv���!S�	3(gc������
���[��=~%k}.`�9A��xx��z��#``��r�hRBc�1A�(����;���X�yr]
�Ih��I�Gi��#���C��|�n���Z��w������b�9���L�W]u����r����|'�j������\9�����1k
A�#�V�ZeD�Y�*��R�B^��G~��6N���lS���>�x)$R��Y���}�'�����gg���q{��D�������`�hD�!�?x����	�����?�����
k��UY[��T��ON�Y?�N��e���_����L��xP�������)SBR�o&�UVYE��eO��{OB����w��<N���,�������=:^m��r�7�b��>e�H
$�)���b�f?��O�O���������yi�����SNi���0��M��C��|����8N���P>�e�]<�t��[��=~%k}�����L$�`�mD�C�)D���V
����:u�;6/�k��V��5NW\qE������h)�����O�U������u�{�����;�c�G��o�:?���OZ���]o����I|�B��S��,���������?�o!��d��V�e������0a��'Np>"6/����
ku�_�Z_j`����N$�1�����K]7eBg,�.��m�m������;��tL�q*6���m[��cG���^{mu��e��KA��z�+��oJ�u������_W��i@HCA�;���f/�S�����|NF��������q�L��(<��@
X�N��|�}���t0����j_��?��zXj`�@������ 
�Si�Q�mF���x���<��#�d/[{�,�mq*6�%�\��>3	3s
�f�).Z=fC�'ek�����Y�S������P7���M�����/ ���K�Bf�b�7��x���$��W_}U�!�KE�U��O>Q�@q`-�����O�vIFu 
�:�4],%��>�6�0a����G��zH
'�V�4m������/�I�����5Nl� ��;N��?����`�6
��7O]��g��0�L~��$*G�u����3H�\����������������r^��)��7�<���x�X?�a/s�f;�E��-��B���5
�l)�6��jT0`��q.��B����k

k���������o?��{w�X��p$aB���(����={jIO�+��7�pCF-��$�����I����3����K��'��L�������&�����q��k4,���8y�@F9� ^4�u!��Ic�+'s�RA���+�`�i6���-ZPX���E)�*�_|�E6[��a	������|��7-�N�S
@��/��6�Y���t��(�'|�icR^�%1����^9���5�w\��`���b��w��e{|�d�9
��oS>��vZ����oReX���
da�{.y��Tc(�8�����?������0����h�qb��~�����?�z���j���{��8;��	��� ��b>`����N�&��o�]�Y������8�
���=�l��x��\�4&�l�4��
+���+G8�	�t������+��Yg���6A�����l�'��72�gj�<���/�dd��9�'���~���>}���l�R��d���/���/V�\�7���ig{o�YN�������,�.���X������O��lE���Ro�4(cT[0`���lto�
+,K-��N8b����A���^{-���U�3&��7�\�29a:�<yrF�R�<��3ZG�.]r���=�������0���b�������������}�'�pb�QG�0�o�m�����Z���P��i�������m��������yIC��S1���������O���������CO���Q���s
<x
�'fa����c6��� �9<����%��
���Y��@IDAT���TE�������kB���#����P��(�ATP�*������0�s���s^�~������^���s����;3wB����s�tWWW�����P��D������'����x�/����9}��{n����)��O?��}�F���~z?l�0���{j�9����L%�������������m�����~��v�d��
����R�����~�������kW���O�aO���s���������O?��M���J���g����w�3�8c�L�l����wr������[n��|���3���L��S��97������O?���=��9�/���9m�c�9|���SO=���k��L������{��$~���;����O��$�x�������]|��n��Q���po���{��w�"�,��^zi��S�����2�,�f�i��|�zp�����7���w�}�g��y�~��g����s�����o�����'��=���Kn��fs+����������C������}������g7���G��\T���������������+I��_J{��KVA�?��C�+��o�s������on�p��~�[c�5��/��6�X��{�7.�f��q9���{7���v�5���[[�������������f��Vl�5�X���'l�k�4d�M���� �X�h����@'[��
�v��%��D7o�
kqY��
���h��5��Z �
kt�5��0`m�
[��
kKts�F����0
��o �X�h����@'[��
�v��%��D7o�
kqY��
���h��5��Z �
kt�5��0`m�
[��
kKts�F����0
��o �X�h����@'[��
�v��%��D7o�
kqY��
���h��5��Z �
kt�5��0`m�
[��
kKts�FV{'*�{����\rI��J+��g�9��W_}�}�Q��V[������0[o���v�is���i�Q5`���=W[�
Xk�������r�-n���v����[l���M7����3'1;���t�����;4�!�x�m�
7��v�`��R4`�Z��Z7�
k��}N������w����<x��@�Z�x�����#>+�*A��xv@Z���/���4i����K4"HNevchP
�6h��Xl�+�^�K�=z������^���A��v��]s�5�O�>���O>q.����G}��M���i��5`����W;�
Xk����)�����4���v�����C#�=�\7d���B�?������K���_���4�T0`m���ZcX����b�X���k�@i��������U/���.`�~�����S��?�h��c�M�
�����6�����l\f������Or����������2���[7]�������i�.�Y�����k�
k�t������o����ot�
��H��^x��g�}������<�L���nj
�6u�V�q�Sec3J�������
�{���V^ye�����{��'�����V-���]vY]���_�Yf���b���h��5�R,��X���5��m��e0�}9���J�����:��?�Y���=������<���Np�n�ik*�Z��0`m���h�X+���eVXC�X���[o���EY$$��i�e4`��2]�����K}�S8�6Ok�%���4`�Z��Z��
k��xJ{
XSc����Xc���T
�������V[k���
kyzk�R����)�5`MQ�%�b0`�)�.S5`�����z`��Z�m�-O�����J��Z�����5E1�l�i��5��L��
k�jZ��
kk����<
����V+e��j=��^��X�i �
��2�2U���i����������]t���$ko�t�
k��E=Kb�Z��SC��X���^w����v�����+���{o
���o�X o�fJ���OoVX9�������8�7����O�M��4 �id�b
�O<��51f�/�`�����2q�D�3b���<�<���~��
^���7����������SN������-����z��'���F�Y�Gq�?����|�������������g����~{��2�x9���t�Mmd�'���v�a~��6���o�����������+��c�=��������<�9���Zk��V[�K�&����'�E���~��ff�a��:�x9���w���E��)���L3���%�q���7�x;22
�����?~�-��+���~�������$l�H�Y�>}�j�����>��(!�t`9����=���~��/���K�.�Yh���~�YgU�/��~s��c��-^j~�<���*/��o��&~�=��r�rN���?�~�%?Z����o���/��B*��3�]#�i���G�K���{����<���[l���_|q��j�x=������?m
z�w��I��+����$�O7�t~������{^@����v��3�w3�s����
5`����V
��,�m��V��W_}�X
�UW]5��@���7��~��E��]w]����������kt������>���{��;���BR��5�\S�+���o�{��Q��
��g������z�������V���5�?��3�`�������^�l^�.��R�B���m��</XI*E?�!nN���w��W*	/����${����������Imx�R(�u�����3~�X�����O���H�����.�����O��������M)����^zi��RK%����4`��`V-q����p���G�;_|�6�������:�\s���M�D.4����/�$���x�y����7_��9���_T�l�A<����S�_x����R�SX��)�,@���s�����M5��V/`B_$]���l�,���z*�H��=���#����l�M��������~m�[~���4`���K����B�������z�/�K<��P�
�i��������UY`��b���|�m����������<��y���|�s���k����o9�'�tR������X��	�f�QV���S�z�����_��@Q����N�����g=���|��-H,_u�~�������]bm������g�-��`�0a���k�����Y�?�����Qck������b�WXG��s���/��R�M�����z�<�H�EV�$�����5We�J���P�w��:�w��U��s	`�� �a���W^���s]���+y�����(��@�T�~����V[m�������,R�5_h�\�{�����;8����Xq�����p��R��e�G���WU������,��m�����/�d�\�E.���$A3^,X��sO<���-�|[�J����Gm������h��u���U����F�Z��
��yV�T��q�x
��zp������xY�q��3�4�^'���������k[��M�2%Z�K����yUVVN��o��qQ�\���
KhH
�6d�U^�J+��l�4��}�����;p
m�)�e�,��e�r�-Yp���-�3x���������
L�F��_��[M������� #VNU����x�a���=X�I�|�������<+��}��7���nO{�M��BV(��,�c�9y�-F��b+�a������Gy�K����%6�X���#p1`e0�7�����K���6�|s�����@�u�B�,���9��q��n�ii��W���\rIN-a
Jr%h����
���_?���!C4��`
�_���<|W��q��k,2���R�x���I��Br��r�+'s�TA�
$^�Zq��f���~!��/��"o���b���Yzcj���1���RVV�2�$������m��{y�g���|��S��0/�ku��7��.l��c�=r��g�K��^L��\!����,t�UW)�a���^���U�g�vz��9YJ���V�M���U
�����_Z��Y�{����'@u��A��j������������[������:T��������t����i6���Z��Oren�\{�Y�$���s�9������#�%I:��Tj~�������l�@&,�8�?i���)��,�2^�l/�d$�����H�_����L�����'}��'q���.��x	���pk3��\���r���O�����[�Xe�a��<��������[y����e���
�T~=U]XyC����.,�`��i�v_}�����U�<c@\y���,�����;'%n���j;��c^v�G���Yg��n�|Tj��&���c�z�#R�]�G�qb��;���H����5�X��*� k%�<�+��G�C������Za�����x��\�Jm/�%:����rXs�"ur���9�����Az(q�������@��oe%w�J�/������r�-Y�X�k����
���

+2spsVa���c��j�$,�0?G>9����Z��I����
�|o�l�	su�
��������-�i��f���'�l#2��R�������%��p,�R�e�+h�+4�������N���^f����x&�Yx���,TJ{�m>���m�"�y^x��p�`��%�y~����k^��T*�x���aQVr
#ky�W}
Xt�/1r.Kt
GD�w�}�-��"N�_s����>
	�G,'s���*�@�'��N19�F�����tn�UVq����'�Xj~�(�O'�i�XjN�*�.��m��X�n��gwb
����?z�������E��X����q1*U?��z^J{��KV!vN\��W����x`
��a)�m��_T� /�N�9Y��%���
�v�����,�Z�r'e�9.'��'�����[��-Xk+��VO0`����@Y�X;P�Vuh���:�M4`�����
k3�����0`������
ks�c�[a��n���
ktr�h�Z
%6�f�EkC�5`�Zm
7����v�����*4-�
���
4���Jl������jk����n������n�
k�UhZ@�-��h�
k��,X���
���
k�5��
X�����|�*��:�/��[�_;�3����7'�	����C��H��R������C9��1N����C�y���y�w�w89���Qs����%���C����u��I0qw�1�-/14��I��~NH��b����x�	M����{��c��j���-�������$P���k����3�0��6�:U�?5�jh
�;+�`��m�YtN���p\_�����+��e9����������r���O:��(X���������,����r�X��n��Y��td�w~l�?q[9���!�l��@����&�5�}�v_z���#����:��x�8c����L�4`���N=��4������d\����`�K������L��o�='����������������e�]�<?��#O��d��������5<�z��5�U��P�	_F�@^��A��#�R��]g��|����biI
�&5���i��:�}�`".�`"���X�Y2�
�.��@��%�*
��z������\`E�1K4�/���k�i�`��������V��h�;��7�h�����6`��n�{!X���j#`!`�T�Fs
!�p��	K�A��bq��&���A�	
��G�?���O+��p�<g�]�kb��C�����G����?��C�x���s������J�3��y���y]���^&L��	<�����_�[o�U���X�A�����cmR,���o������C�8���/��"���� Ii���_����+����+?����8���od��!����?i_�d�!=��x_��x+x!��E�?~Li���J�Y5���X[����V2
����@>N�HxX�|�-��#F�Y5�7s��&�=��������I#���[n�_V�p3����9��c#��b��~^|��5?A���m��e���_~���[�v��'G���mA'�>��C#����������@�G���������
�����n;
��\q�k/x��k\��^�	�
X�=��tt��k~�WT��������8qb���I`�E�~!������E��i`N�x���g��Q��X�5����w^�M���V'L�:�>����<�H����{Y��e���+�������90`���j��^}��:(�Ybpj���:Jd�
�1�s�a%�
y���|����
8���2�y��:��j\��9�b�P��+����k���<d�%�R�J[%v�������=�+�>�^ ��i��F�/�����
N�U���j���?��3�8��kn�w�g����S����4�����Xy�t�;�X���f��3�>�\�X�X{�'��<�~y�Zj����R'���F���O�,\��N;��
��w��Ki��oI2`��%�c�j�
�@1`��(��ub<�v�������I?������/�W�����7e��X�q
o����u�b�!kX� I��"�VxJl��X����X����:)��M:�B�M�gv����/�P��H�Oi��I��%�{�W_}u���s�f.=�<HZ�����Lq`�
V�S/9��RgV<q�Wg>�5@6>�o��1z�g����[������I��}>
���J�VT��{� �����\��3�����)S4�
�
�y���3�t�]v����5�&+�f�d�����*D
��`���8��������u\��U��wW�����_���/�X��AD�X��,�LH���v�M��_,x�+�>��@91N���>@�8��2�m�5������9������w��2�:�0xR�4v�X���SN�x�~eM����X�h����o@5�X~�a�L�\rIuA�1����[7�����(��������W_E�|W�����.��K�,X�5�t�eV��x#p7������:u����bM�\�@p]�]������
��t��6����
k����0`��v�w`e�<0���HQ��J���\�I.

+�t���J[q�����6��yV���Y-`e�z�@�V��X�Ip�����,f�S>0O����n�m�X����jj�����m �Y����8��7�;����D���L�|���I9V�z��74-�
f���o!
������k)�6�jU�N8��(k{V�$������ ��A��>����T����_��
V�-s��t��������X��c]�����L���g|U8<�[7K��d!'`�)�����b�+���
kXg�������0`m���|V`
��a���������L��x�U����E\w��dE+[MX����������9���pV{�=:$�7��]t�E��g�
��B�"�`5�
�Yq
��~�B�k�zp���

:��a��-��-*a�
����s��Y�����Z�d�V5��@�_R�Y%��5���������u`�.d��Ox����`U��
t�Ygi������/���*fd`���k��}�,X�Wg>2�LyV%8p���,�#_r�T(���9��F�>�X�i���+�a��W`��z�����>�@-����<���9��X�t�}Vs�����@��O�ae����Xt��Z��
b���^GV����m1����P?�.;:�#��a��2����
�������UW�n����VX!R�����d�G�UVY%��9�9� ����d
�(��X�!?����d�V�Y]��
+t��N��p�
�EP���������^a��������+���V�����P�:�
i�ox�y�W'������4^R���;N_^�����O?]���i����#�b��h������s�����E3'�y,[5��Pl��X�XuNNrbI�Z���N���E��t�X�	�T�w�L��u�VI#
����r28����������u�W�"k�r��FL���P����i�f)SJ�nS��Z7�
k��}N�K
���vch!
��Pg�����P^35`m����TK���ls�5`m��,�5�e��
��X[����T�v(����6SoZ[��
�ji����6W����Ug[H�-���h�
k;��LE
X��7�-���
k�4�\|
X��?�n�
k����-�
���v4����k��������ji���Z�m.�����e����l�Y���
kuv;�j���5S�FV9������owr\�[i���������$x�����7��$��������e9��y�N�89N��U��.�����+1`m�.�L
X�/��"��7�89d�����i|�p���O�#w�����BNb�V���r����k	�������2�u�=u#�
k�tE�
R�����X��{l���}�:9��%�5(A��89���9�>$����/���,H������K�0�[o�
kY��r�X[���7��u�w�������H�
��������Dsq;�u��9�M�_<����{��������zl�
k=�J��d�Z}�!V"�HX0�F$��Z��#�*q�����=@��an���v6,s��K�8�N�;	e�$����n�%�<����5:��fs,���0')�L�����$�\N�-a���&sX�n�IH�(Q_pk���/��\x���g�.���[-����k��_~�6����B��I8;�x�������O?���q9��O�?�������|�I88'!���e
��;�����
�GZO�������s6�}���N��9	y�yxQ���������3�t6O-��o,G��1`��7b���`d�����B�WJ\�}���Kx1�GJ�U���_�8�9�pr��C������2/X������G-K][n�e�z��w���;��6�lQ=�G`��!~+��e�� ��K��Ey.���sJ���c�j��+��2z.��qK������2M����B��<����T��E���4d�4i�'�,z�W���nXC6_���=��������J<��_|;V,��1y	�N<S��R>�'�8}H�b��Hi|\*:th�������o��_/��^z����@���?5�.
X�
����/������2���95
&-�c400��O>��2T�,]*���e���)�u�E�?���6g�&@9$�����#��\������GqD�����,Dpq��
�i3� 0{��������*���n�i�#�E���2e�l��~����m�����V�������!�X� ��I���K@q����
Xe�����K�.^V4�o�����u�u�'�����X��g�K.�����&
X�e�L�l���tq�j:`F>YL��/X0�/��n��9��2@�����n�����`�0a�~'���������e�!^_��VX!*�`ma-�Q��=e�n��
\�y�M��St��[ ������*�w����{�����5}��������Y\�Z6+ic�����O>�[��gX���>��S!���
X��K�R ������X��7/s�:(b��	�#�;�<���=VI�p��,���B�
XeuiN����%J��\�c�]v��!C����G�����9��c��j�]w�U��8��^�t#W��Uj:�'�s�U��d��������='���K^T��n���|�I���O�9���U��{�L���R��	}@�8��f'��SN��A������QZ���~:J���z�������>���R�V�.?>���n�L/�����b �zV���~����7X|jh�,Z����I@�,�v��;�h���_?-�����>�Hy��_?u��~
XI/WV�
+�xA`;���?d��j�?-�
��
���V�1�����D�2�,���V��)�����{�l��w^��^&�'OV+PVk�/��=�Ee��^�
`��M2���t7��{�9X,��B���EYP��^���#���re�Q�~���Z/;�E^�(*
��ex���Y����+�
k��MM%K����N�A�����Z},>(
X����t���+�
<�\���e�v�i'M����%/��!z� x��Zk-]�����p�r>���+�����+�����H��h�m��6��=�\?b�-�U��<2�a� ���v���
�5�\���}��%�K���*GSjqW0/Q�7�"�B���:��:����;�V��
�����|���}�* �|�Ge���*fP�5�A��c�94=� ��@�e�0�]��gd�g��U��'V�^{��Z��?M�}�����k��������U�l�+s��E�Y���U�������������0wH9�lAB��Yu�e���e�����G����k��<Vw��`��d���� ��!��g�a��d��2#{��N;M�������6�l�o��/�n��u�m�Z7]Q����u��N�4`E�����%� 	���h�!��YrX��"���C��Fps�z����d��,�X��5��^��%
�|���i<+�
z����
��`��`c-1���%�	'��s��m��������d����9�8�i�h�����EV�o�UW���m@P9�`e;���q1S{Z�=�����0fn=��9��u/e�Q������������Dlc�����$e���Q�������N^��b#��������l��di���	;����5j���>��dA���T�	<�,D]2���<��TOS0���d�tm�����ZO*���K���I�!!�N��w����8!J��N��]��b�:N��+N��o9Y�PT�,�y��o�c)��r�W��Ao��T�������
k��]E%����p���zT!��u����q�F�5`�Z\G��9V��j	��8�
6��������J?�P�y`�%���F�5`�Z\G����~h���jJ�x
�������r�0L�!���V��	/]�J`�Z�}SO���ZO�����v��;�j"�0w'"�`��k��5]7���0`��.Z��������Q����X[��o�4a���t
������W���EK_��t�[�3j��5��Z<�
k��B�
X�&��4��
�t����j�����h�+���~k|F
�fTT�g3`m�@h�
k��}��81���^rx��|����;�prTb�D6`����"�����	o�Z9]��i@��9��J�'�>;���FV��\�������0`����K5`m�.��zN�&8	RP�FW�^	��^{�5���%�f�0`����"�����	o�Z9]�3�g�}�u���I<[�����������$�����s���A
�����a+2`m��������y���';j�8�:���v$�{��'5b�7��)I	9��
���~s;U#�����n���s?����?�f8���� �.������6xbE�OB�i^��\b�jD��[LyI�4��f�i�Q�?.$\���z%lZ+�X}�A���z�F���Z`����KV�sO�	����n	M��W'��j���
D��z����h�2�7�� ���D�z��h?g�y���%,W�����#a�����c%d`-�P�����#!�T|�����k{~��'�w�����\!O)�fk��}�@cE�3i��
b_?3I?�����k����&(�@�t�9�x
/��_?�&n�I'�e!���;��1?	^��;x���@�1L��*?V/!�4�x�K,����|�������k��$<1[�yJ�U9H^�}��~�y��r�~x��20F1L�LPvb�O�:$���`���-V|.��m�%�\���������J�%��n���~�W�W\q���������?���g��rv��B����k���U:t��H/�;%-
��[<��;f��w���?�R�b����7��g�~;d"x����������	t!��|��)&eJ�7���O�����:Oj�E���U,���g��XL
TAM7�x���D�m[v����P���z��c�9FA"+i�f��J�HZV�@DG��c%���B�|���|��5�K�.^V�z��&�l�i�)P��}�E!�oTe�I�R�/����:�GqD`�8����D�QY��X���
�CF?~���m{���?��3�
l�@����"�}��z$V�uwlT��q�T�|0JK���,��<=C��C=�%�.I�S���7����:M�i�i��4`=��#u1b�g�����RK-��}�'L�}�Eq�a�)H�n�i�Fo���V�6���UV�jZV�f�w�qQ}��I�����z"-����.�L����GyD�i[�h_�,�a��;��������J\�
�q��kqs��W_�?������u$���m��s(�L!o�;��l�G��G��"0<���T��L�z��v�~�=�X`�
M	��u�
+���UN��F�`4`s�2r�����_�F�����1���(�ba�<%����g�y���\���s�q#���`�G�v�f����m���	���.�H�2w����$��Np�N�	:
E'�C'��H�\�#V�;���tn��b9�SN9�
>\��y��N��N����m�����z����|g�o���N^.��3����g'�h'���+�
���]�7�y4���2+^�;s��������u�����~�i�_�Y��aA�`��g��m1�z�V��7�o����f�������3����X���x����*��l�5�]�����k�ZHi+*��R�1�
��,>��~�]vI��{��y���7�t��3w(X'�,V
p-���o�by���D^
��9su�+y�]w����2���E�������G�h����d�s�kW�C]�XOp\�TC&���%���;�L9���	�#}���0
��Zk�<�,p���F����PE���f��K7���S#^\t��M�H��3e���~3�5�a�N�����4�b�i�*1;u������b���#����^������{���������
�EFq
�@�gY���F����$Y��/
�g
V����X�IV�
~Y�c!���U,,b�2���6|C����
�Yd
<��
XY(
X��CYY]��
p
�Po��s����'��q*�Y�*��X��`�i0`M�L���
��!C��J#m��O�<P��Wd�cN3N�t�"7�>�c�9���~e�9�����T�n���(_�`��A�
���^�&�@��U�����Jd�c~4>p�k�u��DY�w����GEO�� ��I>Zy���9� ,Vt�j�@i��E��#��J`�oo{�a�-��W ��I�;P�z��v(��	���H��
�I�S���7����:M�i�i��B���#���{��7g`��Q�g�c���mz����Zd���Y�C��N;��5�\�V���4\,R\���
e�vp�u��U���{n����o?���c��zM��4<Q/����"���^[�:u���Nd���<�g�.��>h,�8X��E�mC�m��;�m)�F�,�B>�/�`�
�8�d��M�$���m��F�����=��.<y�a�����������b�D��d���~���}��PS	�[os���7����:M�i�i��B��< �>VY�a��8N�V	��4�q�>W��u�-%�5�>�X����J}^x��R~���^J�j��v�P<l�c)������=���u.@L��o�n,en��_���q*V4�&��b�
\�z��
��
V�+r�QGi���Yg���bC�#��Bq_L�(c��-P���?+��\rI�'��z�G���wB��n��KA�z�;T��C��G=������f7��|TL�r���5��--�
��FZ�>
XqQBB�1�*�*E�4$��u��;�D��0�_����2l/����@�|�����|S��������i��6��"'<��e/�)���8�	H����i2*�K����+����b����������	���-���<��8T"�7f�6���~3`��u�c�m�5�(}�M�ts�UW�
D[_�U��5
TR�����l^^����%���R��c-�I	��_*+��n��������P���:N���F�����z����
X9hBVl����D��P
Y�S��k�@e4`�Z=6;�f����������m.9R���4��vc�3
��Y���8�u�1��V�Z�vY}��Jj�����l^^����%����$uY���
k�v|��6`-Qa�����Y{��UI
�VR������y�������.���0`m��/���%*�Y��6k�Z�*�
�Jj�yy�6o����FV9�Gc��q�N�wr�����{I�mOf9�H���������
ZoD�V9^�I�'�[�T<9I��Y���a{�s��X��'��;|�4���4�G����~������VC~B�q���k�x��j�U.�
�0q��'�|�o���	V^���Z�iXU�6
s;+�i��}
i$`
-%
�Vm`���p�|�;���W`%��u�]�	�]"B���?U��K/�F���V��
k��xJ{X���W`�a��Fm����w�Y��W`���7���i��4�M���y�1�5h��[I6�ZeW{��/6�*�m����gk���[x���4M,;��SO���z�-��B�W�^N�I���v������o��t^��O?u/����I,R'q\sx~��':���*s�N-�y�������q=�3�rd�2�,��_~����n$�{�����>M��pg�U'1au�P��9	��$��� �@�@��g��������t9U!�Xp.�����:	u��GB������1w+qZ�D����O��U�I�>'1p�D|q'W������4�&��4y�dm3�����q����?�Y���3�<�IxA��j�9����`u�&Mr��=��C�'�-Iiu~���N�*�d�����+N"�5} !����l����m"���amM�@!�U�h<U,D�����N�aFt��_�x�+���#e`�}����b�F�r�w����!qU7�xc���+�+������G����,���b
�����`����oV�	2�����y��������K�3�G�*���JY(X��vX4�K;(//��?��`�>`�X���}���5��q��RK-�u�����Yg�u|������[F���[o�Uy�n�]w���������n��*�$ �y�Cx=y����W_}��3%�i���u<���CUo�D|W��Z
+�C��O< =�
���&�
qw�8���_�k��S%B���vms�i������F_�
��7�������*Mq�����@�
n�����I�k�0�G~���A�k������W�����
H�S>`}��4�9�m!z����
{ ���m�Mq�9�(�"�,�~�a
�M��K�rH,U��#G�=�&@����]vYMC��\s��]�G���PG�'�<}���5��
`�E���POX��T]3�
�`����u�����`E�#�<R���*�\s���,u�~K�����x�@�
���J��]�
kc�_��OV�)q��	&DcL�5(��?�x�<�#_V�)���J
�l�Yg��b)�������4j�P��2�E>`�y�x��5�����k9�����X�P�-��"'_V��@�]v��<x�&aR|���I�e"�I;����;�u��y��%���������7���5OX
8��tt����"j9��(s�"��LD�
X�����B��,�/����a�"v����,"�%��R��[�
k��]E%��X��2qY&?2w�n���;+0���S�n��|��x`���^���������<��
�����G^�������&-����w����p����.�6d�����;��c���j>��P.�&��w/�p�C2��� -���t�M~�������}�������q�g��2����@Y�:�M~`}����G
Xy������\�WJ�������JY�����q�U�6`����������7"�2P2��W�8�`q+(��*Y4)��Jf�����U��?x\����T.�20#3����+�t����m��9R@�Vt
�3_��7z�h��J
^Z��f�j��
+�q=3���)���5c�?�k�u����*7�=n{���\�*�d�ZQu6��|-eIDAT-3������X?��c�dp��,a�L���@�`-�b���+�'����
O�(��m��I���,��S�5�x�;��xQ�����

+�Y0�L�z�.n������N�49]I]���;���r�����k?|�������k}���l>
����@Y,�R��?z�<�@=$$�W�o�Zi���1`m����������3fLN],�a%*�E����3Na.6k�UW]5���X�,N	�xe�����X����j�S�������p���A�r�]p�Z�l���0����3'�N;�����`�E�.N?b0H�2��v��)�TX�������lC����y�PG>`�5-��B���+O@
���Q��EX�,�F���F��+�9W�
^��:�OLy>AQ�5�0`�����
�&��J4!
X/���h �=���B�o��ed��V}���k\�XU���s�B�(+��;���#�����qp�2���>l�	y���.����9I�:~Px�e����Y���X_�_����y�.]���]�	���k��uc�
@�5�N���v�������"������c�6��l����/���a�&)X�,
���2���9���,}x���\22�b�G[!+��4���k�=������I9�&���uT������m������jI���9-�7���r4�P�����<g�<��={$��
/���#��O�-��t��`Y�]�W\�Gu�2u
,��5��7��K@!���na��T8���P�������/�\�8��V�p)����gae0s��[�<^,�'Eg�b��(�g���
"�=��7��>ORW�z�����
.��_<�+�U� ���2ey��S����
�l�
m�B��&�P�.^��
0��|a���:
�|�� O5�
X�����m��<}���V�z�C
���*cQ.�@���g�
c ��I�z[2���4��}���9���/��7@.�6\��y�P>�7�\�W.�A����Q�y��:�N{�(K_'��B����b�Y�������y�� ���z��iG�+��s��4,GG���b��q}��(����Z���B,����S
�+c����!�������������:WX��������tb�8Y�R�\���8��h�O��v�hD��v[�
����x��eq%wHSX;D�
W�
k�uYu�$�rx���� �HK�m9A��:��f��l�q2���*s�����vX�E�����������j�VXeaS�&,��4�M��o�@V
�~W'�42N��EY��������^�����-�$����MM�
�&��4���Jn�*X��L����
kG�@c�o���Tu)
X��b��	4`���X�&��@��P�
k#�����0`��h��
X���.�
k�Ul4�X��k��(��0`�m/����'��}�r��c���*��V[�0`-Ye-Y���%��m�
X����)�Fu��g:����8@����
6���U�
h���Jl
�-��Y�h��EK��#a�D�1`��j�����*jm:��M���5���<���T������
X����7`�������J�_X?���\ s��J����9��Gu[m����oN�������I7-��������/�J]���~�a���uI�T�)�Y4m�����F���qn���|T�m��fn������y������q�1�����_vf��"�R'�I�4�Y��V�d2`
�h�o������������@Z�8B�G���W\q��%��XY*�3�<���n;��J��=��3��y�}�iN���������+�����������e���C������O����>:��z���G�)'`N�5��������;�����<���s3�0�D�]PV��K�ob��3F���B�6����O�%s����,l\�wP��g�X��#:Z�4`���:��@��lLgt�������h�����DG��y��5��"�s�����'i
X�������kA�	.���&�,�*�j�$�X�
�r6m$�P�
Xy�E6����;w�x�7Ny�A��5ROC\�6D7u�����!@��[�������PA
�X���O��2��x ����Z��;�8?a��r�!�����O?E<�}���|�b�����i
XI���-i�]v��
:�[�E�����n�e��n�A����U��w�]�^CTy�'��5h�1�
X��:ZJ����:�?
X�x����j�����g�1<��P
%�<�q1d�M���k-���\e��c�j��rJ�j��I��0�'�x���;N7�tS/�e�-����{�]�	z�!�
k�Dc|�6F?u����uR!`=��3�l���._�Yd�S
��
k���H�~���5�wV���Y��c�=
�^{���U6�u�Zk-���s�=�%F��q�	'D�X#U4��
kCtS�i���]P�
��W^�`��
���b��M�����'ON&G�Y�+�f�#��Y`�r-V^6VYe���G��
k\��ym����Vk�
Xk��:�/
XW^yeGAX�s�1�.`
MI�X�0
�����;���L���|��P�����g�F������Xe[����,r\|��Z���IHQ���F���YdcE0��Yg��s��������;���<��v0`����&�F��
������6�K.�D�����}��G������g�9��m����k���l�����������n�9��cJ?z�h-����j=�������^S��8p��#��N;M����E^4��.���G��jZX�����Ok9��"��*������gd��������8�q�����Q�h���~���%1`�����li������A�U�Gu�.`�C"�Yg���>
>����;#��}�]�����`bl!b����Yg������/���3�<x��C!����#�^V)�~����>X�����
�!Z9�R�^Yx��5�l���-[���F~,tf^@�
�/)��r���u���Nz�n��/����X���^;yIF/#�
��$�b��_�y��7R��_��m���-9z�2�,�X�e��SN,=-#.g��S''`=/��������s�N�������
0���%'��S]����K'/N@^����#�[sx��W����3��W��*�
;y��:mF�����e������i�`���{�i�e
)O�����J��Z�����5�bd�;����n���������d�-�
����v5���]�k����}�
��s�U�Q����?��b���v�R0`m��.���e���
�����$vr�~N�,���99�v�R0`m��.���e���
�6WZk��
�������6[����2g�ZJ�-��e7���l�5WA���OkMu4`�Z�6W�f��2�c�Z���XKi�������������*h��\�i���X���f�:
��5[��=�k���v��m���������;��^|�E'GQ�H����h`��'����&+�Q~�'��7Y��9���i@�pvJ���r*mJN�
n�n-�Q�
.]gV��4`�����rZl�Z�����
kv�5��0`��J���
kSvk��2`-]gV��4`��z}^N�
X��Z�1`m�N�&U\�WiS24`m�n-�Q����J��X[���i�
k9Zk�2�M�����k����*mJ��M���7���t�Y����
k��y9-6`-GkMX��N�<�M�0����{n�%�t��_�,����z�-w��6�FZ�I�r�{��'�4g�}��j|	�5`-AY-�����;?��B�z����=��������������w7�x������v�m�n���8+�n@
,�����>Pp]c�5�����6zn�ZX�+$�r������~��W�������[k�����j�m���c�=V!)j�����V[����J����j|������w���j=����h�B�-������=��[u�U]�.]����f�i��
����������QZ#]<����{��������F���dm%=���/�~e1`�����di������e�YFe�{����g��f�a�H������;/����/�P��/���p+.���9�?��cu's������|�A������7��M7�t9y?����������W�[a�r�c9?���n���r���?�g�}���zk7���j�������rK-���p�
u^���W/���g�UVYE-��g�Y�kg+3������/����?e�z%���a~z��6r��:kN{hg��!#}u����K���.�X`
7�����E�e�]������[h���	��[�[l1�/^��6�,:S�����{mo���3t�>�e���[����UVp���(#�������O<��6������Yg5�G,<��k����9�/������~��N:I������z^@��y
��/���?��y���h��+�����
��O/��	
=��3^�w}�N���^��#^b��[n���9��~��1Z�|���e�8�?t�P/.n-#���[n��|�M�]�
d�����l�����_}��^,z��?�Y�����h��N;yq;��O>�����*����)&3/��/���%�\������C�#�d�2�q�\�pI��>���{���G�����K,�D�_
/��WH�7�|��g�y���n���w_�' ������9���>'s�o\pA����i��\����u4��h�_����R�g�Q
�
0N��I�����Xf^Bly���X�|�;�8/��4i���[M'
:��s�>��o���
i	��������#^bM�3�8�o��&^�*�v��Y��g��q�W��(M,fM�v�V�
dWA^���mx
Q��}|��'������^�b��1X���u��8����8@A��,�G���^Y����o�����4������X��_���@!_\��D�����_��������������>�������v��e���}WQ�
+
"Gy�ZY
�8���*�Z/�r�O�t;�0�'+O��������/����[N����W_}u/1b�^��
*�r�{��KY�`�bE�i��w�;3Jw����*��eiz@#F��>�<#f\<��#Q��3��A���X��F������le@
zP�5c�?P��J�}������.�4,�@
X�z�@��wCV�k:�4���ie��n�Zm
7����v����
(�]w]8q�B2'�i�c&?X��$�V^ye���H8���{����E���H�����q��6���{�������@
�~�iM���w��O?}Twh[���������|�������������/1t�GV��V(�������+�ftv�)�D�h+i��%����,r�|X�q�UMy\���P�+[�����ls�5`m��,�5i���/����,I�p��
��
+e�/��;Qq\�����Z[d�E4O�PXqG���Z
�����v���N�#Wg��m��,\R��6��[pQ\0����y��"3�5�+����Z�����$��������'��bRh��X��y�������n��X��**y��
P/@����w���3���������d
x��M����Z�`
��X�@Y���U�Y��

+���mx��W�-�<|�p�KD�^{��I�5j�����J>��UIu�Ef��e��`����zV���GB%�5�����
�q��g%��(K����v�
k�5��
X�����4`�:b�k���w$�d�e��{�����^�6(
X���-+�y4�bx��C,�9��C0i��f������f��(���(�Y	Ls����a��!>��=��������|_���7w�.\���B�,2�����G����K/��
�,�G�Y�U�>i}qWp>=�b�4(�&V5���*���}Z�j��V[������9����HV,FB�U�j_@6Xa�[����.�H�����,�a
�3d����W����4�������,[K���vx�	��z��
�)�[�g�~�W
T��=�<�H�l��6
l���|�6PyB���O4����k��ypw:��V_L���,2��3��a�m������	��l
4p�@�y���!��v�i���[Di��(�V=+�Y�+B��
��{���.��},{M/
Xk�������a�����
+
.U�)�H�/��&q�����
�@
XY��~�n���<*+��!+C��b,\�Gu�<�:�,-�!��w���=��C�[�hYm+3xE@�}�[���
@����
 �/�fx�7P.D�P����r���I��)&3�����c)�R@��=������b,�l���e("+}u�������-
���J�#n_��� �:�����b}�\i�,_�{�jj�yx��K2r9�v��X��L��z^�jN��8qA:���n���N�!�t���#����$@��x�	���i6�"�����"�Y�����Zo���]�'	h�X�����h 
���8j��������o���x�B������AO�J�-$3��^O��$#�����|
��W�k2��#u�&�g~3��C����4��M�������(o���\��i�Z�`����q�F��F��
k#�V��j��q����k��
,�@�������U��05��
kM�����6l�UV�Z�o��d���
`^i<W91���1n��*i���J�m2��M���6��J�t�^�DT,X#�@#h���z��e4`��>�	j
�u�P�4�
��Cy-T����:�PS
Xi����O��K��
�,Zj�<�-����vk����*l	�-���i�Z\G��4`�j��,0`����c���lMl�X����``���\����UNw�����N��s�~$G�9�V��'�q�N�>����w/^Y�s�a�N��:����o��h
����[��$V%�����A3�1��������^rr����5)Mg!�oNY�s��]��I,�,�*�Gh?�Q�zJT�^��1�S��o��>*XQ=4`� �7Z�2���\���G�z�,�8,��b!����X�����/oTU���s-2����Ek 2y���}��g��%����s����v��
����<���O>�7�`���}�&"l��;+����?�9�<i��R�(�zH������_;�����&rH|������3�<*�����#�Z��V��=�*C1`%z�X^,V���^xan-f.���B�����/����`���b:���.���(@@HK���
���bD�Y�4�S\ig����|���}T�oG?3`��h��
X���.e!`=����j���20@���@����p`%�	@PX7�x���k�p�,a�~���x��\
�H<m/D�t���;�������l��X	DN�"���C6-x{���~Y����|n����o��
X���*i!`��=
O�1�{��sd�7��
��X�X�?�p��g�?���9��C�2�0�����xQ.��=~�x���DeQWw1`��6���Z�9��c���
7�t��M�����w���=PX�y����z����>��c��J�u��@2G���o>�3�<S��=I�(5��>��G}�����!Z=�y��DZ�~��P�<�@&�:�S�o�Zk�1�3`m�~�������
�u�J�,9��D	^N�rY��	�M�n����}-Kpp�|2x�5��TS�"��C=�y��"S�����!H:qRW\q��9��M���C�!�!N�t�K��=z(O	��qV
4(+�C0�P����!�:�8m���+|�>}��8q��:t��7eq��5_\���~�����/�����8�Gq��)VO(�o"_
��,���k���l�UW�y�����
X��G�S�����KUX�����*��o�Sr�M
��92��5�:@��4���E3X�PV^��8�B���~��
 e!�����S@ w�g�-����r�B%x
�q
��3��X���~�������mM;���4
�f�q�`��b���+�����
$���#�������>����G2`��^�?�X��O:D�B�ZL�� ���������&L���Ic�����%��Hg� a��j����7h+�"n�<@'iu)�e;�4h��w+��<+�}��&V-��bZ/@a������J��v-����
���SO�_B��i����&���#^��#FD�N=�T��%��e���P�u�%#P%�5�2h�7.���k \����
�$�������W�A;�Q�{���
�z9��Api7w9|����;�xV��q��Go�n��E+��t�n�g�1z�5>w=L\�
k�V��d��_]��i���/���C��m�Z��R2��_�t�D�V�!����aX��,�R�o��������Yy���x]t�E�����/X)M{�L�Rb��f����fO1����[��S���J
k�z�0+��w�����'��|p��
k��L}�e�Z_��a�TX�����;�M{8@��4N;�������v����%+�R��K���E�r�Q�}������s���r)+�����be.P����
+�n�Q����xgV����������hG��|�����
X����(o���\�����(�A 
�V���Z��e'��)(0�2/���J
V������_�W�s��������0'���1�Q����h?'H�	�'=�
f��t�b��w�}����x
��A���}������s�f��Y�W������6!��%����{��d2`��>����k����,T��N��� $�b����~�P)�E$�PM?����6q��+��0@�~N@g����>>i,F
$gk�h�V�K/��?���|�m�=��l)��_�oN�B_]�t��\s��-�0	�8��.�lN����_����M.s�8�%��Y(FY<�i���f�m�-����������Rv����d��
$���#dBV��!�aQo��5�c�Zc�7hu�
�q��`e�
�"�dZh!�{�������<4�����EK�m�]2Y����`
��>��}�����b�/���/�7W����4o�n�4������3���*���q�:���
NG�R|��s�����~� /��e��K,����p�
����������}�,f�,0}�$�s��E>\��<�U���
��&��`e^x�UV��@_t

~R�Z���Z��Y�E������B�m����X^N
pF��d�����NV�����+�A���:t�;w����W#��;�|��X���6���J� C����AQ"�%��;��ISNmEi�^d�'+OY���������E;$�E���7\��
�e��Z�Zi��i�c4`��1zo�Z
X���$�
k�kl�J�M��Uk�
k�T�X�
X��L����
k����j5`m�����UR��m*
�6UwV�1�USmc16`m��2i;F���F�����z�J��VI����4`��T�Y���VM�����������
�v���V�F��*�k�Z%�����
kSug�c�Z5�6c���/��c4`��1zo�Z
X���$o>`��o��?���W'����'�\����i��85J�utrf��H=N����T��+�A�3`m���������;���L��u�o������*]��3
�j@B�99c�IH=w�
��O?=5o���J�
^OcqlRWZi���K~�^��"U�y�����/cR��&N����#F���
��C�+��&fek��UJ�|+�% ����{�,V	j���Kx�J�n|��@����7����v[w�!���������1����lVn����%�++�����
������Jmj!�
�;	��$&�[x��5���������s�9N����o�y�q��J�����n��I$a��b�-����(�V[mkD�|uF�"M.�������GuD��	��$@���in������{My	��p�/��R�6�2�,m�N�2�=��s�l��6r��:k�y����^r����c�_w�u��x��|�����'�p}��qD���^{m'����$����wp���:	]�$�F�!�O!*��R��|��=���U��G����5j�;���Q�g�5Qs�W����5�
�����)�r��I0p��g���95hz
&��_?��a������������f~�-���2%�w�z�Ai�p��$��]w��#z�����o�bmE��3d*����pQ����� �?����&�7
���sN?f�m+A�	6��?D�%T��i���j���e
�_���,���zjIy��
�^�zi�Y������Pv^^x4����N�qb=����h����u�������Kj���c�zx�<g$c��b�/E�����z����T=���?j�k��k�f�����!+�@�
�m���28��L/���.i�^{�
s��7*�0�C�Xmh���X�Zf��7���g��v�t�ad�/����e��'�x�,u���7�x���.]��������'�����y��Q�����}�q��i9��������X��D�v@Q,V}� �X��DG��z���E�0N�����!X���������KX�GD�d���l������R���������hG��u��
kGj�q�6`m�������U\�9����~:�a}A�n�+����0aB�	��a��y����[�}�#�K�%����
&���;�</.�p���,ramR�d��.�L��Vyx�w�}w��;�[���
���#��}r�
�j�,y���j�s�q�E��yGM�n���
_�����t���X���Ht/.w����/������)�~�e���~�e��8]x��*�
k\+v]�0`����|��u���9�`	2X�����K��M����L�4I��
�w��9��Ye�U��9��>/��S��u�:���nO�9��S�7���=��3J+t!s�~���S��r��,X��4��g��!*^��~�9�������D���	i����|��e>����q6���O�,�����O8���zn��&M��]dk=�B��`�Z�}T	KVY�����O��,2)(k��N!��X�
���l���
���g��`��]=V���,y��e�'_>��{f�T&�K/�Te�
<����e]����il�
�?���*���o_|����xP���������[��"�S����?��k���z��i��x}V�K�W^yEBY
���
`�\]�n����,\J|��X�B���ba������`�)��N}��\Y����c%bi��k(	���3N��d����X�<����U2�����_���XY�w�}���dep���>K���E�|���g��.Xs�a7
����P���(�IEND�B`�

From b6cbf986e1f0b32009ed060ad52a145a01e999d0 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v6 1/4] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 ++++++----
 src/backend/utils/adt/pg_locale.c             | 83 ++++++++++++++-----
 .../regress/expected/collate.icu.utf8.out     |  6 ++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 +
 5 files changed, 127 insertions(+), 51 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 2969a2bb21..dd6cd2682f 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (strcmp(colliculocale, "C") == 0 ||
+				strcmp(colliculocale, "POSIX") == 0)
 			{
-				char	   *langtag = icu_language_tag(colliculocale,
-													   icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char	   *langtag = icu_language_tag(colliculocale,
+														   icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 99d4080ea9..bfce8dc348 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (strcmp(dbiculocale, "C") == 0 ||
+			strcmp(dbiculocale, "POSIX") == 0)
 		{
-			char	   *langtag = icu_language_tag(dbiculocale,
-												   icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char	   *langtag = icu_language_tag(dbiculocale,
+													   icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 31e3b16ae0..986dcbd2a7 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,13 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (strcmp(iculocstr, "C") == 0 || strcmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1685,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (strcmp("C", collcollate) == 0 || strcmp("POSIX", collcollate) == 0)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1706,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2493,13 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (strcmp(loc_str, "C") == 0 || strcmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index c658ee1404..bfc28ecfcf 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1043,12 +1043,18 @@ ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 7bd0901281..572dc5a50a 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,9 +379,13 @@ CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, nee
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
-- 
2.34.1

From 05654d18cbc4156fcd0c6a0ec1267c60a36294db Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v6 2/4] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 +++--
 doc/src/sgml/ref/createdb.sgml                |  5 +++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++---
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++++----
 src/bin/initdb/initdb.c                       | 11 ++++++---
 src/bin/scripts/createdb.c                    |  5 ++++
 src/bin/scripts/t/020_createdb.pl             |  4 ++--
 src/test/icu/t/010_database.pl                | 23 ++++++++++++-------
 .../regress/expected/collate.icu.utf8.out     | 22 +++++++++---------
 10 files changed, 65 insertions(+), 35 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index dd6cd2682f..c165922121 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index bfce8dc348..a478a2287f 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 31156e863b..fc27804413 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2163,7 +2163,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2405,7 +2409,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2421,6 +2425,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = pg_strdup(locale);
 	}
 
 	/*
@@ -3296,7 +3302,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..58e98ebb9c 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -219,6 +219,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index d0830a4a1d..c3cd674440 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -137,7 +137,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -145,7 +145,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index d3901f5d3f..ea2be008af 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,17 +51,24 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
 );
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index bfc28ecfcf..fdf07db36e 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1200,9 +1200,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1210,7 +1210,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1227,12 +1227,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1241,7 +1241,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1269,13 +1269,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1325,9 +1325,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 80f04077dc9224ddad479d373d82c2b1f5a1fc41 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 12:15:33 -0700
Subject: [PATCH v6 3/4] pg_upgrade: check for ICU locale C in versions 15 and
 earlier.

ICU collations with locale "C" (and equivalently "POSIX") were not
supported before version 16, but could still be created. Users with
such collations would not get expected behavior.

Reject upgrading clusters with such collations, so that we can support
"C" and "POSIX" locales for the ICU provider without risk of
corrupting indexes during pg_upgrade.
---
 src/bin/pg_upgrade/check.c | 97 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 97 insertions(+)

diff --git a/src/bin/pg_upgrade/check.c b/src/bin/pg_upgrade/check.c
index 64024e3b9e..17c75cf186 100644
--- a/src/bin/pg_upgrade/check.c
+++ b/src/bin/pg_upgrade/check.c
@@ -26,6 +26,7 @@ static void check_for_tables_with_oids(ClusterInfo *cluster);
 static void check_for_composite_data_type_usage(ClusterInfo *cluster);
 static void check_for_reg_data_type_usage(ClusterInfo *cluster);
 static void check_for_aclitem_data_type_usage(ClusterInfo *cluster);
+static void check_icu_c_before_16(ClusterInfo *cluster);
 static void check_for_jsonb_9_4_usage(ClusterInfo *cluster);
 static void check_for_pg_role_prefix(ClusterInfo *cluster);
 static void check_for_new_tablespace_dir(ClusterInfo *new_cluster);
@@ -164,6 +165,9 @@ check_and_dump_old_cluster(bool live_check)
 	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 905)
 		check_for_pg_role_prefix(&old_cluster);
 
+	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 1500)
+		check_icu_c_before_16(&old_cluster);
+
 	if (GET_MAJOR_VERSION(old_cluster.major_version) == 904 &&
 		old_cluster.controldata.cat_ver < JSONB_FORMAT_CHANGE_CAT_VER)
 		check_for_jsonb_9_4_usage(&old_cluster);
@@ -1233,6 +1237,99 @@ check_for_aclitem_data_type_usage(ClusterInfo *cluster)
 		check_ok();
 }
 
+/*
+ * check_icu_c_before_16
+ *
+ *  Version 16 adds support for the ICU C locale, but it was possible to
+ *  (incorrectly) create it in prior versions. Check for this invalid ICU
+ *  locale name in version 15 and earlier.
+ */
+static void
+check_icu_c_before_16(ClusterInfo *cluster)
+{
+	PGresult   *dat_res;
+	PGconn	   *template1_conn = connectToServer(cluster, "template1");
+	int			dat_ntups;
+	int			i_datname;
+	FILE	   *script = NULL;
+	char		output_path[MAXPGPATH];
+
+	prep_status("Checking for ICU collations with locale \"C\"");
+
+	snprintf(output_path, sizeof(output_path), "%s/%s",
+			 log_opts.basedir,
+			 "icu_c_before_16.txt");
+
+	/* check pg_database */
+	dat_res = executeQueryOrDie(template1_conn,
+								"SELECT datname "
+								"FROM pg_catalog.pg_database "
+								"WHERE datlocprovider='i' "
+								"AND daticulocale IN ('C','POSIX')");
+
+	i_datname = PQfnumber(dat_res, "datname");
+
+	dat_ntups = PQntuples(dat_res);
+
+	for (int rowno = 0; rowno < dat_ntups; rowno++)
+	{
+		if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+			pg_fatal("could not open file \"%s\": %s",
+					 output_path, strerror(errno));
+		fprintf(script, "default collation for database %s\n",
+				PQgetvalue(dat_res, rowno, i_datname));
+	}
+
+	PQclear(dat_res);
+	PQfinish(template1_conn);
+
+	/* check pg_collation in each database */
+	for (int dbnum = 0; dbnum < cluster->dbarr.ndbs; dbnum++)
+	{
+		DbInfo	   *active_db = &cluster->dbarr.dbs[dbnum];
+		PGconn	   *db_conn = connectToServer(cluster, active_db->db_name);
+		PGresult   *coll_res;
+		int			coll_ntups;
+		int			i_collid;
+		int			i_collname;
+
+		coll_res = executeQueryOrDie(db_conn,
+									 "SELECT oid as collid, collname "
+									 "FROM pg_catalog.pg_collation "
+									 "WHERE collprovider='i' "
+									 "AND colliculocale IN ('C','POSIX')");
+
+		i_collid = PQfnumber(coll_res, "collid");
+		i_collname = PQfnumber(coll_res, "collname");
+
+		coll_ntups = PQntuples(coll_res);
+
+		for (int rowno = 0; rowno < coll_ntups; rowno++)
+		{
+			if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+				pg_fatal("could not open file \"%s\": %s",
+						 output_path, strerror(errno));
+			fprintf(script, "database %s collation %s (oid=%s)\n",
+					active_db->db_name,
+					PQgetvalue(coll_res, rowno, i_collname),
+					PQgetvalue(coll_res, rowno, i_collid));
+		}
+
+		PQclear(coll_res);
+		PQfinish(db_conn);
+	}
+
+	if (script)
+	{
+		pg_log(PG_REPORT, "fatal");
+		pg_fatal("Your installation contains ICU collations with the locale \"C\" or \"POSIX\",\n"
+				 "which are not supported until version 16. Earlier versions using ICU \n"
+				 "collations with the \"C\" or \"POSIX\" locales cannot be upgraded.");
+	}
+
+	check_ok();
+}
+
 /*
  * check_for_jsonb_9_4_usage()
  *
-- 
2.34.1

From 184644da55144df9295d81ac3b4cd061473d98ce Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 09:53:02 -0700
Subject: [PATCH v6 4/4] Use database default collation's provider as default
 for CREATE COLLATION.

---
 doc/src/sgml/ref/create_collation.sgml           |  9 ++++++---
 src/backend/commands/collationcmds.c             |  7 ++++++-
 src/test/regress/expected/collate.linux.utf8.out | 10 +++++-----
 src/test/regress/sql/collate.linux.utf8.sql      | 10 +++++-----
 4 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..a6927a7d1d 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -121,9 +121,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
       <para>
        Specifies the provider to use for locale services associated with this
        collation.  Possible values are
-       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
-       (if the server was built with ICU support) or <literal>libc</literal>.
-       <literal>libc</literal> is the default.  See <xref
+       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
+       the server was built with ICU support) or <literal>libc</literal>.  If
+       <replaceable>lc_colllate</replaceable> or
+       <replaceable>lc_ctype</replaceable> is specified, the default is
+       <literal>libc</literal>; otherwise, the default is the same as the
+       database default collation's provider.  See <xref
        linkend="locale-providers"/> for details.
       </para>
      </listitem>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c165922121..8fc0ff1903 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -226,7 +226,12 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+		{
+			if (lccollateEl || lcctypeEl)
+				collprovider = COLLPROVIDER_LIBC;
+			else
+				collprovider = default_locale.provider;
+		}
 
 		if (localeEl)
 		{
diff --git a/src/test/regress/expected/collate.linux.utf8.out b/src/test/regress/expected/collate.linux.utf8.out
index 6d34667ceb..6b0cc95ae8 100644
--- a/src/test/regress/expected/collate.linux.utf8.out
+++ b/src/test/regress/expected/collate.linux.utf8.out
@@ -1026,7 +1026,7 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
@@ -1034,7 +1034,7 @@ CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 ERROR:  collation "test0" already exists
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
 NOTICE:  collation "test0" already exists, skipping
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 NOTICE:  collation "test0" for encoding "UTF8" already exists, skipping
 do $$
 BEGIN
@@ -1046,7 +1046,7 @@ END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
 ERROR:  parameter "lc_ctype" must be specified
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 ERROR:  could not create locale "nonsense": No such file or directory
 DETAIL:  The operating system could not find any locale data for the locale name "nonsense".
 CREATE COLLATION test4 FROM nonsense;
@@ -1166,8 +1166,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 
 -- nondeterministic collations
 -- (not supported with libc provider)
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 ERROR:  nondeterministic collations not supported with this provider
 -- cleanup
 SET client_min_messages TO warning;
diff --git a/src/test/regress/sql/collate.linux.utf8.sql b/src/test/regress/sql/collate.linux.utf8.sql
index 2b787507c5..cc25f95ac3 100644
--- a/src/test/regress/sql/collate.linux.utf8.sql
+++ b/src/test/regress/sql/collate.linux.utf8.sql
@@ -358,13 +358,13 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
@@ -374,7 +374,7 @@ BEGIN
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
@@ -455,8 +455,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 -- nondeterministic collations
 -- (not supported with libc provider)
 
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 
 
 -- cleanup
-- 
2.34.1

diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 31156e863b..09a5c98cc0 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2354,42 +2354,13 @@ icu_validate_locale(const char *loc_str)
 }
 
 /*
- * Determine default ICU locale by opening the default collator and reading
- * its locale.
- *
- * NB: The default collator (opened using NULL) is different from the collator
- * for the root locale (opened with "", "und", or "root"). The former depends
- * on the environment (useful at initdb time) and the latter does not.
+ * Determine the default ICU locale
  */
 static char *
 default_icu_locale(void)
 {
 #ifdef USE_ICU
-	UCollator  *collator;
-	UErrorCode	status;
-	const char *valid_locale;
-	char	   *default_locale;
-
-	status = U_ZERO_ERROR;
-	collator = ucol_open(NULL, &status);
-	if (U_FAILURE(status))
-		pg_fatal("could not open collator for default locale: %s",
-				 u_errorName(status));
-
-	status = U_ZERO_ERROR;
-	valid_locale = ucol_getLocaleByType(collator, ULOC_VALID_LOCALE,
-										&status);
-	if (U_FAILURE(status))
-	{
-		ucol_close(collator);
-		pg_fatal("could not determine default ICU locale");
-	}
-
-	default_locale = pg_strdup(valid_locale);
-
-	ucol_close(collator);
-
-	return default_locale;
+	return pg_strdup(uloc_getDefault());
 #else
 	pg_fatal("ICU is not supported in this build");
 #endif

From b542c98403b382ad634ea4915057ee170434fcea Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v8 1/4] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 ++++++----
 src/backend/utils/adt/pg_locale.c             | 83 ++++++++++++++-----
 .../regress/expected/collate.icu.utf8.out     |  6 ++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 +
 5 files changed, 127 insertions(+), 51 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 2969a2bb21..dd6cd2682f 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (strcmp(colliculocale, "C") == 0 ||
+				strcmp(colliculocale, "POSIX") == 0)
 			{
-				char	   *langtag = icu_language_tag(colliculocale,
-													   icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char	   *langtag = icu_language_tag(colliculocale,
+														   icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 99d4080ea9..bfce8dc348 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (strcmp(dbiculocale, "C") == 0 ||
+			strcmp(dbiculocale, "POSIX") == 0)
 		{
-			char	   *langtag = icu_language_tag(dbiculocale,
-												   icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char	   *langtag = icu_language_tag(dbiculocale,
+													   icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 31e3b16ae0..986dcbd2a7 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,13 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (strcmp(iculocstr, "C") == 0 || strcmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1685,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (strcmp("C", collcollate) == 0 || strcmp("POSIX", collcollate) == 0)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1706,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2493,13 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (strcmp(loc_str, "C") == 0 || strcmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index c658ee1404..bfc28ecfcf 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1043,12 +1043,18 @@ ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 7bd0901281..572dc5a50a 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,9 +379,13 @@ CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, nee
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
-- 
2.34.1

From 0a02ac57d65a81080eb34d9ba028e032857d1188 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 12:15:33 -0700
Subject: [PATCH v8 2/4] pg_upgrade: check for ICU locale C in versions 15 and
 earlier.

ICU collations with locale "C" (and equivalently "POSIX") were not
supported before version 16, but could still be created. Users with
such collations would not get expected behavior.

Reject upgrading clusters with such collations, so that we can support
"C" and "POSIX" locales for the ICU provider without risk of
corrupting indexes during pg_upgrade.
---
 src/bin/pg_upgrade/check.c | 97 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 97 insertions(+)

diff --git a/src/bin/pg_upgrade/check.c b/src/bin/pg_upgrade/check.c
index 64024e3b9e..17c75cf186 100644
--- a/src/bin/pg_upgrade/check.c
+++ b/src/bin/pg_upgrade/check.c
@@ -26,6 +26,7 @@ static void check_for_tables_with_oids(ClusterInfo *cluster);
 static void check_for_composite_data_type_usage(ClusterInfo *cluster);
 static void check_for_reg_data_type_usage(ClusterInfo *cluster);
 static void check_for_aclitem_data_type_usage(ClusterInfo *cluster);
+static void check_icu_c_before_16(ClusterInfo *cluster);
 static void check_for_jsonb_9_4_usage(ClusterInfo *cluster);
 static void check_for_pg_role_prefix(ClusterInfo *cluster);
 static void check_for_new_tablespace_dir(ClusterInfo *new_cluster);
@@ -164,6 +165,9 @@ check_and_dump_old_cluster(bool live_check)
 	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 905)
 		check_for_pg_role_prefix(&old_cluster);
 
+	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 1500)
+		check_icu_c_before_16(&old_cluster);
+
 	if (GET_MAJOR_VERSION(old_cluster.major_version) == 904 &&
 		old_cluster.controldata.cat_ver < JSONB_FORMAT_CHANGE_CAT_VER)
 		check_for_jsonb_9_4_usage(&old_cluster);
@@ -1233,6 +1237,99 @@ check_for_aclitem_data_type_usage(ClusterInfo *cluster)
 		check_ok();
 }
 
+/*
+ * check_icu_c_before_16
+ *
+ *  Version 16 adds support for the ICU C locale, but it was possible to
+ *  (incorrectly) create it in prior versions. Check for this invalid ICU
+ *  locale name in version 15 and earlier.
+ */
+static void
+check_icu_c_before_16(ClusterInfo *cluster)
+{
+	PGresult   *dat_res;
+	PGconn	   *template1_conn = connectToServer(cluster, "template1");
+	int			dat_ntups;
+	int			i_datname;
+	FILE	   *script = NULL;
+	char		output_path[MAXPGPATH];
+
+	prep_status("Checking for ICU collations with locale \"C\"");
+
+	snprintf(output_path, sizeof(output_path), "%s/%s",
+			 log_opts.basedir,
+			 "icu_c_before_16.txt");
+
+	/* check pg_database */
+	dat_res = executeQueryOrDie(template1_conn,
+								"SELECT datname "
+								"FROM pg_catalog.pg_database "
+								"WHERE datlocprovider='i' "
+								"AND daticulocale IN ('C','POSIX')");
+
+	i_datname = PQfnumber(dat_res, "datname");
+
+	dat_ntups = PQntuples(dat_res);
+
+	for (int rowno = 0; rowno < dat_ntups; rowno++)
+	{
+		if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+			pg_fatal("could not open file \"%s\": %s",
+					 output_path, strerror(errno));
+		fprintf(script, "default collation for database %s\n",
+				PQgetvalue(dat_res, rowno, i_datname));
+	}
+
+	PQclear(dat_res);
+	PQfinish(template1_conn);
+
+	/* check pg_collation in each database */
+	for (int dbnum = 0; dbnum < cluster->dbarr.ndbs; dbnum++)
+	{
+		DbInfo	   *active_db = &cluster->dbarr.dbs[dbnum];
+		PGconn	   *db_conn = connectToServer(cluster, active_db->db_name);
+		PGresult   *coll_res;
+		int			coll_ntups;
+		int			i_collid;
+		int			i_collname;
+
+		coll_res = executeQueryOrDie(db_conn,
+									 "SELECT oid as collid, collname "
+									 "FROM pg_catalog.pg_collation "
+									 "WHERE collprovider='i' "
+									 "AND colliculocale IN ('C','POSIX')");
+
+		i_collid = PQfnumber(coll_res, "collid");
+		i_collname = PQfnumber(coll_res, "collname");
+
+		coll_ntups = PQntuples(coll_res);
+
+		for (int rowno = 0; rowno < coll_ntups; rowno++)
+		{
+			if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+				pg_fatal("could not open file \"%s\": %s",
+						 output_path, strerror(errno));
+			fprintf(script, "database %s collation %s (oid=%s)\n",
+					active_db->db_name,
+					PQgetvalue(coll_res, rowno, i_collname),
+					PQgetvalue(coll_res, rowno, i_collid));
+		}
+
+		PQclear(coll_res);
+		PQfinish(db_conn);
+	}
+
+	if (script)
+	{
+		pg_log(PG_REPORT, "fatal");
+		pg_fatal("Your installation contains ICU collations with the locale \"C\" or \"POSIX\",\n"
+				 "which are not supported until version 16. Earlier versions using ICU \n"
+				 "collations with the \"C\" or \"POSIX\" locales cannot be upgraded.");
+	}
+
+	check_ok();
+}
+
 /*
  * check_for_jsonb_9_4_usage()
  *
-- 
2.34.1

From 4959605b5695e03491fc355aa3e13b522103f988 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v8 3/4] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 +++--
 doc/src/sgml/ref/createdb.sgml                |  5 +++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++---
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++++----
 src/bin/initdb/initdb.c                       | 11 ++++++---
 src/bin/scripts/createdb.c                    |  5 ++++
 src/bin/scripts/t/020_createdb.pl             |  4 ++--
 src/test/icu/t/010_database.pl                | 23 ++++++++++++-------
 .../regress/expected/collate.icu.utf8.out     | 22 +++++++++---------
 10 files changed, 65 insertions(+), 35 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index dd6cd2682f..c165922121 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index bfce8dc348..a478a2287f 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 09a5c98cc0..c7696dda10 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2163,7 +2163,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2376,7 +2380,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2392,6 +2396,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = pg_strdup(locale);
 	}
 
 	/*
@@ -3267,7 +3273,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..58e98ebb9c 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -219,6 +219,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index d0830a4a1d..c3cd674440 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -137,7 +137,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -145,7 +145,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index d3901f5d3f..ea2be008af 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,17 +51,24 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
 );
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index bfc28ecfcf..fdf07db36e 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1200,9 +1200,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1210,7 +1210,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1227,12 +1227,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1241,7 +1241,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1269,13 +1269,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1325,9 +1325,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From a02b6d160ddbc4db584b13c842b8ed82644baf9f Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 09:53:02 -0700
Subject: [PATCH v8 4/4] Use database default collation's provider as default
 for CREATE COLLATION.

---
 doc/src/sgml/ref/create_collation.sgml           |  9 ++++++---
 src/backend/commands/collationcmds.c             |  7 ++++++-
 src/test/regress/expected/collate.linux.utf8.out | 10 +++++-----
 src/test/regress/sql/collate.linux.utf8.sql      | 10 +++++-----
 4 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..a6927a7d1d 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -121,9 +121,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
       <para>
        Specifies the provider to use for locale services associated with this
        collation.  Possible values are
-       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
-       (if the server was built with ICU support) or <literal>libc</literal>.
-       <literal>libc</literal> is the default.  See <xref
+       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
+       the server was built with ICU support) or <literal>libc</literal>.  If
+       <replaceable>lc_colllate</replaceable> or
+       <replaceable>lc_ctype</replaceable> is specified, the default is
+       <literal>libc</literal>; otherwise, the default is the same as the
+       database default collation's provider.  See <xref
        linkend="locale-providers"/> for details.
       </para>
      </listitem>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c165922121..8fc0ff1903 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -226,7 +226,12 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+		{
+			if (lccollateEl || lcctypeEl)
+				collprovider = COLLPROVIDER_LIBC;
+			else
+				collprovider = default_locale.provider;
+		}
 
 		if (localeEl)
 		{
diff --git a/src/test/regress/expected/collate.linux.utf8.out b/src/test/regress/expected/collate.linux.utf8.out
index 6d34667ceb..6b0cc95ae8 100644
--- a/src/test/regress/expected/collate.linux.utf8.out
+++ b/src/test/regress/expected/collate.linux.utf8.out
@@ -1026,7 +1026,7 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
@@ -1034,7 +1034,7 @@ CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 ERROR:  collation "test0" already exists
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
 NOTICE:  collation "test0" already exists, skipping
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 NOTICE:  collation "test0" for encoding "UTF8" already exists, skipping
 do $$
 BEGIN
@@ -1046,7 +1046,7 @@ END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
 ERROR:  parameter "lc_ctype" must be specified
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 ERROR:  could not create locale "nonsense": No such file or directory
 DETAIL:  The operating system could not find any locale data for the locale name "nonsense".
 CREATE COLLATION test4 FROM nonsense;
@@ -1166,8 +1166,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 
 -- nondeterministic collations
 -- (not supported with libc provider)
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 ERROR:  nondeterministic collations not supported with this provider
 -- cleanup
 SET client_min_messages TO warning;
diff --git a/src/test/regress/sql/collate.linux.utf8.sql b/src/test/regress/sql/collate.linux.utf8.sql
index 2b787507c5..cc25f95ac3 100644
--- a/src/test/regress/sql/collate.linux.utf8.sql
+++ b/src/test/regress/sql/collate.linux.utf8.sql
@@ -358,13 +358,13 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
@@ -374,7 +374,7 @@ BEGIN
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
@@ -455,8 +455,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 -- nondeterministic collations
 -- (not supported with libc provider)
 
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 
 
 -- cleanup
-- 
2.34.1

From 879f68b254f2e3f531b0c92c08bebf20298376bc Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 12:15:33 -0700
Subject: [PATCH v9 2/4] pg_upgrade: check for ICU locale C in versions 15 and
 earlier.

ICU collations with locale "C" (and equivalently "POSIX") were not
supported before version 16, but could still be created. Users with
such collations would not get expected behavior.

Reject upgrading clusters with such collations, so that we can support
"C" and "POSIX" locales for the ICU provider without risk of
corrupting indexes during pg_upgrade.
---
 src/bin/pg_upgrade/check.c | 97 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 97 insertions(+)

diff --git a/src/bin/pg_upgrade/check.c b/src/bin/pg_upgrade/check.c
index 64024e3b9e..17c75cf186 100644
--- a/src/bin/pg_upgrade/check.c
+++ b/src/bin/pg_upgrade/check.c
@@ -26,6 +26,7 @@ static void check_for_tables_with_oids(ClusterInfo *cluster);
 static void check_for_composite_data_type_usage(ClusterInfo *cluster);
 static void check_for_reg_data_type_usage(ClusterInfo *cluster);
 static void check_for_aclitem_data_type_usage(ClusterInfo *cluster);
+static void check_icu_c_before_16(ClusterInfo *cluster);
 static void check_for_jsonb_9_4_usage(ClusterInfo *cluster);
 static void check_for_pg_role_prefix(ClusterInfo *cluster);
 static void check_for_new_tablespace_dir(ClusterInfo *new_cluster);
@@ -164,6 +165,9 @@ check_and_dump_old_cluster(bool live_check)
 	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 905)
 		check_for_pg_role_prefix(&old_cluster);
 
+	if (GET_MAJOR_VERSION(old_cluster.major_version) <= 1500)
+		check_icu_c_before_16(&old_cluster);
+
 	if (GET_MAJOR_VERSION(old_cluster.major_version) == 904 &&
 		old_cluster.controldata.cat_ver < JSONB_FORMAT_CHANGE_CAT_VER)
 		check_for_jsonb_9_4_usage(&old_cluster);
@@ -1233,6 +1237,99 @@ check_for_aclitem_data_type_usage(ClusterInfo *cluster)
 		check_ok();
 }
 
+/*
+ * check_icu_c_before_16
+ *
+ *  Version 16 adds support for the ICU C locale, but it was possible to
+ *  (incorrectly) create it in prior versions. Check for this invalid ICU
+ *  locale name in version 15 and earlier.
+ */
+static void
+check_icu_c_before_16(ClusterInfo *cluster)
+{
+	PGresult   *dat_res;
+	PGconn	   *template1_conn = connectToServer(cluster, "template1");
+	int			dat_ntups;
+	int			i_datname;
+	FILE	   *script = NULL;
+	char		output_path[MAXPGPATH];
+
+	prep_status("Checking for ICU collations with locale \"C\"");
+
+	snprintf(output_path, sizeof(output_path), "%s/%s",
+			 log_opts.basedir,
+			 "icu_c_before_16.txt");
+
+	/* check pg_database */
+	dat_res = executeQueryOrDie(template1_conn,
+								"SELECT datname "
+								"FROM pg_catalog.pg_database "
+								"WHERE datlocprovider='i' "
+								"AND daticulocale IN ('C','POSIX')");
+
+	i_datname = PQfnumber(dat_res, "datname");
+
+	dat_ntups = PQntuples(dat_res);
+
+	for (int rowno = 0; rowno < dat_ntups; rowno++)
+	{
+		if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+			pg_fatal("could not open file \"%s\": %s",
+					 output_path, strerror(errno));
+		fprintf(script, "default collation for database %s\n",
+				PQgetvalue(dat_res, rowno, i_datname));
+	}
+
+	PQclear(dat_res);
+	PQfinish(template1_conn);
+
+	/* check pg_collation in each database */
+	for (int dbnum = 0; dbnum < cluster->dbarr.ndbs; dbnum++)
+	{
+		DbInfo	   *active_db = &cluster->dbarr.dbs[dbnum];
+		PGconn	   *db_conn = connectToServer(cluster, active_db->db_name);
+		PGresult   *coll_res;
+		int			coll_ntups;
+		int			i_collid;
+		int			i_collname;
+
+		coll_res = executeQueryOrDie(db_conn,
+									 "SELECT oid as collid, collname "
+									 "FROM pg_catalog.pg_collation "
+									 "WHERE collprovider='i' "
+									 "AND colliculocale IN ('C','POSIX')");
+
+		i_collid = PQfnumber(coll_res, "collid");
+		i_collname = PQfnumber(coll_res, "collname");
+
+		coll_ntups = PQntuples(coll_res);
+
+		for (int rowno = 0; rowno < coll_ntups; rowno++)
+		{
+			if (script == NULL && (script = fopen_priv(output_path, "w")) == NULL)
+				pg_fatal("could not open file \"%s\": %s",
+						 output_path, strerror(errno));
+			fprintf(script, "database %s collation %s (oid=%s)\n",
+					active_db->db_name,
+					PQgetvalue(coll_res, rowno, i_collname),
+					PQgetvalue(coll_res, rowno, i_collid));
+		}
+
+		PQclear(coll_res);
+		PQfinish(db_conn);
+	}
+
+	if (script)
+	{
+		pg_log(PG_REPORT, "fatal");
+		pg_fatal("Your installation contains ICU collations with the locale \"C\" or \"POSIX\",\n"
+				 "which are not supported until version 16. Earlier versions using ICU \n"
+				 "collations with the \"C\" or \"POSIX\" locales cannot be upgraded.");
+	}
+
+	check_ok();
+}
+
 /*
  * check_for_jsonb_9_4_usage()
  *
-- 
2.34.1

From 361b81f1ea5153745b51d8215b3dcaeac4fdeedc Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v9 3/4] Make LOCALE apply to ICU_LOCALE for CREATE DATABASE.

LOCALE is now an alias for LC_COLLATE, LC_CTYPE, and (if the provider
is ICU) ICU_LOCALE. The ICU provider accepts more locale names than
libc (e.g. language tags and locale names containing collation
attributes), so in some cases LC_COLLATE, LC_CTYPE, and ICU_LOCALE
will still need to be specified separately.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE (and
similarly for --locale in initdb and createdb). That could lead to
confusion when the provider is implicit, such as when it is inherited
from the template database, or when ICU was made default at initdb
time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_database.sgml         |  6 +++--
 doc/src/sgml/ref/createdb.sgml                |  5 +++-
 doc/src/sgml/ref/initdb.sgml                  |  7 +++---
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 15 ++++++++----
 src/bin/initdb/initdb.c                       | 11 ++++++---
 src/bin/scripts/createdb.c                    |  5 ++++
 src/bin/scripts/t/020_createdb.pl             |  4 ++--
 src/test/icu/t/010_database.pl                | 23 ++++++++++++-------
 .../regress/expected/collate.icu.utf8.out     | 22 +++++++++---------
 10 files changed, 65 insertions(+), 35 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..844773ff44 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,10 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        This is a shortcut for setting <symbol>LC_COLLATE</symbol>,
+        <symbol>LC_CTYPE</symbol> and <symbol>ICU_LOCALE</symbol> at
+        once. Some locales are only valid for ICU, and must be set separately
+        with <symbol>ICU_LOCALE</symbol>.
        </para>
        <tip>
         <para>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index dd6cd2682f..c165922121 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -288,7 +288,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					if (langtag && strcmp(colliculocale, langtag) != 0)
 					{
 						ereport(NOTICE,
-								(errmsg("using standard form \"%s\" for locale \"%s\"",
+								(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 										langtag, colliculocale)));
 
 						colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index bfce8dc348..a478a2287f 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1080,7 +1087,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 				if (langtag && strcmp(dbiculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, dbiculocale)));
 
 					dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 7aa2d871e3..93b19b952b 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2163,7 +2163,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2376,7 +2380,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2392,6 +2396,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = pg_strdup(locale);
 	}
 
 	/*
@@ -3276,7 +3282,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..58e98ebb9c 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -219,6 +219,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index d0830a4a1d..c3cd674440 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -137,7 +137,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -145,7 +145,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index d3901f5d3f..ea2be008af 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,17 +51,24 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
 );
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
-
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index bfc28ecfcf..fdf07db36e 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1200,9 +1200,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1210,7 +1210,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1227,12 +1227,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1241,7 +1241,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1269,13 +1269,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1325,9 +1325,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 6a1b9ca2a247f3e5111988292e148ba86d802005 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 09:53:02 -0700
Subject: [PATCH v9 4/4] Use database default collation's provider as default
 for CREATE COLLATION.

---
 doc/src/sgml/ref/create_collation.sgml           |  9 ++++++---
 src/backend/commands/collationcmds.c             |  7 ++++++-
 src/test/regress/expected/collate.linux.utf8.out | 10 +++++-----
 src/test/regress/sql/collate.linux.utf8.sql      | 10 +++++-----
 4 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..a6927a7d1d 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -121,9 +121,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
       <para>
        Specifies the provider to use for locale services associated with this
        collation.  Possible values are
-       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
-       (if the server was built with ICU support) or <literal>libc</literal>.
-       <literal>libc</literal> is the default.  See <xref
+       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
+       the server was built with ICU support) or <literal>libc</literal>.  If
+       <replaceable>lc_colllate</replaceable> or
+       <replaceable>lc_ctype</replaceable> is specified, the default is
+       <literal>libc</literal>; otherwise, the default is the same as the
+       database default collation's provider.  See <xref
        linkend="locale-providers"/> for details.
       </para>
      </listitem>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index c165922121..8fc0ff1903 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -226,7 +226,12 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+		{
+			if (lccollateEl || lcctypeEl)
+				collprovider = COLLPROVIDER_LIBC;
+			else
+				collprovider = default_locale.provider;
+		}
 
 		if (localeEl)
 		{
diff --git a/src/test/regress/expected/collate.linux.utf8.out b/src/test/regress/expected/collate.linux.utf8.out
index 6d34667ceb..6b0cc95ae8 100644
--- a/src/test/regress/expected/collate.linux.utf8.out
+++ b/src/test/regress/expected/collate.linux.utf8.out
@@ -1026,7 +1026,7 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
@@ -1034,7 +1034,7 @@ CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 ERROR:  collation "test0" already exists
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
 NOTICE:  collation "test0" already exists, skipping
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 NOTICE:  collation "test0" for encoding "UTF8" already exists, skipping
 do $$
 BEGIN
@@ -1046,7 +1046,7 @@ END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
 ERROR:  parameter "lc_ctype" must be specified
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 ERROR:  could not create locale "nonsense": No such file or directory
 DETAIL:  The operating system could not find any locale data for the locale name "nonsense".
 CREATE COLLATION test4 FROM nonsense;
@@ -1166,8 +1166,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 
 -- nondeterministic collations
 -- (not supported with libc provider)
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 ERROR:  nondeterministic collations not supported with this provider
 -- cleanup
 SET client_min_messages TO warning;
diff --git a/src/test/regress/sql/collate.linux.utf8.sql b/src/test/regress/sql/collate.linux.utf8.sql
index 2b787507c5..cc25f95ac3 100644
--- a/src/test/regress/sql/collate.linux.utf8.sql
+++ b/src/test/regress/sql/collate.linux.utf8.sql
@@ -358,13 +358,13 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal(current_setting('lc_collate')) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
@@ -374,7 +374,7 @@ BEGIN
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
@@ -455,8 +455,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 -- nondeterministic collations
 -- (not supported with libc provider)
 
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 
 
 -- cleanup
-- 
2.34.1

From 54a171f643504d0f8beb2a1f8f7e29dc9639654e Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 24 Apr 2023 15:46:17 -0700
Subject: [PATCH v9 1/4] ICU: support locale "C" with the same behavior as
 libc.

The "C" locale doesn't actually use a provider at all, it's a special
locale that uses memcmp() and built-in character classification. Make
it behave the same in ICU as libc (even though it doesn't actually
make use of either provider).

Discussion: https://postgr.es/m/87v8hoexdv.fsf@news-spur.riddles.org.uk
---
 src/backend/commands/collationcmds.c          | 43 ++++++----
 src/backend/commands/dbcommands.c             | 42 ++++++----
 src/backend/utils/adt/pg_locale.c             | 83 ++++++++++++++-----
 src/bin/initdb/initdb.c                       | 25 ++++--
 .../regress/expected/collate.icu.utf8.out     |  6 ++
 src/test/regress/sql/collate.icu.utf8.sql     |  4 +
 6 files changed, 144 insertions(+), 59 deletions(-)

diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 2969a2bb21..dd6cd2682f 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -264,26 +264,39 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 						 errmsg("parameter \"locale\" must be specified")));
 
-			/*
-			 * During binary upgrade, preserve the locale string. Otherwise,
-			 * canonicalize to a language tag.
-			 */
-			if (!IsBinaryUpgrade)
+			if (strcmp(colliculocale, "C") == 0 ||
+				strcmp(colliculocale, "POSIX") == 0)
 			{
-				char	   *langtag = icu_language_tag(colliculocale,
-													   icu_validation_level);
-
-				if (langtag && strcmp(colliculocale, langtag) != 0)
+				if (!collisdeterministic)
+					ereport(ERROR,
+							(errmsg("nondeterministic collations not supported for C or POSIX locale")));
+				if (collicurules != NULL)
+					ereport(ERROR,
+							(errmsg("RULES not supported for C or POSIX locale")));
+			}
+			else
+			{
+				/*
+				 * During binary upgrade, preserve the locale
+				 * string. Otherwise, canonicalize to a language tag.
+				 */
+				if (!IsBinaryUpgrade)
 				{
-					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
-									langtag, colliculocale)));
+					char	   *langtag = icu_language_tag(colliculocale,
+														   icu_validation_level);
+
+					if (langtag && strcmp(colliculocale, langtag) != 0)
+					{
+						ereport(NOTICE,
+								(errmsg("using standard form \"%s\" for locale \"%s\"",
+										langtag, colliculocale)));
 
-					colliculocale = langtag;
+						colliculocale = langtag;
+					}
 				}
-			}
 
-			icu_validate_locale(colliculocale);
+				icu_validate_locale(colliculocale);
+			}
 		}
 
 		/*
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 99d4080ea9..bfce8dc348 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1058,27 +1058,37 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 					 errmsg("ICU locale must be specified")));
 
-		/*
-		 * During binary upgrade, or when the locale came from the template
-		 * database, preserve locale string. Otherwise, canonicalize to a
-		 * language tag.
-		 */
-		if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
+		if (strcmp(dbiculocale, "C") == 0 ||
+			strcmp(dbiculocale, "POSIX") == 0)
 		{
-			char	   *langtag = icu_language_tag(dbiculocale,
-												   icu_validation_level);
-
-			if (langtag && strcmp(dbiculocale, langtag) != 0)
+			if (dbicurules != NULL)
+				ereport(ERROR,
+						(errmsg("ICU_RULES not supported for C or POSIX locale")));
+		}
+		else
+		{
+			/*
+			 * During binary upgrade, or when the locale came from the
+			 * template database, preserve locale string. Otherwise,
+			 * canonicalize to a language tag.
+			 */
+			if (!IsBinaryUpgrade && dbiculocale != src_iculocale)
 			{
-				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
-								langtag, dbiculocale)));
+				char	   *langtag = icu_language_tag(dbiculocale,
+													   icu_validation_level);
+
+				if (langtag && strcmp(dbiculocale, langtag) != 0)
+				{
+					ereport(NOTICE,
+							(errmsg("using standard form \"%s\" for locale \"%s\"",
+									langtag, dbiculocale)));
 
-				dbiculocale = langtag;
+					dbiculocale = langtag;
+				}
 			}
-		}
 
-		icu_validate_locale(dbiculocale);
+			icu_validate_locale(dbiculocale);
+		}
 	}
 	else
 	{
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 31e3b16ae0..986dcbd2a7 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1246,8 +1246,15 @@ lookup_collation_cache(Oid collation, bool set_flags)
 		}
 		else
 		{
-			cache_entry->collate_is_c = false;
-			cache_entry->ctype_is_c = false;
+			Datum		datum;
+			const char *colliculocale;
+
+			datum = SysCacheGetAttrNotNull(COLLOID, tp, Anum_pg_collation_colliculocale);
+			colliculocale = TextDatumGetCString(datum);
+
+			cache_entry->collate_is_c = ((strcmp(colliculocale, "C") == 0) ||
+										 (strcmp(colliculocale, "POSIX") == 0));
+			cache_entry->ctype_is_c = cache_entry->collate_is_c;
 		}
 
 		cache_entry->flags_valid = true;
@@ -1279,16 +1286,27 @@ lc_collate_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_COLLATE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_COLLATE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_COLLATE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_COLLATE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1332,16 +1350,27 @@ lc_ctype_is_c(Oid collation)
 	if (collation == DEFAULT_COLLATION_OID)
 	{
 		static int	result = -1;
-		char	   *localeptr;
-
-		if (default_locale.provider == COLLPROVIDER_ICU)
-			return false;
+		const char *localeptr;
 
 		if (result >= 0)
 			return (bool) result;
-		localeptr = setlocale(LC_CTYPE, NULL);
-		if (!localeptr)
-			elog(ERROR, "invalid LC_CTYPE setting");
+
+		if (default_locale.provider == COLLPROVIDER_ICU)
+		{
+#ifdef USE_ICU
+			localeptr = default_locale.info.icu.locale;
+#else
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("ICU is not supported in this build")));
+#endif
+		}
+		else
+		{
+			localeptr = setlocale(LC_CTYPE, NULL);
+			if (!localeptr)
+				elog(ERROR, "invalid LC_CTYPE setting");
+		}
 
 		if (strcmp(localeptr, "C") == 0)
 			result = true;
@@ -1375,7 +1404,13 @@ make_icu_collator(const char *iculocstr,
 #ifdef USE_ICU
 	UCollator  *collator;
 
-	collator = pg_ucol_open(iculocstr);
+	if (strcmp(iculocstr, "C") == 0 || strcmp(iculocstr, "POSIX") == 0)
+	{
+		Assert(icurules == NULL);
+		collator = NULL;
+	}
+	else
+		collator = pg_ucol_open(iculocstr);
 
 	/*
 	 * If rules are specified, we extract the rules of the standard collation,
@@ -1650,6 +1685,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (strcmp("C", collcollate) == 0 || strcmp("POSIX", collcollate) == 0)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
@@ -1668,9 +1706,7 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 	else
 #endif
 		if (collprovider == COLLPROVIDER_LIBC &&
-			pg_strcasecmp("C", collcollate) != 0 &&
-			pg_strncasecmp("C.", collcollate, 2) != 0 &&
-			pg_strcasecmp("POSIX", collcollate) != 0)
+			pg_strncasecmp("C.", collcollate, 2) != 0)
 	{
 #if defined(__GLIBC__)
 		/* Use the glibc version because we don't have anything better. */
@@ -2457,6 +2493,13 @@ pg_ucol_open(const char *loc_str)
 	if (loc_str == NULL)
 		elog(ERROR, "opening default collator is not supported");
 
+	/*
+	 * Must never open special values C or POSIX, which are treated specially
+	 * and not passed to the provider.
+	 */
+	if (strcmp(loc_str, "C") == 0 || strcmp(loc_str, "POSIX") == 0)
+		elog(ERROR, "unexpected ICU locale string: %s", loc_str);
+
 	/*
 	 * In ICU versions 54 and earlier, "und" is not a recognized spelling of
 	 * the root locale. If the first component of the locale is "und", replace
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 09a5c98cc0..7aa2d871e3 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2428,14 +2428,23 @@ setlocales(void)
 			printf(_("Using default ICU locale \"%s\".\n"), icu_locale);
 		}
 
-		/* canonicalize to a language tag */
-		langtag = icu_language_tag(icu_locale);
-		printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
-			   langtag, icu_locale);
-		pg_free(icu_locale);
-		icu_locale = langtag;
-
-		icu_validate_locale(icu_locale);
+		if (strcmp(icu_locale, "C") == 0 ||
+			strcmp(icu_locale, "POSIX") == 0)
+		{
+			if (icu_rules != NULL)
+				pg_fatal("RULES not supported for C or POSIX locale");
+		}
+		else
+		{
+			/* canonicalize to a language tag */
+			langtag = icu_language_tag(icu_locale);
+			printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
+				   langtag, icu_locale);
+			pg_free(icu_locale);
+			icu_locale = langtag;
+
+			icu_validate_locale(icu_locale);
+		}
 
 		/*
 		 * In supported builds, the ICU locale ID will be opened during
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index c658ee1404..bfc28ecfcf 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1043,12 +1043,18 @@ ERROR:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
 ERROR:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+ERROR:  nondeterministic collations not supported for C or POSIX locale
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
+ERROR:  RULES not supported for C or POSIX locale
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 WARNING:  could not convert locale name "@colStrength=primary;nonsense=yes" to language tag: U_ILLEGAL_ARGUMENT_ERROR
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
 WARNING:  ICU locale "nonsense-nowhere" has unknown language "nonsense"
 HINT:  To disable ICU locale validation, set parameter icu_validation_level to DISABLED.
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 CREATE COLLATION test4 FROM nonsense;
 ERROR:  collation "nonsense" for encoding "UTF8" does not exist
 CREATE COLLATION test5 FROM test0;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index 7bd0901281..572dc5a50a 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -379,9 +379,13 @@ CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, nee
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', deterministic = false); -- fails
+CREATE COLLATION testx (provider = icu, locale = 'C', rules = '&V << w <<< W'); -- fails
 RESET icu_validation_level;
 CREATE COLLATION testx (provider = icu, locale = '@colStrength=primary;nonsense=yes'); DROP COLLATION testx;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'C'); DROP COLLATION testx;
+CREATE COLLATION testx (provider = icu, locale = 'POSIX'); DROP COLLATION testx;
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
-- 
2.34.1

From b9a27c9536c4cc2a748a293fb095e985eec9de7b Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 1 May 2023 15:38:29 -0700
Subject: [PATCH v10 1/4] Introduce collation provider "builtin".

Only supports locale "C" (or equivalently, "POSIX"). Provides
locale-unaware semantics that are implemented as fast byte operations
in Postgres, independent of the operating system or any provider
libraries.

Equivalent (in semantics and implementation) to the libc provider with
locale "C", except that LC_COLLATE and LC_CTYPE can be set
independently.

Use provider "builtin" for built-in collation "ucs_basic" instead of
libc.

Discussion: https://postgr.es/m/ab925f69-5f9d-f85e-b87c-bd2a44798659@joeconway.com
---
 doc/src/sgml/charset.sgml              |  89 +++++++++++++++++----
 doc/src/sgml/ref/create_collation.sgml |  11 ++-
 doc/src/sgml/ref/create_database.sgml  |   8 +-
 doc/src/sgml/ref/createdb.sgml         |   2 +-
 doc/src/sgml/ref/initdb.sgml           |   7 +-
 src/backend/catalog/pg_collation.c     |   7 +-
 src/backend/commands/collationcmds.c   | 103 +++++++++++++++++++++----
 src/backend/commands/dbcommands.c      |  69 ++++++++++++++---
 src/backend/utils/adt/pg_locale.c      |  27 ++++++-
 src/backend/utils/init/postinit.c      |  10 ++-
 src/bin/initdb/initdb.c                |  24 ++++--
 src/bin/initdb/t/001_initdb.pl         |  47 +++++++++++
 src/bin/pg_dump/pg_dump.c              |  49 +++++++++---
 src/bin/pg_upgrade/t/002_pg_upgrade.pl |  35 +++++++--
 src/bin/psql/describe.c                |   4 +-
 src/bin/scripts/createdb.c             |   2 +-
 src/bin/scripts/t/020_createdb.pl      |  56 ++++++++++++++
 src/include/catalog/catversion.h       |   2 +-
 src/include/catalog/pg_collation.dat   |   3 +-
 src/include/catalog/pg_collation.h     |   3 +
 src/test/icu/t/010_database.pl         |   9 +++
 src/test/regress/expected/collate.out  |  25 +++++-
 src/test/regress/sql/collate.sql       |  10 +++
 23 files changed, 519 insertions(+), 83 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index ed84465996..b38cf82f83 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -342,22 +342,14 @@ initdb --locale=sv_SE
    <title>Locale Providers</title>
 
    <para>
-    <productname>PostgreSQL</productname> supports multiple <firstterm>locale
-    providers</firstterm>.  This specifies which library supplies the locale
-    data.  One standard provider name is <literal>libc</literal>, which uses
-    the locales provided by the operating system C library.  These are the
-    locales used by most tools provided by the operating system.  Another
-    provider is <literal>icu</literal>, which uses the external
-    ICU<indexterm><primary>ICU</primary></indexterm> library.  ICU locales can
-    only be used if support for ICU was configured when PostgreSQL was built.
+    A locale provider specifies which library defines the locale behavior for
+    collations and character classifications.
    </para>
 
    <para>
     The commands and tools that select the locale settings, as described
-    above, each have an option to select the locale provider.  The examples
-    shown earlier all use the <literal>libc</literal> provider, which is the
-    default.  Here is an example to initialize a database cluster using the
-    ICU provider:
+    above, each have an option to select the locale provider. Here is an
+    example to initialize a database cluster using the ICU provider:
 <programlisting>
 initdb --locale-provider=icu --icu-locale=en
 </programlisting>
@@ -370,12 +362,75 @@ initdb --locale-provider=icu --icu-locale=en
    </para>
 
    <para>
-    Which locale provider to use depends on individual requirements.  For most
-    basic uses, either provider will give adequate results.  For the libc
-    provider, it depends on what the operating system offers; some operating
-    systems are better than others.  For advanced uses, ICU offers more locale
-    variants and customization options.
+    Regardless of the locale provider, the operating system is still used to
+    provide some locale-aware behavior, such as messages (see <xref
+    linkend="guc-lc-messages"/>).
    </para>
+
+   <para>
+    The available locale providers are listed below.
+   </para>
+
+   <sect3 id="locale-provider-builtin">
+    <title>Builtin</title>
+    <para>
+     The <literal>builtin</literal> provider uses simple built-in operations
+     which are not locale-aware. Only the <literal>C</literal> (or
+     equivalently, <literal>POSIX</literal>) locales are supported for this
+     provider.
+    </para>
+    <para>
+     The collation and character classification behavior is equivalent to
+     using the <literal>libc</literal> provider with locale
+     <literal>C</literal>, except that <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently.
+    </para>
+    <note>
+     <para>
+      When using the <literal>builtin</literal> locale provider, behavior may
+      depend on the database encoding.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-icu">
+    <title>ICU</title>
+    <para>
+     The <literal>icu</literal> provider uses the external
+     ICU<indexterm><primary>ICU</primary></indexterm>
+     library. <productname>PostgreSQL</productname> must have been configured
+     with support.
+    </para>
+    <para>
+     ICU provides collation and character classification behavior that is
+     independent of the operating system and database encoding, which is
+     preferable if you expect to transition to other platforms without any
+     change in results. <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
+    </para>
+    <note>
+     <para>
+      For the ICU provider, results may depend on the version of the ICU
+      library used, as it is updated to reflect changes in natural language
+      over time.
+     </para>
+    </note>
+   </sect3>
+   <sect3 id="locale-provider-libc">
+    <title>libc</title>
+    <para>
+     The <literal>libc</literal> provider uses the operating system's C
+     library. The collation and character classification behavior is
+     controlled by the settings <literal>LC_COLLATE</literal> and
+     <literal>LC_CTYPE</literal>, so they cannot be set independently.
+    </para>
+    <note>
+     <para>
+      The same locale name may have different behavior on different platforms
+      when using the libc provider.
+     </para>
+    </note>
+   </sect3>
+
   </sect2>
   <sect2 id="icu-locales">
    <title>ICU Locales</title>
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..c63a350c1e 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -89,6 +89,11 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
        and <symbol>LC_CTYPE</symbol> at once.  If you specify this,
        you cannot specify either of those parameters.
       </para>
+      <para>
+       If <replaceable>provider</replaceable> is <literal>builtin</literal>,
+       then <replaceable>locale</replaceable> must be specified and set to
+       either <literal>C</literal> or <literal>POSIX</literal>.
+      </para>
      </listitem>
     </varlistentry>
 
@@ -120,9 +125,9 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
      <listitem>
       <para>
        Specifies the provider to use for locale services associated with this
-       collation.  Possible values are
-       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
-       (if the server was built with ICU support) or <literal>libc</literal>.
+       collation.  Possible values are <literal>builtin</literal>,
+       <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
+       the server was built with ICU support) or <literal>libc</literal>.
        <literal>libc</literal> is the default.  See <xref
        linkend="locale-providers"/> for details.
       </para>
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..5655d6c823 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -148,6 +148,12 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
         This is a shortcut for setting <symbol>LC_COLLATE</symbol>
         and <symbol>LC_CTYPE</symbol> at once.
        </para>
+       <para>
+        If <xref linkend="create-database-locale-provider"/> is
+        <literal>builtin</literal>, then <replaceable>locale</replaceable>
+        must be specified and set to either <literal>C</literal> or
+        <literal>POSIX</literal>.
+       </para>
        <tip>
         <para>
          The other locale settings <xref linkend="guc-lc-messages"/>, <xref
@@ -212,7 +218,7 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <listitem>
        <para>
         Specifies the provider to use for the default collation in this
-        database.  Possible values are
+        database.  Possible values are <literal>builtin</literal>,
         <literal>icu</literal><indexterm><primary>ICU</primary></indexterm>
         (if the server was built with ICU support) or <literal>libc</literal>.
         By default, the provider is the same as that of the <xref
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..ee4644818d 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -168,7 +168,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry>
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>builtin</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         Specifies the locale provider for the database's default collation.
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..96f84dc8ca 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -294,6 +294,11 @@ PostgreSQL documentation
         environment that <command>initdb</command> runs in. Locale
         support is described in <xref linkend="locale"/>.
        </para>
+       <para>
+        If <option>--locale-provider</option> is <literal>builtin</literal>,
+        <option>--locale</option> must be specified and set to
+        <literal>C</literal> (or equivalently, <literal>POSIX</literal>).
+       </para>
       </listitem>
      </varlistentry>
 
@@ -323,7 +328,7 @@ PostgreSQL documentation
      </varlistentry>
 
      <varlistentry id="app-initdb-option-locale-provider">
-      <term><option>--locale-provider={<literal>libc</literal>|<literal>icu</literal>}</option></term>
+      <term><option>--locale-provider={<literal>builtin</literal>|<literal>libc</literal>|<literal>icu</literal>}</option></term>
       <listitem>
        <para>
         This option sets the locale provider for databases created in the new
diff --git a/src/backend/catalog/pg_collation.c b/src/backend/catalog/pg_collation.c
index fd022e6fc2..0b1ba359b6 100644
--- a/src/backend/catalog/pg_collation.c
+++ b/src/backend/catalog/pg_collation.c
@@ -68,7 +68,12 @@ CollationCreate(const char *collname, Oid collnamespace,
 	Assert(collname);
 	Assert(collnamespace);
 	Assert(collowner);
-	Assert((collcollate && collctype) || colliculocale);
+	Assert((collprovider == COLLPROVIDER_BUILTIN &&
+			!collcollate && !collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_LIBC &&
+			 collcollate &&  collctype && !colliculocale) ||
+		   (collprovider == COLLPROVIDER_ICU &&
+			!collcollate && !collctype &&  colliculocale));
 
 	/*
 	 * Make sure there is no existing collation of same name & encoding.
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 2969a2bb21..4748946499 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -66,6 +66,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 	DefElem    *deterministicEl = NULL;
 	DefElem    *rulesEl = NULL;
 	DefElem    *versionEl = NULL;
+	char	   *builtin_locale = NULL;
 	char	   *collcollate;
 	char	   *collctype;
 	char	   *colliculocale;
@@ -215,7 +216,9 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 
 		if (collproviderstr)
 		{
-			if (pg_strcasecmp(collproviderstr, "icu") == 0)
+			if (pg_strcasecmp(collproviderstr, "builtin") == 0)
+				collprovider = COLLPROVIDER_BUILTIN;
+			else if (pg_strcasecmp(collproviderstr, "icu") == 0)
 				collprovider = COLLPROVIDER_ICU;
 			else if (pg_strcasecmp(collproviderstr, "libc") == 0)
 				collprovider = COLLPROVIDER_LIBC;
@@ -230,7 +233,11 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 
 		if (localeEl)
 		{
-			if (collprovider == COLLPROVIDER_LIBC)
+			if (collprovider == COLLPROVIDER_BUILTIN)
+			{
+				builtin_locale = defGetString(localeEl);
+			}
+			else if (collprovider == COLLPROVIDER_LIBC)
 			{
 				collcollate = defGetString(localeEl);
 				collctype = defGetString(localeEl);
@@ -245,7 +252,22 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		if (lcctypeEl)
 			collctype = defGetString(lcctypeEl);
 
-		if (collprovider == COLLPROVIDER_LIBC)
+		if (collprovider == COLLPROVIDER_BUILTIN)
+		{
+			if (!builtin_locale)
+				ereport(ERROR,
+						(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
+						 errmsg("parameter \"locale\" must be specified")));
+
+			if (strcmp(builtin_locale, "C") != 0 &&
+				strcmp(builtin_locale, "POSIX") != 0)
+				ereport(ERROR,
+						(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+						 errmsg("collation provider \"builtin\" does not support locale \"%s\"",
+								builtin_locale),
+						 errhint("The built-in collation provider only supports the \"C\" and \"POSIX\" locales.")));
+		}
+		else if (collprovider == COLLPROVIDER_LIBC)
 		{
 			if (!collcollate)
 				ereport(ERROR,
@@ -302,7 +324,17 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 					(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
 					 errmsg("ICU rules cannot be specified unless locale provider is ICU")));
 
-		if (collprovider == COLLPROVIDER_ICU)
+		if (collprovider == COLLPROVIDER_BUILTIN)
+		{
+			/*
+			 * Behavior may be different in different encodings, so set
+			 * collencoding to the current database encoding. No validation is
+			 * required, because the "builtin" provider is compatible with any
+			 * encoding.
+			 */
+			collencoding = GetDatabaseEncoding();
+		}
+		else if (collprovider == COLLPROVIDER_ICU)
 		{
 #ifdef USE_ICU
 			/*
@@ -331,7 +363,18 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 	}
 
 	if (!collversion)
-		collversion = get_collation_actual_version(collprovider, collprovider == COLLPROVIDER_ICU ? colliculocale : collcollate);
+	{
+		char *locale;
+
+		if (collprovider == COLLPROVIDER_ICU)
+			locale = colliculocale;
+		else if (collprovider == COLLPROVIDER_LIBC)
+			locale = collcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+		collversion = get_collation_actual_version(collprovider, locale);
+	}
 
 	newoid = CollationCreate(collName,
 							 collNamespace,
@@ -406,6 +449,7 @@ AlterCollation(AlterCollationStmt *stmt)
 	Form_pg_collation collForm;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 	ObjectAddress address;
@@ -430,8 +474,20 @@ AlterCollation(AlterCollationStmt *stmt)
 	datum = SysCacheGetAttr(COLLOID, tup, Anum_pg_collation_collversion, &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = SysCacheGetAttrNotNull(COLLOID, tup, collForm->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
-	newversion = get_collation_actual_version(collForm->collprovider, TextDatumGetCString(datum));
+	if (collForm->collprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_colliculocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (collForm->collprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(COLLOID, tup, Anum_pg_collation_collcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+	newversion = get_collation_actual_version(collForm->collprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -495,11 +551,18 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_database) GETSTRUCT(dbtup))->datlocprovider;
 
-		datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_daticulocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(DATABASEOID, dbtup, Anum_pg_database_datcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
 
 		ReleaseSysCache(dbtup);
 	}
@@ -516,11 +579,19 @@ pg_collation_actual_version(PG_FUNCTION_ARGS)
 
 		provider = ((Form_pg_collation) GETSTRUCT(colltp))->collprovider;
 		Assert(provider != COLLPROVIDER_DEFAULT);
-		datum = SysCacheGetAttrNotNull(COLLOID, colltp,
-									   provider == COLLPROVIDER_ICU ?
-									   Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
-		locale = TextDatumGetCString(datum);
+		if (provider == COLLPROVIDER_ICU)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_colliculocale);
+			locale = TextDatumGetCString(datum);
+		}
+		else if (provider == COLLPROVIDER_LIBC)
+		{
+			datum = SysCacheGetAttrNotNull(COLLOID, colltp, Anum_pg_collation_collcollate);
+			locale = TextDatumGetCString(datum);
+		}
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
 
 		ReleaseSysCache(colltp);
 	}
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 99d4080ea9..016852644f 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -909,7 +909,9 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	{
 		char	   *locproviderstr = defGetString(dlocprovider);
 
-		if (pg_strcasecmp(locproviderstr, "icu") == 0)
+		if (pg_strcasecmp(locproviderstr, "builtin") == 0)
+			dblocprovider = COLLPROVIDER_BUILTIN;
+		else if (pg_strcasecmp(locproviderstr, "icu") == 0)
 			dblocprovider = COLLPROVIDER_ICU;
 		else if (pg_strcasecmp(locproviderstr, "libc") == 0)
 			dblocprovider = COLLPROVIDER_LIBC;
@@ -1177,9 +1179,17 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 */
 	if (src_collversion && !dcollversion)
 	{
-		char	   *actual_versionstr;
+		char	*actual_versionstr;
+		char	*locale;
 
-		actual_versionstr = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+		actual_versionstr = get_collation_actual_version(dblocprovider, locale);
 		if (!actual_versionstr)
 			ereport(ERROR,
 					(errmsg("template database \"%s\" has a collation version, but no actual collation version could be determined",
@@ -1207,7 +1217,18 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	 * collation version, which is normally only the case for template0.
 	 */
 	if (dbcollversion == NULL)
-		dbcollversion = get_collation_actual_version(dblocprovider, dblocprovider == COLLPROVIDER_ICU ? dbiculocale : dbcollate);
+	{
+		char *locale;
+
+		if (dblocprovider == COLLPROVIDER_ICU)
+			locale = dbiculocale;
+		else if (dblocprovider == COLLPROVIDER_LIBC)
+			locale = dbcollate;
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+		dbcollversion = get_collation_actual_version(dblocprovider, locale);
+	}
 
 	/* Resolve default tablespace for new database */
 	if (dtablespacename && dtablespacename->arg)
@@ -2403,6 +2424,7 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	ObjectAddress address;
 	Datum		datum;
 	bool		isnull;
+	char	   *locale;
 	char	   *oldversion;
 	char	   *newversion;
 
@@ -2429,10 +2451,24 @@ AlterDatabaseRefreshColl(AlterDatabaseRefreshCollStmt *stmt)
 	datum = heap_getattr(tuple, Anum_pg_database_datcollversion, RelationGetDescr(rel), &isnull);
 	oldversion = isnull ? NULL : TextDatumGetCString(datum);
 
-	datum = heap_getattr(tuple, datForm->datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
-	if (isnull)
-		elog(ERROR, "unexpected null in pg_database");
-	newversion = get_collation_actual_version(datForm->datlocprovider, TextDatumGetCString(datum));
+	if (datForm->datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_daticulocale, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datForm->datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = heap_getattr(tuple, Anum_pg_database_datcollate, RelationGetDescr(rel), &isnull);
+		if (isnull)
+			elog(ERROR, "unexpected null in pg_database");
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+	newversion = get_collation_actual_version(datForm->datlocprovider, locale);
 
 	/* cannot change from NULL to non-NULL or vice versa */
 	if ((!oldversion && newversion) || (oldversion && !newversion))
@@ -2617,6 +2653,7 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 	HeapTuple	tp;
 	char		datlocprovider;
 	Datum		datum;
+	char	   *locale;
 	char	   *version;
 
 	tp = SearchSysCache1(DATABASEOID, ObjectIdGetDatum(dbid));
@@ -2627,8 +2664,20 @@ pg_database_collation_actual_version(PG_FUNCTION_ARGS)
 
 	datlocprovider = ((Form_pg_database) GETSTRUCT(tp))->datlocprovider;
 
-	datum = SysCacheGetAttrNotNull(DATABASEOID, tp, datlocprovider == COLLPROVIDER_ICU ? Anum_pg_database_daticulocale : Anum_pg_database_datcollate);
-	version = get_collation_actual_version(datlocprovider, TextDatumGetCString(datum));
+	if (datlocprovider == COLLPROVIDER_ICU)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_daticulocale);
+		locale = TextDatumGetCString(datum);
+	}
+	else if (datlocprovider == COLLPROVIDER_LIBC)
+	{
+		datum = SysCacheGetAttrNotNull(DATABASEOID, tp, Anum_pg_database_datcollate);
+		locale = TextDatumGetCString(datum);
+	}
+	else
+		locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+	version = get_collation_actual_version(datlocprovider, locale);
 
 	ReleaseSysCache(tp);
 
diff --git a/src/backend/utils/adt/pg_locale.c b/src/backend/utils/adt/pg_locale.c
index 31e3b16ae0..d4d7affba6 100644
--- a/src/backend/utils/adt/pg_locale.c
+++ b/src/backend/utils/adt/pg_locale.c
@@ -1228,7 +1228,12 @@ lookup_collation_cache(Oid collation, bool set_flags)
 			elog(ERROR, "cache lookup failed for collation %u", collation);
 		collform = (Form_pg_collation) GETSTRUCT(tp);
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_BUILTIN)
+		{
+			cache_entry->collate_is_c = true;
+			cache_entry->ctype_is_c = true;
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 			Datum		datum;
 			const char *collcollate;
@@ -1281,6 +1286,9 @@ lc_collate_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_BUILTIN)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1334,6 +1342,9 @@ lc_ctype_is_c(Oid collation)
 		static int	result = -1;
 		char	   *localeptr;
 
+		if (default_locale.provider == COLLPROVIDER_BUILTIN)
+			return true;
+
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return false;
 
@@ -1487,8 +1498,10 @@ pg_newlocale_from_collation(Oid collid)
 	{
 		if (default_locale.provider == COLLPROVIDER_ICU)
 			return &default_locale;
-		else
+		else if (default_locale.provider == COLLPROVIDER_LIBC)
 			return (pg_locale_t) 0;
+		else
+			elog(ERROR, "cannot open collation with provider \"builtin\"");
 	}
 
 	cache_entry = lookup_collation_cache(collid, false);
@@ -1513,7 +1526,11 @@ pg_newlocale_from_collation(Oid collid)
 		result.provider = collform->collprovider;
 		result.deterministic = collform->collisdeterministic;
 
-		if (collform->collprovider == COLLPROVIDER_LIBC)
+		if (collform->collprovider == COLLPROVIDER_BUILTIN)
+		{
+			elog(ERROR, "cannot open collation with provider \"builtin\"");
+		}
+		else if (collform->collprovider == COLLPROVIDER_LIBC)
 		{
 #ifdef HAVE_LOCALE_T
 			const char *collcollate;
@@ -1599,6 +1616,7 @@ pg_newlocale_from_collation(Oid collid)
 
 			collversionstr = TextDatumGetCString(datum);
 
+			Assert(collform->collprovider != COLLPROVIDER_BUILTIN);
 			datum = SysCacheGetAttrNotNull(COLLOID, tp, collform->collprovider == COLLPROVIDER_ICU ? Anum_pg_collation_colliculocale : Anum_pg_collation_collcollate);
 
 			actual_versionstr = get_collation_actual_version(collform->collprovider,
@@ -1650,6 +1668,9 @@ get_collation_actual_version(char collprovider, const char *collcollate)
 {
 	char	   *collversion = NULL;
 
+	if (collprovider == COLLPROVIDER_BUILTIN)
+		return NULL;
+
 #ifdef USE_ICU
 	if (collprovider == COLLPROVIDER_ICU)
 	{
diff --git a/src/backend/utils/init/postinit.c b/src/backend/utils/init/postinit.c
index 561bd13ed2..12c36d12e6 100644
--- a/src/backend/utils/init/postinit.c
+++ b/src/backend/utils/init/postinit.c
@@ -461,10 +461,18 @@ CheckMyDatabase(const char *name, bool am_superuser, bool override_allow_connect
 	{
 		char	   *actual_versionstr;
 		char	   *collversionstr;
+		char	   *locale;
 
 		collversionstr = TextDatumGetCString(datum);
 
-		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, dbform->datlocprovider == COLLPROVIDER_ICU ? iculocale : collate);
+		if (dbform->datlocprovider == COLLPROVIDER_ICU)
+			locale = iculocale;
+		else if (dbform->datlocprovider == COLLPROVIDER_LIBC)
+			locale = collate;
+		else
+			locale = NULL; /* COLLPROVIDER_BUILTIN */
+
+		actual_versionstr = get_collation_actual_version(dbform->datlocprovider, locale);
 		if (!actual_versionstr)
 			/* should not happen */
 			elog(WARNING,
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 09a5c98cc0..6fc19c8d64 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2472,7 +2472,7 @@ usage(const char *progname)
 			 "                            set default locale in the respective category for\n"
 			 "                            new databases (default taken from environment)\n"));
 	printf(_("      --no-locale           equivalent to --locale=C\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={builtin|libc|icu}\n"
 			 "                            set default locale provider for new databases\n"));
 	printf(_("      --pwfile=FILE         read password for the new superuser from file\n"));
 	printf(_("  -T, --text-search-config=CFG\n"
@@ -2622,7 +2622,15 @@ setup_locale_encoding(void)
 {
 	setlocales();
 
-	if (locale_provider == COLLPROVIDER_LIBC &&
+	if (locale_provider == COLLPROVIDER_BUILTIN &&
+		strcmp(lc_ctype, "C") == 0 &&
+		strcmp(lc_collate, "C") == 0 &&
+		strcmp(lc_time, "C") == 0 &&
+		strcmp(lc_numeric, "C") == 0 &&
+		strcmp(lc_monetary, "C") == 0 &&
+		strcmp(lc_messages, "C") == 0)
+		printf(_("The database cluster will be initialized with no locale.\n"));
+	else if (locale_provider == COLLPROVIDER_LIBC &&
 		strcmp(lc_ctype, lc_collate) == 0 &&
 		strcmp(lc_ctype, lc_time) == 0 &&
 		strcmp(lc_ctype, lc_numeric) == 0 &&
@@ -2633,9 +2641,11 @@ setup_locale_encoding(void)
 	else
 	{
 		printf(_("The database cluster will be initialized with this locale configuration:\n"));
-		printf(_("  provider:    %s\n"), collprovider_name(locale_provider));
-		if (icu_locale)
-			printf(_("  ICU locale:  %s\n"), icu_locale);
+		printf(_("  default collation provider:  %s\n"), collprovider_name(locale_provider));
+		if (locale_provider == COLLPROVIDER_BUILTIN)
+			printf(_("  default collation locale:    %s\n"), "C");
+		else if (locale_provider == COLLPROVIDER_ICU)
+			printf(_("  default collation locale:    %s\n"), icu_locale);
 		printf(_("  LC_COLLATE:  %s\n"
 				 "  LC_CTYPE:    %s\n"
 				 "  LC_MESSAGES: %s\n"
@@ -3296,7 +3306,9 @@ main(int argc, char *argv[])
 										 "-c debug_discard_caches=1");
 				break;
 			case 15:
-				if (strcmp(optarg, "icu") == 0)
+				if (strcmp(optarg, "builtin") == 0)
+					locale_provider = COLLPROVIDER_BUILTIN;
+				else if (strcmp(optarg, "icu") == 0)
 					locale_provider = COLLPROVIDER_ICU;
 				else if (strcmp(optarg, "libc") == 0)
 					locale_provider = COLLPROVIDER_LIBC;
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index fa00bb3dab..157d6acfd4 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -154,6 +154,53 @@ else
 		'locale provider ICU fails since no ICU support');
 }
 
+command_ok(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', "$tempdir/data6"
+	],
+	'locale provider builtin'
+);
+
+command_ok(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=C',
+		"$tempdir/data7"
+	],
+	'locale provider builtin with --locale'
+);
+
+command_ok(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--lc-collate=C',
+		"$tempdir/data8"
+	],
+	'locale provider builtin with --lc-collate'
+);
+
+command_ok(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--lc-ctype=C',
+		"$tempdir/data9"
+	],
+	'locale provider builtin with --lc-ctype'
+);
+
+command_fails(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--icu-locale=en',
+		"$tempdir/dataX"
+	],
+	'fails for locale provider builtin with ICU locale'
+);
+
+command_fails(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--icu-rules=""',
+		"$tempdir/dataX"
+	],
+	'fails for locale provider builtin with ICU rules'
+);
+
 command_fails(
 	[ 'initdb', '--no-sync', '--locale-provider=xyz', "$tempdir/dataX" ],
 	'fails for invalid locale provider');
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index 5dab1ba9ea..c75818c3a4 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -3070,7 +3070,9 @@ dumpDatabase(Archive *fout)
 	}
 
 	appendPQExpBufferStr(creaQry, " LOCALE_PROVIDER = ");
-	if (datlocprovider[0] == 'c')
+	if (datlocprovider[0] == 'b')
+		appendPQExpBufferStr(creaQry, "builtin");
+	else if (datlocprovider[0] == 'c')
 		appendPQExpBufferStr(creaQry, "libc");
 	else if (datlocprovider[0] == 'i')
 		appendPQExpBufferStr(creaQry, "icu");
@@ -13432,7 +13434,9 @@ dumpCollation(Archive *fout, const CollInfo *collinfo)
 					  fmtQualifiedDumpable(collinfo));
 
 	appendPQExpBufferStr(q, "provider = ");
-	if (collprovider[0] == 'c')
+	if (collprovider[0] == 'b')
+		appendPQExpBufferStr(q, "builtin");
+	else if (collprovider[0] == 'c')
 		appendPQExpBufferStr(q, "libc");
 	else if (collprovider[0] == 'i')
 		appendPQExpBufferStr(q, "icu");
@@ -13446,13 +13450,42 @@ dumpCollation(Archive *fout, const CollInfo *collinfo)
 	if (strcmp(PQgetvalue(res, 0, i_collisdeterministic), "f") == 0)
 		appendPQExpBufferStr(q, ", deterministic = false");
 
-	if (colliculocale != NULL)
+	if (collprovider[0] == 'd')
 	{
+		Assert(colliculocale == NULL);
+		Assert(collicurules == NULL);
+		Assert(collcollate == NULL);
+		Assert(collctype == NULL);
+
+		/* no locale -- cannot be reloaded anyway */
+	}
+	else if (collprovider[0] == 'b')
+	{
+		Assert(colliculocale == NULL);
+		Assert(collicurules == NULL);
+		Assert(collcollate == NULL);
+		Assert(collctype == NULL);
+		appendPQExpBufferStr(q, ", locale = 'C'");
+	}
+	else if (collprovider[0] == 'i')
+	{
+		Assert(colliculocale != NULL);
+		Assert(collcollate == NULL);
+		Assert(collctype == NULL);
+
 		appendPQExpBufferStr(q, ", locale = ");
 		appendStringLiteralAH(q, colliculocale, fout);
+
+		if (collicurules)
+		{
+			appendPQExpBufferStr(q, ", rules = ");
+			appendStringLiteralAH(q, collicurules, fout);
+		}
 	}
-	else
+	else if (collprovider[0] == 'c')
 	{
+		Assert(colliculocale == NULL);
+		Assert(collicurules == NULL);
 		Assert(collcollate != NULL);
 		Assert(collctype != NULL);
 
@@ -13469,12 +13502,8 @@ dumpCollation(Archive *fout, const CollInfo *collinfo)
 			appendStringLiteralAH(q, collctype, fout);
 		}
 	}
-
-	if (collicurules)
-	{
-		appendPQExpBufferStr(q, ", rules = ");
-		appendStringLiteralAH(q, collicurules, fout);
-	}
+	else
+		pg_fatal("unrecognized collation provider '%c'", collprovider[0]);
 
 	/*
 	 * For binary upgrade, carry over the collation version.  For normal
diff --git a/src/bin/pg_upgrade/t/002_pg_upgrade.pl b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
index 41fce089d6..22d9fc10a0 100644
--- a/src/bin/pg_upgrade/t/002_pg_upgrade.pl
+++ b/src/bin/pg_upgrade/t/002_pg_upgrade.pl
@@ -114,22 +114,45 @@ my $original_locale = "C";
 my $original_iculocale = "";
 my $provider_field = "'c' AS datlocprovider";
 my $iculocale_field = "NULL AS daticulocale";
-if ($oldnode->pg_version >= 15 && $ENV{with_icu} eq 'yes')
+if ($oldnode->pg_version >= 15)
 {
 	$provider_field = "datlocprovider";
 	$iculocale_field = "daticulocale";
-	$original_provider = "i";
-	$original_iculocale = "fr-CA";
+
+	if ($ENV{with_icu} eq 'yes')
+	{
+		$original_provider = "i";
+		$original_iculocale = "fr-CA";
+	}
+}
+
+# use builtin provider instead of libc, if supported
+if ($oldnode->pg_version >= 16 && $ENV{with_icu} ne 'yes')
+{
+	$original_provider = "b";
 }
 
 my @initdb_params = @custom_opts;
 
 push @initdb_params, ('--encoding', 'UTF-8');
 push @initdb_params, ('--locale', $original_locale);
-if ($original_provider eq "i")
+
+# add --locale-provider, if supported
+if ($oldnode->pg_version >= 15)
 {
-	push @initdb_params, ('--locale-provider', 'icu');
-	push @initdb_params, ('--icu-locale', 'fr-CA');
+	if ($original_provider eq "b")
+	{
+		push @initdb_params, ('--locale-provider', 'builtin');
+	}
+	elsif ($original_provider eq "i")
+	{
+		push @initdb_params, ('--locale-provider', 'icu');
+		push @initdb_params, ('--icu-locale', 'fr-CA');
+	}
+	elsif ($original_provider eq "c")
+	{
+		push @initdb_params, ('--locale-provider', 'libc');
+	}
 }
 
 $node_params{extra} = \@initdb_params;
diff --git a/src/bin/psql/describe.c b/src/bin/psql/describe.c
index 9325a46b8f..5642638dfb 100644
--- a/src/bin/psql/describe.c
+++ b/src/bin/psql/describe.c
@@ -932,7 +932,7 @@ listAllDbs(const char *pattern, bool verbose)
 					  gettext_noop("Encoding"));
 	if (pset.sversion >= 150000)
 		appendPQExpBuffer(&buf,
-						  "  CASE d.datlocprovider WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
+						  "  CASE d.datlocprovider WHEN 'b' THEN 'builtin' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
 						  gettext_noop("Locale Provider"));
 	else
 		appendPQExpBuffer(&buf,
@@ -4873,7 +4873,7 @@ listCollations(const char *pattern, bool verbose, bool showSystem)
 
 	if (pset.sversion >= 100000)
 		appendPQExpBuffer(&buf,
-						  "  CASE c.collprovider WHEN 'd' THEN 'default' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
+						  "  CASE c.collprovider WHEN 'd' THEN 'default' WHEN 'b' THEN 'builtin' WHEN 'c' THEN 'libc' WHEN 'i' THEN 'icu' END AS \"%s\",\n",
 						  gettext_noop("Provider"));
 	else
 		appendPQExpBuffer(&buf,
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..41a8de659e 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -299,7 +299,7 @@ help(const char *progname)
 	printf(_("      --lc-ctype=LOCALE        LC_CTYPE setting for the database\n"));
 	printf(_("      --icu-locale=LOCALE      ICU locale setting for the database\n"));
 	printf(_("      --icu-rules=RULES        ICU rules setting for the database\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --locale-provider={builtin|libc|icu}\n"
 			 "                               locale provider for the database's default collation\n"));
 	printf(_("  -O, --owner=OWNER            database user to own the new database\n"));
 	printf(_("  -S, --strategy=STRATEGY      database creation strategy wal_log or file_copy\n"));
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index d0830a4a1d..f1d6db0f48 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -94,6 +94,62 @@ else
 		'create database with ICU fails since no ICU support');
 }
 
+$node->command_ok(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'tbuiltin1'
+	],
+	'create database with provider "builtin"'
+);
+
+$node->command_ok(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--locale=C', 'tbuiltin2'
+	],
+	'create database with provider "builtin" and locale "C"'
+);
+
+$node->command_ok(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--lc-collate=C', 'tbuiltin3'
+	],
+	'create database with provider "builtin" and LC_COLLATE=C'
+);
+
+$node->command_ok(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--lc-ctype=C', 'tbuiltin4'
+	],
+	'create database with provider "builtin" and LC_CTYPE=C'
+);
+
+$node->command_fails(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--icu-locale=en', 'tbuiltin5'
+	],
+	'create database with provider "builtin" and ICU_LOCALE="en"'
+);
+
+$node->command_fails(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--icu-rules=""', 'tbuiltin6'
+	],
+	'create database with provider "builtin" and ICU_RULES=""'
+);
+
+$node->command_fails(
+	[
+		'createdb', '-T', 'template1', '--locale-provider=builtin',
+		'--locale=C', 'tbuiltin7'
+	],
+	'create database with provider "builtin" not matching template'
+);
+
 $node->command_fails([ 'createdb', 'foobar1' ],
 	'fails if database already exists');
 
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index c784937a0e..52e17baafa 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	202305211
+#define CATALOG_VERSION_NO	202306081
 
 #endif
diff --git a/src/include/catalog/pg_collation.dat b/src/include/catalog/pg_collation.dat
index b6a69d1d42..cfb53807ed 100644
--- a/src/include/catalog/pg_collation.dat
+++ b/src/include/catalog/pg_collation.dat
@@ -24,8 +24,7 @@
   collname => 'POSIX', collprovider => 'c', collencoding => '-1',
   collcollate => 'POSIX', collctype => 'POSIX' },
 { oid => '962', descr => 'sorts by Unicode code point',
-  collname => 'ucs_basic', collprovider => 'c', collencoding => '6',
-  collcollate => 'C', collctype => 'C' },
+  collname => 'ucs_basic', collprovider => 'b', collencoding => '6' },
 { oid => '963',
   descr => 'sorts using the Unicode Collation Algorithm with default settings',
   collname => 'unicode', collprovider => 'i', collencoding => '-1',
diff --git a/src/include/catalog/pg_collation.h b/src/include/catalog/pg_collation.h
index bfa3568451..4009c4ec93 100644
--- a/src/include/catalog/pg_collation.h
+++ b/src/include/catalog/pg_collation.h
@@ -65,6 +65,7 @@ DECLARE_UNIQUE_INDEX_PKEY(pg_collation_oid_index, 3085, CollationOidIndexId, on
 #ifdef EXPOSE_TO_CLIENT_CODE
 
 #define COLLPROVIDER_DEFAULT	'd'
+#define COLLPROVIDER_BUILTIN	'b'
 #define COLLPROVIDER_ICU		'i'
 #define COLLPROVIDER_LIBC		'c'
 
@@ -73,6 +74,8 @@ collprovider_name(char c)
 {
 	switch (c)
 	{
+		case COLLPROVIDER_BUILTIN:
+			return "builtin";
 		case COLLPROVIDER_ICU:
 			return "icu";
 		case COLLPROVIDER_LIBC:
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index d3901f5d3f..26f71e1155 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -63,5 +63,14 @@ like(
 	qr/ERROR:  ICU locale must be specified/,
 	"ICU locale must be specified for ICU provider: error message");
 
+my ($ret, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu LOCALE_PROVIDER builtin LOCALE 'C' TEMPLATE dbicu}
+);
+isnt($ret, 0,
+	"locale provider must match template: exit code not 0");
+like(
+	$stderr,
+	qr/ERROR:  new locale provider \(builtin\) does not match locale provider of the template database \(icu\)/,
+	"locale provider must match template: error message");
 
 done_testing();
diff --git a/src/test/regress/expected/collate.out b/src/test/regress/expected/collate.out
index 0649564485..5b28de1b47 100644
--- a/src/test/regress/expected/collate.out
+++ b/src/test/regress/expected/collate.out
@@ -650,6 +650,27 @@ EXPLAIN (COSTS OFF)
 (3 rows)
 
 -- CREATE/DROP COLLATION
+CREATE COLLATION builtin_c ( PROVIDER = builtin, LOCALE = "C" );
+CREATE COLLATION builtin_posix ( PROVIDER = builtin, LOCALE = "POSIX" );
+SELECT b FROM collate_test1 ORDER BY b COLLATE builtin_c;
+  b  
+-----
+ ABD
+ Abc
+ abc
+ bbc
+(4 rows)
+
+CREATE COLLATION builtin2 ( PROVIDER = builtin ); -- fails
+ERROR:  parameter "locale" must be specified
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LOCALE = "en_US" ); -- fails
+ERROR:  collation provider "builtin" does not support locale "en_US"
+HINT:  The built-in collation provider only supports the "C" and "POSIX" locales.
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LC_CTYPE = "C", LC_COLLATE = "C" ); -- fails
+ERROR:  parameter "locale" must be specified
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LOCALE = "POSIX", LC_CTYPE = "POSIX" ); -- fails
+ERROR:  conflicting or redundant options
+DETAIL:  LOCALE cannot be specified together with LC_COLLATE or LC_CTYPE.
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
@@ -754,7 +775,7 @@ DETAIL:  FROM cannot be specified together with any other options.
 -- must get rid of them.
 --
 DROP SCHEMA collate_tests CASCADE;
-NOTICE:  drop cascades to 19 other objects
+NOTICE:  drop cascades to 21 other objects
 DETAIL:  drop cascades to table collate_test1
 drop cascades to table collate_test_like
 drop cascades to table collate_test2
@@ -771,6 +792,8 @@ drop cascades to function dup(anyelement)
 drop cascades to table collate_test20
 drop cascades to table collate_test21
 drop cascades to table collate_test22
+drop cascades to collation builtin_c
+drop cascades to collation builtin_posix
 drop cascades to collation mycoll2
 drop cascades to table collate_test23
 drop cascades to view collate_on_int
diff --git a/src/test/regress/sql/collate.sql b/src/test/regress/sql/collate.sql
index c3d40fc195..01d5c69fe4 100644
--- a/src/test/regress/sql/collate.sql
+++ b/src/test/regress/sql/collate.sql
@@ -244,6 +244,16 @@ EXPLAIN (COSTS OFF)
 
 -- CREATE/DROP COLLATION
 
+CREATE COLLATION builtin_c ( PROVIDER = builtin, LOCALE = "C" );
+CREATE COLLATION builtin_posix ( PROVIDER = builtin, LOCALE = "POSIX" );
+
+SELECT b FROM collate_test1 ORDER BY b COLLATE builtin_c;
+
+CREATE COLLATION builtin2 ( PROVIDER = builtin ); -- fails
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LOCALE = "en_US" ); -- fails
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LC_CTYPE = "C", LC_COLLATE = "C" ); -- fails
+CREATE COLLATION builtin2 ( PROVIDER = builtin, LOCALE = "POSIX", LC_CTYPE = "POSIX" ); -- fails
+
 CREATE COLLATION mycoll1 FROM "C";
 CREATE COLLATION mycoll2 ( LC_COLLATE = "POSIX", LC_CTYPE = "POSIX" );
 CREATE COLLATION mycoll3 FROM "default";  -- intentionally unsupported
-- 
2.34.1

From c48d91ab80e151093dbae0ada6a9691a2f112c5a Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 8 May 2023 13:48:01 -0700
Subject: [PATCH v10 2/4] ICU: for locale "C", automatically use "builtin"
 provider instead.

Postgres expects locale C to be optimizable to simple locale-unaware
byte operations; while ICU does not recognize the locale "C" at all,
and falls back to the root locale.

If the user specifies locale "C" when creating a new collation or a
new database with the ICU provider, automatically switch it to the
"builtin" provider.

If provider is libc, behavior is unchanged.

Discussion: https://postgr.es/m/ab925f69-5f9d-f85e-b87c-bd2a44798659@joeconway.com
---
 doc/src/sgml/charset.sgml                     |  6 +++
 doc/src/sgml/ref/create_collation.sgml        |  6 +++
 doc/src/sgml/ref/create_database.sgml         |  5 +++
 doc/src/sgml/ref/createdb.sgml                |  5 +++
 doc/src/sgml/ref/initdb.sgml                  | 23 ++++++-----
 src/backend/commands/collationcmds.c          | 20 ++++++++++
 src/backend/commands/dbcommands.c             | 21 ++++++++++
 src/bin/initdb/initdb.c                       | 17 ++++++++
 src/bin/initdb/t/001_initdb.pl                | 39 +++++++++++++++++++
 src/bin/scripts/t/020_createdb.pl             | 18 +++++++++
 .../regress/expected/collate.icu.utf8.out     | 14 +++++--
 src/test/regress/sql/collate.icu.utf8.sql     |  6 +++
 12 files changed, 168 insertions(+), 12 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index b38cf82f83..9a7f9263ec 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -407,6 +407,12 @@ initdb --locale-provider=icu --icu-locale=en
      change in results. <literal>LC_COLLATE</literal> and
      <literal>LC_CTYPE</literal> can be set independently of the ICU locale.
     </para>
+    <para>
+     The ICU provider does not accept the <literal>C</literal>
+     locale. Commands that create collations or database with the
+     <literal>icu</literal> provider and ICU locale <literal>C</literal> use
+     the provider <literal>builtin</literal> instead.
+    </para>
     <note>
      <para>
       For the ICU provider, results may depend on the version of the ICU
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index c63a350c1e..67ea247b39 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -131,6 +131,12 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
        <literal>libc</literal> is the default.  See <xref
        linkend="locale-providers"/> for details.
       </para>
+      <para>
+       If the provider is <literal>icu</literal> and the locale is
+       <literal>C</literal> or <literal>POSIX</literal>, the provider is
+       automatically set to <literal>builtin</literal>; as the ICU provider
+       doesn't support an ICU locale of <literal>C</literal>.
+      </para>
      </listitem>
     </varlistentry>
 
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 5655d6c823..17550c19bb 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -196,6 +196,11 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
        <para>
         Specifies the ICU locale ID if the ICU locale provider is used.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>builtin</literal>, as the
+        ICU provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index ee4644818d..5352cce744 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -154,6 +154,11 @@ PostgreSQL documentation
         Specifies the ICU locale ID to be used in this database, if the
         ICU locale provider is selected.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>builtin</literal>, as the
+        ICU provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
       </listitem>
      </varlistentry>
 
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 96f84dc8ca..d9620e5931 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -94,13 +94,14 @@ PostgreSQL documentation
 
   <para>
    By default, <command>initdb</command> uses the ICU library to provide
-   locale services if the server was built with ICU support; otherwise it uses
-   the <literal>libc</literal> locale provider (see <xref
-   linkend="locale-providers"/>). To choose the specific ICU locale ID to
-   apply, use the option <option>--icu-locale</option>.  Note that for
-   implementation reasons and to support legacy code,
-   <command>initdb</command> will still select and initialize libc locale
-   settings when the ICU locale provider is used.
+   locale services if the server was built with ICU support. To choose the
+   specific ICU locale ID to apply, use the option
+   <option>--icu-locale</option>.  Note that for implementation reasons and to
+   support legacy code, <command>initdb</command> will still select and
+   initialize libc locale settings when the ICU locale provider is used. If
+   <literal>--icu-locale</literal> is <literal>C</literal> (or equivalently
+   <literal>POSIX</literal>), the <literal>builtin</literal> provider will be
+   selected instead of ICU.
   </para>
 
   <para>
@@ -109,8 +110,7 @@ PostgreSQL documentation
    <literal>--locale-provider=libc</literal>, or build the server without ICU
    support. The <literal>libc</literal> locale provider takes the locale
    settings from the environment, and determines the encoding from the locale
-   settings.  This is almost always sufficient, unless there are special
-   requirements.
+   settings.
   </para>
 
   <para>
@@ -250,6 +250,11 @@ PostgreSQL documentation
         Specifies the ICU locale when the ICU provider is used. Locale support
         is described in <xref linkend="locale"/>.
        </para>
+       <para>
+        If specified as <literal>C</literal> or <literal>POSIX</literal>, the
+        provider is automatically set to <literal>builtin</literal>, as the
+        ICU provider doesn't support an ICU locale of <literal>C</literal>.
+       </para>
        <para>
         If this option is not specified, the locale is inherited from the
         environment in which <command>initdb</command> runs. The environment's
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 4748946499..3b2ba38250 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -252,6 +252,26 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 		if (lcctypeEl)
 			collctype = defGetString(lcctypeEl);
 
+		/*
+		 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+		 * optimizable to byte operations (memcmp(), pg_ascii_tolower(),
+		 * etc.); use the builtin provider instead of ICU if the locale is C.
+		 *
+		 * Don't change the provider during binary upgrade.
+		 */
+		if (!IsBinaryUpgrade && collprovider == COLLPROVIDER_ICU &&
+			colliculocale &&
+			(strcmp(colliculocale, "C") == 0 ||
+			 strcmp(colliculocale, "POSIX") == 0))
+		{
+			ereport(NOTICE,
+					(errmsg("using locale provider \"builtin\" for ICU locale \"%s\"",
+							colliculocale)));
+			builtin_locale = colliculocale;
+			colliculocale = NULL;
+			collprovider = COLLPROVIDER_BUILTIN;
+		}
+
 		if (collprovider == COLLPROVIDER_BUILTIN)
 		{
 			if (!builtin_locale)
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 016852644f..5c8caeac2d 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1043,6 +1043,27 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
 
+	/*
+	 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+	 * optimizable to byte operations (memcmp(), pg_ascii_tolower(), etc.);
+	 * use the builtin provider instead of ICU if the locale is C.
+	 *
+	 * Don't change the provider during binary upgrade or when both the
+	 * provider and ICU locale are unchanged from the template.
+	 */
+	if (!IsBinaryUpgrade && dblocprovider == COLLPROVIDER_ICU &&
+		dbiculocale &&
+		(dblocprovider != src_locprovider ||
+		 strcmp(dbiculocale, src_iculocale) != 0) &&
+		(strcmp(dbiculocale, "C") == 0 || strcmp(dbiculocale, "POSIX") == 0))
+	{
+		ereport(NOTICE,
+				(errmsg("using locale provider \"builtin\" for ICU locale \"%s\"",
+						dbiculocale)));
+		dbiculocale = NULL;
+		dblocprovider = COLLPROVIDER_BUILTIN;
+	}
+
 	if (dblocprovider == COLLPROVIDER_ICU)
 	{
 		if (!(is_encoding_supported_by_icu(encoding)))
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 6fc19c8d64..897521cf35 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2394,6 +2394,23 @@ setlocales(void)
 			lc_messages = locale;
 	}
 
+	/*
+	 * Postgres defines the "C" (and equivalently, "POSIX") locales to be
+	 * optimizable to byte operations (memcmp(), pg_ascii_tolower(), etc.);
+	 * use the builtin provider instead of ICU if the locale is C.
+	 */
+	if (icu_locale && locale_provider == COLLPROVIDER_ICU &&
+		(strcmp(icu_locale, "C") == 0 ||
+		 strcmp(icu_locale, "POSIX") == 0 ||
+		 strncmp(icu_locale, "C.", 2) == 0 ||
+		 strncmp(icu_locale, "POSIX.", 6) == 0))
+	{
+		pg_log_info("using locale provider \"builtin\" for ICU locale \"%s\"",
+					 icu_locale);
+		icu_locale = NULL;
+		locale_provider = COLLPROVIDER_BUILTIN;
+	}
+
 	/*
 	 * canonicalize locale names, and obtain any missing values from our
 	 * current environment
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index 157d6acfd4..f3f832a413 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -111,6 +111,45 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'option --icu-locale');
 
+	# transformed to provider=builtin
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			"$tempdir/data4a"
+		],
+		'option --icu-locale=C');
+
+	# transformed to provider=builtin
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--locale=C',
+			"$tempdir/data4b"
+		],
+		'option --icu-locale=C --locale=C');
+
+	# transformed to provider=builtin
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-collate=C',
+			"$tempdir/data4c"
+		],
+		'option --icu-locale=C --lc-collate=C');
+
+	# transformed to provider=builtin
+	command_ok(
+		[
+			'initdb',                '--no-sync',
+			'--locale-provider=icu', '--icu-locale=C',
+			'--lc-ctype=C',
+			"$tempdir/data4d"
+		],
+		'option --icu-locale=C --lc-ctype=C');
+
 	command_fails_like(
 		[
 			'initdb', '--no-sync',
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index f1d6db0f48..121ad5112b 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -86,6 +86,24 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'create database with icu locale from template database with icu provider'
 	);
+
+	# transformed into provider "builtin"
+	$node->command_ok(
+		[
+			'createdb', '-T', 'template0', '--locale-provider=icu',
+			'--icu-locale=C', 'test_builtin_icu1'
+		],
+		'create database with provider "icu" and ICU_LOCALE="C"'
+	);
+
+	# transformed into provider "builtin"
+	$node->command_ok(
+		[
+			'createdb', '-T', 'template0', '--locale-provider=icu',
+			'--icu-locale=C', '--lc-ctype=C', 'test_builtin_icu2'
+		],
+		'create database with provider "icu" and ICU_LOCALE="C" and LC_CTYPE=C'
+	);
 }
 else
 {
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 00dee24549..5ba8f75558 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1035,6 +1035,9 @@ BEGIN
 END
 $$;
 RESET client_min_messages;
+-- uses "builtin" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+NOTICE:  using locale provider "builtin" for ICU locale "C"
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 ERROR:  parameter "locale" must be specified
 SET icu_validation_level = ERROR;
@@ -1058,8 +1061,11 @@ SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
  test0
  test1
  test5
-(3 rows)
+ testc
+(4 rows)
 
+DROP COLLATION test1;
+CREATE COLLATION test1 (provider = icu, locale = 'und');
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
 ERROR:  collation "test11" already exists in schema "collate_tests"
@@ -1079,7 +1085,8 @@ SELECT collname, nspname, obj_description(pg_collation.oid, 'pg_collation')
  test0    | collate_tests | US English
  test11   | test_schema   | 
  test5    | collate_tests | 
-(3 rows)
+ testc    | collate_tests | 
+(4 rows)
 
 DROP COLLATION test0, test_schema.test11, test5;
 DROP COLLATION test0; -- fail
@@ -1089,7 +1096,8 @@ NOTICE:  collation "test0" does not exist, skipping
 SELECT collname FROM pg_collation WHERE collname LIKE 'test%';
  collname 
 ----------
-(0 rows)
+ testc
+(1 row)
 
 DROP SCHEMA test_schema;
 DROP ROLE regress_test_role;
diff --git a/src/test/regress/sql/collate.icu.utf8.sql b/src/test/regress/sql/collate.icu.utf8.sql
index a8001b4b8e..122c166966 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -375,6 +375,9 @@ $$;
 
 RESET client_min_messages;
 
+-- uses "builtin" provider instead
+CREATE COLLATION testc (provider = icu, locale='C');
+
 CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, needs "locale"
 SET icu_validation_level = ERROR;
 CREATE COLLATION testx (provider = icu, locale = 'nonsense-nowhere'); -- fails
@@ -388,6 +391,9 @@ CREATE COLLATION test5 FROM test0;
 
 SELECT collname FROM pg_collation WHERE collname LIKE 'test%' ORDER BY 1;
 
+DROP COLLATION test1;
+CREATE COLLATION test1 (provider = icu, locale = 'und');
+
 ALTER COLLATION test1 RENAME TO test11;
 ALTER COLLATION test0 RENAME TO test11; -- fail
 ALTER COLLATION test1 RENAME TO test22; -- fail
-- 
2.34.1

From 7ee9bfc7ded5906f5b27390061513bbab3d87555 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v10 3/4] CREATE DATABASE: make LOCALE apply to all collation
 providers.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE. That could
lead to confusion when the provider is implicit, such as when it is
inherited from the template database, or when ICU was made default at
initdb time in commit 27b62377b4.

Reverts incomplete fix 5cd1a5af4d.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_collation.sgml        | 23 +++++---
 doc/src/sgml/ref/create_database.sgml         | 57 +++++++++++++++----
 doc/src/sgml/ref/createdb.sgml                |  5 +-
 doc/src/sgml/ref/initdb.sgml                  |  7 ++-
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 51 +++++++++++++++--
 src/bin/initdb/initdb.c                       | 27 +++++++--
 src/bin/initdb/t/001_initdb.pl                | 28 +++++----
 src/bin/pg_dump/pg_dump.c                     | 55 ++++++++++++++----
 src/bin/scripts/createdb.c                    | 13 ++---
 src/bin/scripts/t/020_createdb.pl             | 28 +++++++--
 src/test/icu/t/010_database.pl                | 22 ++++---
 .../regress/expected/collate.icu.utf8.out     | 22 +++----
 13 files changed, 251 insertions(+), 89 deletions(-)

diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index 67ea247b39..48b62f77cb 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -85,9 +85,16 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-       and <symbol>LC_CTYPE</symbol> at once.  If you specify this,
-       you cannot specify either of those parameters.
+       The locale name for this collation. See <xref
+       linkend="collation-managing-create-libc"/> and <xref
+       linkend="collation-managing-create-icu"/> for details.
+      </para>
+      <para>
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, this
+       is a shortcut for setting <symbol>LC_COLLATE</symbol> and
+       <symbol>LC_CTYPE</symbol> at once.  If you specify
+       <replaceable>locale</replaceable>, you cannot specify either of those
+       parameters.
       </para>
       <para>
        If <replaceable>provider</replaceable> is <literal>builtin</literal>,
@@ -102,8 +109,9 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       Use the specified operating system locale for
-       the <symbol>LC_COLLATE</symbol> locale category.
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, use
+       the specified operating system locale for the
+       <symbol>LC_COLLATE</symbol> locale category.
       </para>
      </listitem>
     </varlistentry>
@@ -113,8 +121,9 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       Use the specified operating system locale for
-       the <symbol>LC_CTYPE</symbol> locale category.
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, use
+       the specified operating system locale for the <symbol>LC_CTYPE</symbol>
+       locale category.
       </para>
      </listitem>
     </varlistentry>
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 17550c19bb..9f20c93267 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,22 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        Sets the default collation order and character classification in the
+        new database.  Collation affects the sort order applied to strings,
+        e.g., in queries with ORDER BY, as well as the order used in indexes
+        on text columns.  Character classification affects the categorization
+        of characters, e.g., lower, - upper and digit.  Also sets the
+        associated aspects of the operating system environment,
+        <literal>LC_COLLATE</literal> and <literal>LC_CTYPE</literal>.  The
+        default is the same setting as the template database.  See <xref
+        linkend="collation-managing-create-libc"/> and <xref
+        linkend="collation-managing-create-icu"/> for details.
+       </para>
+       <para>
+        Can be overridden by setting <xref
+        linkend="create-database-lc-collate"/>, <xref
+        linkend="create-database-lc-ctype"/>, or <xref
+        linkend="create-database-icu-locale"/> individually.
        </para>
        <para>
         If <xref linkend="create-database-locale-provider"/> is
@@ -170,11 +184,17 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">lc_collate</replaceable></term>
       <listitem>
        <para>
-        Collation order (<literal>LC_COLLATE</literal>) to use in the new database.
-        This affects the sort order applied to strings, e.g., in queries with
-        ORDER BY, as well as the order used in indexes on text columns.
-        The default is to use the collation order of the template database.
-        See below for additional restrictions.
+        Sets <literal>LC_COLLATE</literal> in the database server's operating
+        system environment.  The default is the setting of <xref
+        linkend="create-database-locale"/> if specified; otherwise the same
+        setting as the template database.  See below for additional
+        restrictions.
+       </para>
+       <para>
+        If <xref linkend="create-database-locale-provider"/> is
+        <literal>libc</literal>, also sets the default collation order to use
+        in the new database, overriding the setting <xref
+        linkend="create-database-locale"/>.
        </para>
       </listitem>
      </varlistentry>
@@ -182,10 +202,17 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">lc_ctype</replaceable></term>
       <listitem>
        <para>
-        Character classification (<literal>LC_CTYPE</literal>) to use in the new
-        database. This affects the categorization of characters, e.g., lower,
-        upper and digit. The default is to use the character classification of
-        the template database. See below for additional restrictions.
+        Sets <literal>LC_CTYPE</literal> in the database server's operating
+        system environment.  The default is the setting of <xref
+        linkend="create-database-locale"/> if specified; otherwise the same
+        setting as the template database.  See below for additional
+        restrictions.
+       </para>
+       <para>
+        If <xref linkend="create-database-locale-provider"/> is
+        <literal>libc</literal>, also sets the default character
+        classification to use in the new database, overriding the setting
+        <xref linkend="create-database-locale"/>.
        </para>
       </listitem>
      </varlistentry>
@@ -194,7 +221,13 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">icu_locale</replaceable></term>
       <listitem>
        <para>
-        Specifies the ICU locale ID if the ICU locale provider is used.
+        Specifies the ICU locale (see <xref
+        linkend="collation-managing-create-icu"/>) for the database default
+        collation order and character classification, overriding the setting
+        <xref linkend="create-database-locale"/>.  The <xref
+        linkend="create-database-locale-provider"/> must be ICU.  The default
+        is the setting of <xref linkend="create-database-locale"/> if
+        specified; otherwise the same setting as the template database.
        </para>
        <para>
         If specified as <literal>C</literal> or <literal>POSIX</literal>, the
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index 5352cce744..94937e1d6d 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index d9620e5931..5eabaaec3a 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 3b2ba38250..58316ec19e 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -318,7 +318,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 				if (langtag && strcmp(colliculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, colliculocale)));
 
 					colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 5c8caeac2d..3d53e29c7e 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -710,6 +710,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	char	   *dbname = stmt->dbname;
 	char	   *dbowner = NULL;
 	const char *dbtemplate = NULL;
+	char	   *builtin_locale = NULL;
 	char	   *dbcollate = NULL;
 	char	   *dbctype = NULL;
 	char	   *dbiculocale = NULL;
@@ -894,6 +895,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	}
 	if (dlocale && dlocale->arg)
 	{
+		builtin_locale = defGetString(dlocale);
 		dbcollate = defGetString(dlocale);
 		dbctype = defGetString(dlocale);
 	}
@@ -1018,8 +1020,16 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 		dbctype = src_ctype;
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
+	if (builtin_locale == NULL && dblocprovider == COLLPROVIDER_BUILTIN &&
+		src_locprovider == COLLPROVIDER_BUILTIN)
+		builtin_locale = "C";
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1033,12 +1043,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1060,11 +1072,38 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 		ereport(NOTICE,
 				(errmsg("using locale provider \"builtin\" for ICU locale \"%s\"",
 						dbiculocale)));
+		builtin_locale = dbiculocale;
 		dbiculocale = NULL;
 		dblocprovider = COLLPROVIDER_BUILTIN;
 	}
 
-	if (dblocprovider == COLLPROVIDER_ICU)
+	if (dblocprovider == COLLPROVIDER_BUILTIN)
+	{
+		/* can happen if template is a different provider */
+		if (builtin_locale == NULL)
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+					 errmsg("LOCALE must be specified")));
+
+		if (dbiculocale)
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
+					 errmsg("ICU locale cannot be specified unless locale provider is ICU")));
+
+		if (dbicurules)
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
+					 errmsg("ICU rules cannot be specified unless locale provider is ICU")));
+
+		if (strcmp(builtin_locale, "C") != 0 &&
+			strcmp(builtin_locale, "POSIX") != 0)
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("locale provider \"builtin\" does not support locale \"%s\"",
+							builtin_locale),
+					 errhint("The built-in locale provider only supports the \"C\" and \"POSIX\" locales.")));
+	}
+	else if (dblocprovider == COLLPROVIDER_ICU)
 	{
 		if (!(is_encoding_supported_by_icu(encoding)))
 			ereport(ERROR,
@@ -1079,7 +1118,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 		if (!dbiculocale)
 			ereport(ERROR,
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
-					 errmsg("ICU locale must be specified")));
+					 errmsg("LOCALE or ICU_LOCALE must be specified")));
 
 		/*
 		 * During binary upgrade, or when the locale came from the template
@@ -1094,7 +1133,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 			if (langtag && strcmp(dbiculocale, langtag) != 0)
 			{
 				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
+						(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 								langtag, dbiculocale)));
 
 				dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 897521cf35..0a9307df2b 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2163,7 +2163,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2375,8 +2379,9 @@ static void
 setlocales(void)
 {
 	char	   *canonname;
+	char	   *builtin_locale = locale;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2392,6 +2397,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	/*
@@ -2407,6 +2414,7 @@ setlocales(void)
 	{
 		pg_log_info("using locale provider \"builtin\" for ICU locale \"%s\"",
 					 icu_locale);
+		builtin_locale = "C";
 		icu_locale = NULL;
 		locale_provider = COLLPROVIDER_BUILTIN;
 	}
@@ -2434,7 +2442,15 @@ setlocales(void)
 	lc_messages = canonname;
 #endif
 
-	if (locale_provider == COLLPROVIDER_ICU)
+	if (locale_provider == COLLPROVIDER_BUILTIN)
+	{
+		if (!builtin_locale)
+			pg_fatal("The parameter --locale must be specified for the built-in provider.");
+		if (strcmp(builtin_locale, "C") != 0 &&
+			strcmp(builtin_locale, "POSIX") != 0)
+			pg_fatal("The built-in locale provider only supports the \"C\" and \"POSIX\" locales.");
+	}
+	else if (locale_provider == COLLPROVIDER_ICU)
 	{
 		char	   *langtag;
 
@@ -2687,7 +2703,9 @@ setup_locale_encoding(void)
 		 * If ctype_enc=SQL_ASCII, it's compatible with any encoding. ICU does
 		 * not support SQL_ASCII, so select UTF-8 instead.
 		 */
-		if (locale_provider == COLLPROVIDER_ICU && ctype_enc == PG_SQL_ASCII)
+		if ((locale_provider == COLLPROVIDER_BUILTIN ||
+			 locale_provider == COLLPROVIDER_ICU)
+			&& ctype_enc == PG_SQL_ASCII)
 			ctype_enc = PG_UTF8;
 
 		if (ctype_enc == -1)
@@ -3294,7 +3312,6 @@ main(int argc, char *argv[])
 				break;
 			case 8:
 				locale = "C";
-				locale_provider = COLLPROVIDER_LIBC;
 				break;
 			case 9:
 				pwfilename = pg_strdup(optarg);
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index f3f832a413..9b0c2202a3 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -193,11 +193,19 @@ else
 		'locale provider ICU fails since no ICU support');
 }
 
-command_ok(
+command_fails(
 	[
 		'initdb', '--no-sync', '--locale-provider=builtin', "$tempdir/data6"
 	],
-	'locale provider builtin'
+	'locale provider builtin without --locale'
+);
+
+command_fails(
+	[
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=en',
+		"$tempdir/data6"
+	],
+	'locale provider builtin with invalid --locale'
 );
 
 command_ok(
@@ -210,32 +218,32 @@ command_ok(
 
 command_ok(
 	[
-		'initdb', '--no-sync', '--locale-provider=builtin', '--lc-collate=C',
-		"$tempdir/data8"
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=C',
+		'--lc-collate=C', "$tempdir/data8"
 	],
 	'locale provider builtin with --lc-collate'
 );
 
 command_ok(
 	[
-		'initdb', '--no-sync', '--locale-provider=builtin', '--lc-ctype=C',
-		"$tempdir/data9"
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=C',
+		'--lc-ctype=C', "$tempdir/data9"
 	],
 	'locale provider builtin with --lc-ctype'
 );
 
 command_fails(
 	[
-		'initdb', '--no-sync', '--locale-provider=builtin', '--icu-locale=en',
-		"$tempdir/dataX"
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=C',
+		'--icu-locale=en', "$tempdir/dataX"
 	],
 	'fails for locale provider builtin with ICU locale'
 );
 
 command_fails(
 	[
-		'initdb', '--no-sync', '--locale-provider=builtin', '--icu-rules=""',
-		"$tempdir/dataX"
+		'initdb', '--no-sync', '--locale-provider=builtin', '--locale=C',
+		'--icu-rules=""', "$tempdir/dataX"
 	],
 	'fails for locale provider builtin with ICU rules'
 );
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index c75818c3a4..8dbbd87610 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -3080,13 +3080,10 @@ dumpDatabase(Archive *fout)
 		pg_fatal("unrecognized locale provider: %s",
 				 datlocprovider);
 
-	if (strlen(collate) > 0 && strcmp(collate, ctype) == 0)
-	{
-		appendPQExpBufferStr(creaQry, " LOCALE = ");
-		appendStringLiteralAH(creaQry, collate, fout);
-	}
-	else
+	if (datlocprovider[0] == 'b')
 	{
+		appendPQExpBufferStr(creaQry, " LOCALE = 'C'");
+
 		if (strlen(collate) > 0)
 		{
 			appendPQExpBufferStr(creaQry, " LC_COLLATE = ");
@@ -3098,15 +3095,51 @@ dumpDatabase(Archive *fout)
 			appendStringLiteralAH(creaQry, ctype, fout);
 		}
 	}
-	if (iculocale)
+	else if (datlocprovider[0] == 'i')
 	{
-		appendPQExpBufferStr(creaQry, " ICU_LOCALE = ");
+		Assert(iculocale);
+
+		appendPQExpBufferStr(creaQry, " LOCALE = ");
 		appendStringLiteralAH(creaQry, iculocale, fout);
+
+		if (strlen(collate) > 0)
+		{
+			appendPQExpBufferStr(creaQry, " LC_COLLATE = ");
+			appendStringLiteralAH(creaQry, collate, fout);
+		}
+		if (strlen(ctype) > 0)
+		{
+			appendPQExpBufferStr(creaQry, " LC_CTYPE = ");
+			appendStringLiteralAH(creaQry, ctype, fout);
+		}
+		if (icurules)
+		{
+			appendPQExpBufferStr(creaQry, " ICU_RULES = ");
+			appendStringLiteralAH(creaQry, icurules, fout);
+		}
 	}
-	if (icurules)
+	else if (datlocprovider[0] == 'c')
 	{
-		appendPQExpBufferStr(creaQry, " ICU_RULES = ");
-		appendStringLiteralAH(creaQry, icurules, fout);
+		Assert(!iculocale);
+
+		if (strlen(collate) > 0 && strcmp(collate, ctype) == 0)
+		{
+			appendPQExpBufferStr(creaQry, " LOCALE = ");
+			appendStringLiteralAH(creaQry, collate, fout);
+		}
+		else
+		{
+			if (strlen(collate) > 0)
+			{
+				appendPQExpBufferStr(creaQry, " LC_COLLATE = ");
+				appendStringLiteralAH(creaQry, collate, fout);
+			}
+			if (strlen(ctype) > 0)
+			{
+				appendPQExpBufferStr(creaQry, " LC_CTYPE = ");
+				appendStringLiteralAH(creaQry, ctype, fout);
+			}
+		}
 	}
 
 	/*
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index 41a8de659e..8f8995964c 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
@@ -219,6 +211,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index 121ad5112b..3f5719c913 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -115,7 +115,7 @@ else
 $node->command_ok(
 	[
 		'createdb', '-T', 'template0', '--locale-provider=builtin',
-		'tbuiltin1'
+		'--locale=C', 'tbuiltin1'
 	],
 	'create database with provider "builtin"'
 );
@@ -131,7 +131,7 @@ $node->command_ok(
 $node->command_ok(
 	[
 		'createdb', '-T', 'template0', '--locale-provider=builtin',
-		'--lc-collate=C', 'tbuiltin3'
+		'--locale=C', '--lc-collate=C', 'tbuiltin3'
 	],
 	'create database with provider "builtin" and LC_COLLATE=C'
 );
@@ -139,7 +139,7 @@ $node->command_ok(
 $node->command_ok(
 	[
 		'createdb', '-T', 'template0', '--locale-provider=builtin',
-		'--lc-ctype=C', 'tbuiltin4'
+		'--locale=C', '--lc-ctype=C', 'tbuiltin4'
 	],
 	'create database with provider "builtin" and LC_CTYPE=C'
 );
@@ -168,6 +168,22 @@ $node->command_fails(
 	'create database with provider "builtin" not matching template'
 );
 
+$node->command_fails(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'tbuiltin8'
+	],
+	'create database with provider "builtin" and locale unspecified'
+);
+
+$node->command_fails(
+	[
+		'createdb', '-T', 'template0', '--locale-provider=builtin',
+		'--locale=en', 'tbuiltin8'
+	],
+	'create database with provider "builtin" and locale=en'
+);
+
 $node->command_fails([ 'createdb', 'foobar1' ],
 	'fails if database already exists');
 
@@ -184,7 +200,7 @@ ALTER TABLE tab_foobar owner to role_foobar;
 CREATE POLICY pol_foobar ON tab_foobar FOR ALL TO role_foobar;');
 $node->issues_sql_like(
 	[ 'createdb', '-l', 'C', '-T', 'foobar2', 'foobar3' ],
-	qr/statement: CREATE DATABASE foobar3 TEMPLATE foobar2/,
+	qr/statement: CREATE DATABASE foobar3 TEMPLATE foobar2 LOCALE 'C'/,
 	'create database with template');
 ($ret, $stdout, $stderr) = $node->psql(
 	'foobar3',
@@ -211,7 +227,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -219,7 +235,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index 26f71e1155..6d19488afc 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,17 +51,23 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
-
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+# Test that LOCALE='C' works for ICU
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
 );
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+is($ret1, 0,
+	"C locale works for ICU");
+
+# Test that ICU-specific locale string must be specified with ICU_LOCALE,
+# not LOCALE
+my ($ret2, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary' TEMPLATE template0 ENCODING UTF8});
+isnt($ret2, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 my ($ret, $stdout, $stderr) = $node1->psql('postgres',
 	q{CREATE DATABASE dbicu LOCALE_PROVIDER builtin LOCALE 'C' TEMPLATE dbicu}
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 5ba8f75558..9e49c0d9a7 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1202,9 +1202,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1212,7 +1212,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1229,12 +1229,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1243,7 +1243,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1271,13 +1271,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1327,9 +1327,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From ea72d9f8baae263af0f7ce1b80bf624e0496a571 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 24 May 2023 09:53:02 -0700
Subject: [PATCH v10 4/4] CREATE COLLATION default provider.

If the LC_COLLATE or LC_CTYPE are specified for a new collation, the
default provider is libc. Otherwise, the default provider is the same
as the provider for the database default collation.

Previously, the default provider was always libc.
---
 contrib/citext/expected/create_index_acl.out     |  2 +-
 contrib/citext/sql/create_index_acl.sql          |  2 +-
 doc/src/sgml/ref/create_collation.sgml           | 12 +++++++++---
 src/backend/commands/collationcmds.c             |  7 ++++++-
 src/test/regress/expected/collate.linux.utf8.out | 10 +++++-----
 src/test/regress/sql/collate.linux.utf8.sql      | 10 +++++-----
 6 files changed, 27 insertions(+), 16 deletions(-)

diff --git a/contrib/citext/expected/create_index_acl.out b/contrib/citext/expected/create_index_acl.out
index 33be13a92d..765867d36d 100644
--- a/contrib/citext/expected/create_index_acl.out
+++ b/contrib/citext/expected/create_index_acl.out
@@ -43,7 +43,7 @@ REVOKE ALL ON FUNCTION s.index_row_if FROM PUBLIC;
 -- Even for an empty table, CREATE INDEX checks ii_Predicate permissions.
 GRANT EXECUTE ON FUNCTION s.index_row_if TO regress_minimal;
 -- Non-extension, non-function objects.
-CREATE COLLATION s.coll (LOCALE="C");
+CREATE COLLATION s.coll (PROVIDER=libc, LOCALE="C");
 CREATE TABLE s.x (y s.citext);
 ALTER TABLE s.x OWNER TO regress_minimal;
 -- Empty-table DefineIndex()
diff --git a/contrib/citext/sql/create_index_acl.sql b/contrib/citext/sql/create_index_acl.sql
index 10b5225569..e338ac8799 100644
--- a/contrib/citext/sql/create_index_acl.sql
+++ b/contrib/citext/sql/create_index_acl.sql
@@ -45,7 +45,7 @@ REVOKE ALL ON FUNCTION s.index_row_if FROM PUBLIC;
 -- Even for an empty table, CREATE INDEX checks ii_Predicate permissions.
 GRANT EXECUTE ON FUNCTION s.index_row_if TO regress_minimal;
 -- Non-extension, non-function objects.
-CREATE COLLATION s.coll (LOCALE="C");
+CREATE COLLATION s.coll (PROVIDER=libc, LOCALE="C");
 CREATE TABLE s.x (y s.citext);
 ALTER TABLE s.x OWNER TO regress_minimal;
 -- Empty-table DefineIndex()
diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index 48b62f77cb..2b25eff44e 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -136,9 +136,15 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
        Specifies the provider to use for locale services associated with this
        collation.  Possible values are <literal>builtin</literal>,
        <literal>icu</literal><indexterm><primary>ICU</primary></indexterm> (if
-       the server was built with ICU support) or <literal>libc</literal>.
-       <literal>libc</literal> is the default.  See <xref
-       linkend="locale-providers"/> for details.
+       the server was built with ICU support) or <literal>libc</literal>. See
+       <xref linkend="locale-providers"/> for details.
+      </para>
+      <para>
+       If <replaceable>lc_colllate</replaceable> or
+       <replaceable>lc_ctype</replaceable> is specified, the default is
+       <literal>libc</literal>; otherwise, the default is the same as the
+       provider for the database default collation (see <xref
+       linkend="sql-createdatabase"/>).
       </para>
       <para>
        If the provider is <literal>icu</literal> and the locale is
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 58316ec19e..f2264000cb 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -229,7 +229,12 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 								collproviderstr)));
 		}
 		else
-			collprovider = COLLPROVIDER_LIBC;
+		{
+			if (lccollateEl || lcctypeEl)
+				collprovider = COLLPROVIDER_LIBC;
+			else
+				collprovider = default_locale.provider;
+		}
 
 		if (localeEl)
 		{
diff --git a/src/test/regress/expected/collate.linux.utf8.out b/src/test/regress/expected/collate.linux.utf8.out
index 01664f7c1b..588198d13e 100644
--- a/src/test/regress/expected/collate.linux.utf8.out
+++ b/src/test/regress/expected/collate.linux.utf8.out
@@ -1026,7 +1026,7 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
@@ -1034,7 +1034,7 @@ CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 ERROR:  collation "test0" already exists
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
 NOTICE:  collation "test0" already exists, skipping
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 NOTICE:  collation "test0" for encoding "UTF8" already exists, skipping
 do $$
 BEGIN
@@ -1046,7 +1046,7 @@ END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
 ERROR:  parameter "lc_ctype" must be specified
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 ERROR:  could not create locale "nonsense": No such file or directory
 DETAIL:  The operating system could not find any locale data for the locale name "nonsense".
 CREATE COLLATION test4 FROM nonsense;
@@ -1166,8 +1166,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 
 -- nondeterministic collations
 -- (not supported with libc provider)
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 ERROR:  nondeterministic collations not supported with this provider
 -- cleanup
 SET client_min_messages TO warning;
diff --git a/src/test/regress/sql/collate.linux.utf8.sql b/src/test/regress/sql/collate.linux.utf8.sql
index 132d13af0a..2d031293d1 100644
--- a/src/test/regress/sql/collate.linux.utf8.sql
+++ b/src/test/regress/sql/collate.linux.utf8.sql
@@ -358,13 +358,13 @@ CREATE SCHEMA test_schema;
 -- We need to do this this way to cope with varying names for encodings:
 do $$
 BEGIN
-  EXECUTE 'CREATE COLLATION test0 (locale = ' ||
+  EXECUTE 'CREATE COLLATION test0 (provider = libc, locale = ' ||
           quote_literal((SELECT datcollate FROM pg_database WHERE datname = current_database())) || ');';
 END
 $$;
 CREATE COLLATION test0 FROM "C"; -- fail, duplicate name
 CREATE COLLATION IF NOT EXISTS test0 FROM "C"; -- ok, skipped
-CREATE COLLATION IF NOT EXISTS test0 (locale = 'foo'); -- ok, skipped
+CREATE COLLATION IF NOT EXISTS test0 (provider = libc, locale = 'foo'); -- ok, skipped
 do $$
 BEGIN
   EXECUTE 'CREATE COLLATION test1 (lc_collate = ' ||
@@ -374,7 +374,7 @@ BEGIN
 END
 $$;
 CREATE COLLATION test3 (lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
-CREATE COLLATION testx (locale = 'nonsense'); -- fail
+CREATE COLLATION testx (provider = libc, locale = 'nonsense'); -- fail
 
 CREATE COLLATION test4 FROM nonsense;
 CREATE COLLATION test5 FROM test0;
@@ -455,8 +455,8 @@ SELECT * FROM collate_test2 ORDER BY b COLLATE UCS_BASIC;
 -- nondeterministic collations
 -- (not supported with libc provider)
 
-CREATE COLLATION ctest_det (locale = 'en_US.utf8', deterministic = true);
-CREATE COLLATION ctest_nondet (locale = 'en_US.utf8', deterministic = false);
+CREATE COLLATION ctest_det (provider = libc, locale = 'en_US.utf8', deterministic = true);
+CREATE COLLATION ctest_nondet (provider = libc, locale = 'en_US.utf8', deterministic = false);
 
 
 -- cleanup
-- 
2.34.1

From 4c474b5f0af86611480143dcf89918f9329ba512 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Tue, 25 Apr 2023 15:01:55 -0700
Subject: [PATCH v11] CREATE DATABASE: make LOCALE apply to all collation
 providers.

Previously, LOCALE applied only to LC_COLLATE and LC_CTYPE. That could
lead to confusion when the provider is ICU.

Discussion: https://postgr.es/m/3391932.1682107209@sss.pgh.pa.us
---
 doc/src/sgml/ref/create_collation.sgml        | 23 +++++---
 doc/src/sgml/ref/create_database.sgml         | 57 +++++++++++++++----
 doc/src/sgml/ref/createdb.sgml                |  5 +-
 doc/src/sgml/ref/initdb.sgml                  |  7 ++-
 src/backend/commands/collationcmds.c          |  2 +-
 src/backend/commands/dbcommands.c             | 17 ++++--
 src/bin/initdb/initdb.c                       | 10 +++-
 src/bin/initdb/t/001_initdb.pl                | 11 ++++
 src/bin/scripts/createdb.c                    | 13 ++---
 src/bin/scripts/t/020_createdb.pl             | 15 ++++-
 src/test/icu/t/010_database.pl                | 30 +++++++---
 .../regress/expected/collate.icu.utf8.out     | 22 +++----
 12 files changed, 152 insertions(+), 60 deletions(-)

diff --git a/doc/src/sgml/ref/create_collation.sgml b/doc/src/sgml/ref/create_collation.sgml
index f6353da5c1..b86a9bbb9c 100644
--- a/doc/src/sgml/ref/create_collation.sgml
+++ b/doc/src/sgml/ref/create_collation.sgml
@@ -85,9 +85,16 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-       and <symbol>LC_CTYPE</symbol> at once.  If you specify this,
-       you cannot specify either of those parameters.
+       The locale name for this collation. See <xref
+       linkend="collation-managing-create-libc"/> and <xref
+       linkend="collation-managing-create-icu"/> for details.
+      </para>
+      <para>
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, this
+       is a shortcut for setting <symbol>LC_COLLATE</symbol> and
+       <symbol>LC_CTYPE</symbol> at once.  If you specify
+       <replaceable>locale</replaceable>, you cannot specify either of those
+       parameters.
       </para>
      </listitem>
     </varlistentry>
@@ -97,8 +104,9 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       Use the specified operating system locale for
-       the <symbol>LC_COLLATE</symbol> locale category.
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, use
+       the specified operating system locale for the
+       <symbol>LC_COLLATE</symbol> locale category.
       </para>
      </listitem>
     </varlistentry>
@@ -108,8 +116,9 @@ CREATE COLLATION [ IF NOT EXISTS ] <replaceable>name</replaceable> FROM <replace
 
      <listitem>
       <para>
-       Use the specified operating system locale for
-       the <symbol>LC_CTYPE</symbol> locale category.
+       If <replaceable>provider</replaceable> is <literal>libc</literal>, use
+       the specified operating system locale for the <symbol>LC_CTYPE</symbol>
+       locale category.
       </para>
      </listitem>
     </varlistentry>
diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index 13793bb6b7..dab05950ed 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -145,8 +145,22 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">locale</replaceable></term>
       <listitem>
        <para>
-        This is a shortcut for setting <symbol>LC_COLLATE</symbol>
-        and <symbol>LC_CTYPE</symbol> at once.
+        Sets the default collation order and character classification in the
+        new database.  Collation affects the sort order applied to strings,
+        e.g., in queries with ORDER BY, as well as the order used in indexes
+        on text columns.  Character classification affects the categorization
+        of characters, e.g., lower, - upper and digit.  Also sets the
+        associated aspects of the operating system environment,
+        <literal>LC_COLLATE</literal> and <literal>LC_CTYPE</literal>.  The
+        default is the same setting as the template database.  See <xref
+        linkend="collation-managing-create-libc"/> and <xref
+        linkend="collation-managing-create-icu"/> for details.
+       </para>
+       <para>
+        Can be overridden by setting <xref
+        linkend="create-database-lc-collate"/>, <xref
+        linkend="create-database-lc-ctype"/>, or <xref
+        linkend="create-database-icu-locale"/> individually.
        </para>
        <tip>
         <para>
@@ -164,11 +178,17 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">lc_collate</replaceable></term>
       <listitem>
        <para>
-        Collation order (<literal>LC_COLLATE</literal>) to use in the new database.
-        This affects the sort order applied to strings, e.g., in queries with
-        ORDER BY, as well as the order used in indexes on text columns.
-        The default is to use the collation order of the template database.
-        See below for additional restrictions.
+        Sets <literal>LC_COLLATE</literal> in the database server's operating
+        system environment.  The default is the setting of <xref
+        linkend="create-database-locale"/> if specified; otherwise the same
+        setting as the template database.  See below for additional
+        restrictions.
+       </para>
+       <para>
+        If <xref linkend="create-database-locale-provider"/> is
+        <literal>libc</literal>, also sets the default collation order to use
+        in the new database, overriding the setting <xref
+        linkend="create-database-locale"/>.
        </para>
       </listitem>
      </varlistentry>
@@ -176,10 +196,17 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">lc_ctype</replaceable></term>
       <listitem>
        <para>
-        Character classification (<literal>LC_CTYPE</literal>) to use in the new
-        database. This affects the categorization of characters, e.g., lower,
-        upper and digit. The default is to use the character classification of
-        the template database. See below for additional restrictions.
+        Sets <literal>LC_CTYPE</literal> in the database server's operating
+        system environment.  The default is the setting of <xref
+        linkend="create-database-locale"/> if specified; otherwise the same
+        setting as the template database.  See below for additional
+        restrictions.
+       </para>
+       <para>
+        If <xref linkend="create-database-locale-provider"/> is
+        <literal>libc</literal>, also sets the default character
+        classification to use in the new database, overriding the setting
+        <xref linkend="create-database-locale"/>.
        </para>
       </listitem>
      </varlistentry>
@@ -188,7 +215,13 @@ CREATE DATABASE <replaceable class="parameter">name</replaceable>
       <term><replaceable class="parameter">icu_locale</replaceable></term>
       <listitem>
        <para>
-        Specifies the ICU locale ID if the ICU locale provider is used.
+        Specifies the ICU locale (see <xref
+        linkend="collation-managing-create-icu"/>) for the database default
+        collation order and character classification, overriding the setting
+        <xref linkend="create-database-locale"/>.  The <xref
+        linkend="create-database-locale-provider"/> must be ICU.  The default
+        is the setting of <xref linkend="create-database-locale"/> if
+        specified; otherwise the same setting as the template database.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/createdb.sgml b/doc/src/sgml/ref/createdb.sgml
index e23419ba6c..e4647d5ce7 100644
--- a/doc/src/sgml/ref/createdb.sgml
+++ b/doc/src/sgml/ref/createdb.sgml
@@ -124,7 +124,10 @@ PostgreSQL documentation
       <listitem>
        <para>
         Specifies the locale to be used in this database.  This is equivalent
-        to specifying both <option>--lc-collate</option> and <option>--lc-ctype</option>.
+        to specifying <option>--lc-collate</option>,
+        <option>--lc-ctype</option>, and <option>--icu-locale</option> to the
+        same value. Some locales are only valid for ICU and must be set with
+        <option>--icu-locale</option>.
        </para>
       </listitem>
      </varlistentry>
diff --git a/doc/src/sgml/ref/initdb.sgml b/doc/src/sgml/ref/initdb.sgml
index 87945b4b62..f850dc404d 100644
--- a/doc/src/sgml/ref/initdb.sgml
+++ b/doc/src/sgml/ref/initdb.sgml
@@ -116,9 +116,10 @@ PostgreSQL documentation
   <para>
    To choose a different locale for the cluster, use the option
    <option>--locale</option>.  There are also individual options
-   <option>--lc-*</option> (see below) to set values for the individual locale
-   categories.  Note that inconsistent settings for different locale
-   categories can give nonsensical results, so this should be used with care.
+   <option>--lc-*</option> and <option>--icu-locale</option> (see below) to
+   set values for the individual locale categories.  Note that inconsistent
+   settings for different locale categories can give nonsensical results, so
+   this should be used with care.
   </para>
 
   <para>
diff --git a/src/backend/commands/collationcmds.c b/src/backend/commands/collationcmds.c
index 2969a2bb21..efb8b4d289 100644
--- a/src/backend/commands/collationcmds.c
+++ b/src/backend/commands/collationcmds.c
@@ -276,7 +276,7 @@ DefineCollation(ParseState *pstate, List *names, List *parameters, bool if_not_e
 				if (langtag && strcmp(colliculocale, langtag) != 0)
 				{
 					ereport(NOTICE,
-							(errmsg("using standard form \"%s\" for locale \"%s\"",
+							(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 									langtag, colliculocale)));
 
 					colliculocale = langtag;
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 99d4080ea9..09f1ab41ad 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1017,7 +1017,12 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (dblocprovider == '\0')
 		dblocprovider = src_locprovider;
 	if (dbiculocale == NULL && dblocprovider == COLLPROVIDER_ICU)
-		dbiculocale = src_iculocale;
+	{
+		if (dlocale && dlocale->arg)
+			dbiculocale = defGetString(dlocale);
+		else
+			dbiculocale = src_iculocale;
+	}
 	if (dbicurules == NULL && dblocprovider == COLLPROVIDER_ICU)
 		dbicurules = src_icurules;
 
@@ -1031,12 +1036,14 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	if (!check_locale(LC_COLLATE, dbcollate, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbcollate)));
+				 errmsg("invalid LC_COLLATE locale name: \"%s\"", dbcollate),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbcollate = canonname;
 	if (!check_locale(LC_CTYPE, dbctype, &canonname))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
-				 errmsg("invalid locale name: \"%s\"", dbctype)));
+				 errmsg("invalid LC_CTYPE locale name: \"%s\"", dbctype),
+				 errhint("If the locale name is specific to ICU, use ICU_LOCALE.")));
 	dbctype = canonname;
 
 	check_encoding_locale_matches(encoding, dbcollate, dbctype);
@@ -1056,7 +1063,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 		if (!dbiculocale)
 			ereport(ERROR,
 					(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
-					 errmsg("ICU locale must be specified")));
+					 errmsg("LOCALE or ICU_LOCALE must be specified")));
 
 		/*
 		 * During binary upgrade, or when the locale came from the template
@@ -1071,7 +1078,7 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 			if (langtag && strcmp(dbiculocale, langtag) != 0)
 			{
 				ereport(NOTICE,
-						(errmsg("using standard form \"%s\" for locale \"%s\"",
+						(errmsg("using standard form \"%s\" for ICU locale \"%s\"",
 								langtag, dbiculocale)));
 
 				dbiculocale = langtag;
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index 09a5c98cc0..71a3d26c37 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2163,7 +2163,11 @@ check_locale_name(int category, const char *locale, char **canonname)
 	if (res == NULL)
 	{
 		if (*locale)
-			pg_fatal("invalid locale name \"%s\"", locale);
+		{
+			pg_log_error("invalid locale name \"%s\"", locale);
+			pg_log_error_hint("If the locale name is specific to ICU, use --icu-locale.");
+			exit(1);
+		}
 		else
 		{
 			/*
@@ -2376,7 +2380,7 @@ setlocales(void)
 {
 	char	   *canonname;
 
-	/* set empty lc_* values to locale config if set */
+	/* set empty lc_* and iculocale values to locale config if set */
 
 	if (locale)
 	{
@@ -2392,6 +2396,8 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
+		if (!icu_locale && locale_provider == COLLPROVIDER_ICU)
+			icu_locale = locale;
 	}
 
 	/*
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index fa00bb3dab..cf55a84cd1 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -111,6 +111,17 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'option --icu-locale');
 
+	command_like(
+		[
+			'initdb', '--no-sync', '-A', 'trust',
+			'--locale-provider=icu', '--locale=und',
+			'--lc-collate=C', '--lc-ctype=C', '--lc-messages=C',
+			'--lc-numeric=C', '--lc-monetary=C', '--lc-time=C',
+			"$tempdir/data4"
+		],
+		qr/^\s+ICU locale:\s+und\n/ms,
+		'options --locale-provider=icu --locale=und --lc-*=C');
+
 	command_fails_like(
 		[
 			'initdb', '--no-sync',
diff --git a/src/bin/scripts/createdb.c b/src/bin/scripts/createdb.c
index b4205c4fa5..9ca86a3e53 100644
--- a/src/bin/scripts/createdb.c
+++ b/src/bin/scripts/createdb.c
@@ -164,14 +164,6 @@ main(int argc, char *argv[])
 			exit(1);
 	}
 
-	if (locale)
-	{
-		if (!lc_ctype)
-			lc_ctype = locale;
-		if (!lc_collate)
-			lc_collate = locale;
-	}
-
 	if (encoding)
 	{
 		if (pg_char_to_encoding(encoding) < 0)
@@ -219,6 +211,11 @@ main(int argc, char *argv[])
 		appendPQExpBuffer(&sql, " STRATEGY %s", fmtId(strategy));
 	if (template)
 		appendPQExpBuffer(&sql, " TEMPLATE %s", fmtId(template));
+	if (locale)
+	{
+		appendPQExpBufferStr(&sql, " LOCALE ");
+		appendStringLiteralConn(&sql, locale, conn);
+	}
 	if (lc_collate)
 	{
 		appendPQExpBufferStr(&sql, " LC_COLLATE ");
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index d0830a4a1d..694ec56804 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -86,6 +86,15 @@ if ($ENV{with_icu} eq 'yes')
 		],
 		'create database with icu locale from template database with icu provider'
 	);
+
+	$node2->command_ok(
+		[
+			'createdb', '-T', 'template0', '--locale-provider', 'icu',
+			'--locale', 'en', '--lc-collate', 'C', '--lc-ctype', 'C',
+			'foobar57'
+		],
+		'create database with locale as ICU locale'
+	);
 }
 else
 {
@@ -110,7 +119,7 @@ ALTER TABLE tab_foobar owner to role_foobar;
 CREATE POLICY pol_foobar ON tab_foobar FOR ALL TO role_foobar;');
 $node->issues_sql_like(
 	[ 'createdb', '-l', 'C', '-T', 'foobar2', 'foobar3' ],
-	qr/statement: CREATE DATABASE foobar3 TEMPLATE foobar2/,
+	qr/statement: CREATE DATABASE foobar3 TEMPLATE foobar2 LOCALE 'C'/,
 	'create database with template');
 ($ret, $stdout, $stderr) = $node->psql(
 	'foobar3',
@@ -137,7 +146,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_COLLATE locale name|^createdb: error: database creation failed: ERROR:  new collation \(foo'; SELECT '1\) is incompatible with the collation of the template database/s
 	],
 	'createdb with incorrect --lc-collate');
 $node->command_checks_all(
@@ -145,7 +154,7 @@ $node->command_checks_all(
 	1,
 	[qr/^$/],
 	[
-		qr/^createdb: error: database creation failed: ERROR:  invalid locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
+		qr/^createdb: error: database creation failed: ERROR:  invalid LC_CTYPE locale name|^createdb: error: database creation failed: ERROR:  new LC_CTYPE \(foo'; SELECT '1\) is incompatible with the LC_CTYPE of the template database/s
 	],
 	'createdb with incorrect --lc-ctype');
 
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index d3901f5d3f..b417b1a409 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -51,17 +51,33 @@ b),
 	'sort by explicit collation upper first');
 
 
-# Test error cases in CREATE DATABASE involving locale-related options
+# Test that LOCALE='C' works for ICU
+my $ret1 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu1 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+);
+is($ret1, 0,
+	"C locale works for ICU");
 
-my ($ret, $stdout, $stderr) = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+# Test that LOCALE works for ICU locales if LC_COLLATE and LC_CTYPE
+# are specified
+my $ret2 = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE '@colStrength=primary'
+      LC_COLLATE='C' LC_CTYPE='C' TEMPLATE template0 ENCODING UTF8}
 );
-isnt($ret, 0,
-	"ICU locale must be specified for ICU provider: exit code not 0");
+is($ret2, 0,
+	"LOCALE works for ICU locales if LC_COLLATE and LC_CTYPE are specified");
+
+# Test that ICU-specific LOCALE without LC_COLLATE and LC_CTYPE must
+# be specified with ICU_LOCALE
+my ($ret3, $stdout, $stderr) = $node1->psql('postgres',
+	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary'
+      TEMPLATE template0 ENCODING UTF8});
+isnt($ret3, 0,
+	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-	qr/ERROR:  ICU locale must be specified/,
-	"ICU locale must be specified for ICU provider: error message");
+	qr/ERROR:  invalid LC_COLLATE locale name/,
+	"ICU-specific locale must be specified with ICU_LOCALE: error message");
 
 
 done_testing();
diff --git a/src/test/regress/expected/collate.icu.utf8.out b/src/test/regress/expected/collate.icu.utf8.out
index 00dee24549..dc96e590f7 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1194,9 +1194,9 @@ SELECT 'coté' < 'côte' COLLATE "und-x-icu", 'coté' > 'côte' COLLATE testcoll
 (1 row)
 
 CREATE COLLATION testcoll_lower_first (provider = icu, locale = '@colCaseFirst=lower');
-NOTICE:  using standard form "und-u-kf-lower" for locale "@colCaseFirst=lower"
+NOTICE:  using standard form "und-u-kf-lower" for ICU locale "@colCaseFirst=lower"
 CREATE COLLATION testcoll_upper_first (provider = icu, locale = '@colCaseFirst=upper');
-NOTICE:  using standard form "und-u-kf-upper" for locale "@colCaseFirst=upper"
+NOTICE:  using standard form "und-u-kf-upper" for ICU locale "@colCaseFirst=upper"
 SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcoll_upper_first;
  ?column? | ?column? 
 ----------+----------
@@ -1204,7 +1204,7 @@ SELECT 'aaa' < 'AAA' COLLATE testcoll_lower_first, 'aaa' > 'AAA' COLLATE testcol
 (1 row)
 
 CREATE COLLATION testcoll_shifted (provider = icu, locale = '@colAlternate=shifted');
-NOTICE:  using standard form "und-u-ka-shifted" for locale "@colAlternate=shifted"
+NOTICE:  using standard form "und-u-ka-shifted" for ICU locale "@colAlternate=shifted"
 SELECT 'de-luge' < 'deanza' COLLATE "und-x-icu", 'de-luge' > 'deanza' COLLATE testcoll_shifted;
  ?column? | ?column? 
 ----------+----------
@@ -1221,12 +1221,12 @@ SELECT 'A-21' > 'A-123' COLLATE "und-x-icu", 'A-21' < 'A-123' COLLATE testcoll_n
 (1 row)
 
 CREATE COLLATION testcoll_error1 (provider = icu, locale = '@colNumeric=lower');
-NOTICE:  using standard form "und-u-kn-lower" for locale "@colNumeric=lower"
+NOTICE:  using standard form "und-u-kn-lower" for ICU locale "@colNumeric=lower"
 ERROR:  could not open collator for locale "und-u-kn-lower": U_ILLEGAL_ARGUMENT_ERROR
 -- test that attributes not handled by icu_set_collation_attributes()
 -- (handled by ucol_open() directly) also work
 CREATE COLLATION testcoll_de_phonebook (provider = icu, locale = 'de@collation=phonebook');
-NOTICE:  using standard form "de-u-co-phonebk" for locale "de@collation=phonebook"
+NOTICE:  using standard form "de-u-co-phonebk" for ICU locale "de@collation=phonebook"
 SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE testcoll_de_phonebook;
  ?column? | ?column? 
 ----------+----------
@@ -1235,7 +1235,7 @@ SELECT 'Goldmann' < 'Götz' COLLATE "de-x-icu", 'Goldmann' > 'Götz' COLLATE tes
 
 -- rules
 CREATE COLLATION testcoll_rules1 (provider = icu, locale = '', rules = '&a < g');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test7 (a text);
 -- example from https://unicode-org.github.io/icu/userguide/collation/customization/#syntax
 INSERT INTO test7 VALUES ('Abernathy'), ('apple'), ('bird'), ('Boston'), ('Graham'), ('green');
@@ -1263,13 +1263,13 @@ SELECT * FROM test7 ORDER BY a COLLATE testcoll_rules1;
 
 DROP TABLE test7;
 CREATE COLLATION testcoll_rulesx (provider = icu, locale = '', rules = '!!wrong!!');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 ERROR:  could not open collator for locale "und" with rules "!!wrong!!": U_INVALID_FORMAT_ERROR
 -- nondeterministic collations
 CREATE COLLATION ctest_det (provider = icu, locale = '', deterministic = true);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION ctest_nondet (provider = icu, locale = '', deterministic = false);
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE TABLE test6 (a int, b text);
 -- same string in different normal forms
 INSERT INTO test6 VALUES (1, U&'\00E4bc');
@@ -1319,9 +1319,9 @@ SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
 (2 rows)
 
 CREATE COLLATION case_sensitive (provider = icu, locale = '');
-NOTICE:  using standard form "und" for locale ""
+NOTICE:  using standard form "und" for ICU locale ""
 CREATE COLLATION case_insensitive (provider = icu, locale = '@colStrength=secondary', deterministic = false);
-NOTICE:  using standard form "und-u-ks-level2" for locale "@colStrength=secondary"
+NOTICE:  using standard form "und-u-ks-level2" for ICU locale "@colStrength=secondary"
 SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE case_sensitive;
  ?column? | ?column? 
 ----------+----------
-- 
2.34.1

From 0cd2154f364999091aba52136a139df75f58d1b7 Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <peter@eisentraut.org>
Date: Fri, 16 Jun 2023 16:46:36 +0200
Subject: [PATCH] fixup! CREATE DATABASE: make LOCALE apply to all collation
 providers.

---
 doc/src/sgml/ref/create_database.sgml | 12 ++++++------
 src/test/icu/t/010_database.pl        | 23 +++++++++++++----------
 2 files changed, 19 insertions(+), 16 deletions(-)

diff --git a/doc/src/sgml/ref/create_database.sgml b/doc/src/sgml/ref/create_database.sgml
index dab05950ed..b2c8aef1ad 100644
--- a/doc/src/sgml/ref/create_database.sgml
+++ b/doc/src/sgml/ref/create_database.sgml
@@ -147,9 +147,9 @@ <title>Parameters</title>
        <para>
         Sets the default collation order and character classification in the
         new database.  Collation affects the sort order applied to strings,
-        e.g., in queries with ORDER BY, as well as the order used in indexes
+        e.g., in queries with <literal>ORDER BY</literal>, as well as the order used in indexes
         on text columns.  Character classification affects the categorization
-        of characters, e.g., lower, - upper and digit.  Also sets the
+        of characters, e.g., lower, upper, and digit.  Also sets the
         associated aspects of the operating system environment,
         <literal>LC_COLLATE</literal> and <literal>LC_CTYPE</literal>.  The
         default is the same setting as the template database.  See <xref
@@ -180,7 +180,7 @@ <title>Parameters</title>
        <para>
         Sets <literal>LC_COLLATE</literal> in the database server's operating
         system environment.  The default is the setting of <xref
-        linkend="create-database-locale"/> if specified; otherwise the same
+        linkend="create-database-locale"/> if specified, otherwise the same
         setting as the template database.  See below for additional
         restrictions.
        </para>
@@ -198,7 +198,7 @@ <title>Parameters</title>
        <para>
         Sets <literal>LC_CTYPE</literal> in the database server's operating
         system environment.  The default is the setting of <xref
-        linkend="create-database-locale"/> if specified; otherwise the same
+        linkend="create-database-locale"/> if specified, otherwise the same
         setting as the template database.  See below for additional
         restrictions.
        </para>
@@ -218,8 +218,8 @@ <title>Parameters</title>
         Specifies the ICU locale (see <xref
         linkend="collation-managing-create-icu"/>) for the database default
         collation order and character classification, overriding the setting
-        <xref linkend="create-database-locale"/>.  The <xref
-        linkend="create-database-locale-provider"/> must be ICU.  The default
+        <xref linkend="create-database-locale"/>.  The <link
+        linkend="create-database-locale-provider">locale provider</link> must be ICU.  The default
         is the setting of <xref linkend="create-database-locale"/> if
         specified; otherwise the same setting as the template database.
        </para>
diff --git a/src/test/icu/t/010_database.pl b/src/test/icu/t/010_database.pl
index b417b1a409..cbe5467f3c 100644
--- a/src/test/icu/t/010_database.pl
+++ b/src/test/icu/t/010_database.pl
@@ -52,27 +52,30 @@
 
 
 # Test that LOCALE='C' works for ICU
-my $ret1 = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu1 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
-);
-is($ret1, 0,
+is( $node1->psql(
+		'postgres',
+		q{CREATE DATABASE dbicu1 LOCALE_PROVIDER icu LOCALE 'C' TEMPLATE template0 ENCODING UTF8}
+	),
+	0,
 	"C locale works for ICU");
 
 # Test that LOCALE works for ICU locales if LC_COLLATE and LC_CTYPE
 # are specified
-my $ret2 = $node1->psql('postgres',
-	q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE '@colStrength=primary'
+is( $node1->psql(
+		'postgres',
+		q{CREATE DATABASE dbicu2 LOCALE_PROVIDER icu LOCALE '@colStrength=primary'
       LC_COLLATE='C' LC_CTYPE='C' TEMPLATE template0 ENCODING UTF8}
-);
-is($ret2, 0,
+	),
+	0,
 	"LOCALE works for ICU locales if LC_COLLATE and LC_CTYPE are specified");
 
 # Test that ICU-specific LOCALE without LC_COLLATE and LC_CTYPE must
 # be specified with ICU_LOCALE
-my ($ret3, $stdout, $stderr) = $node1->psql('postgres',
+my ($ret, $stdout, $stderr) = $node1->psql(
+	'postgres',
 	q{CREATE DATABASE dbicu3 LOCALE_PROVIDER icu LOCALE '@colStrength=primary'
       TEMPLATE template0 ENCODING UTF8});
-isnt($ret3, 0,
+isnt($ret, 0,
 	"ICU-specific locale must be specified with ICU_LOCALE: exit code not 0");
 like(
 	$stderr,
-- 
2.41.0