pg_dump/restore encoding woes
Started by Heikki Linnakangasover 12 years ago9 messages
pg_dump and pg_restore don't behave very nicely when the client and
server encodings don't match. Below are three issues that arise from
that. All the examples below use a console with a UTF-8 locale, and the
'latin1db' database uses ISO-8859-1 as the database encoding. In that
database, there is a single table called "pöö".
1. pg_dump verbose output
-------------------------
$ pg_dump -d latin1db -Fc -v -f a.backup
...
pg_dump: finding the columns and types of table "p��"
...
When client encoding is not specified explicitly with the -E option, or
PGCLIENTENCODING env variable, the dump is created in the server encoding.
Alexander Law reported this bug about a year ago, see bug #6742:
/messages/by-id/E1SrOVd-00028F-Sz@wrigleys.postgresql.org.
Now, you can say that it's the user's fault for not specifying
client_encoding correctly, but see #2.
2. pg_dump -t option doesn't work if client_encoding is not set
---------------------------------------------------------------
$ ./pg_dump -d latin1db -Fc -t pöö -f a.backup
pg_dump: No matching tables were found
$ ./pg_dump -d latin1db -Fc -t pöö -f a.backup -E utf-8
(success)
The table name given in the argument is passed to the server without
translation, so client_encoding needs to be set or the server will not
interpret the table name correctly.
Like #1, this is a user-error - he needs to set client_encoding
correctly. Other client programs like vacuumdb have the same problem.
But we could do better. psql sets client_encoding automatically
(client_encoding='auto') based on the locale. Why don't we do the same
in all the client programs?
However, pg_dump is special, because client encoding affects not only
the encoding used to speak to the server, but it also determines how the
resulting dump is encoded. If you have a UTF-8 server, and a LATIN1
console, there is no way to get a UTF-8 encoded dump of a single table
which has non-ASCII characters in its name. There is a good reason to
want to dump in the server encoding regardless of the encoding of the
client: that avoids the costly encoding conversion during the dump, and
very likely another conversion back on restore. (as a convenience, it
would be nice if you could specify "-E server" to mean "same as server
encoding")
The pg_dump -E option just sets client_encoding, but I think it would be
better for -E to only set the encoding used in the dump, and
PGCLIENTENCODING env variable (if set) was used to determine the
encoding of the command-line arguments. Opinions?
3. pg_restore -t option doesn't work if dump is in different encoding
---------------------------------------------------------------------
$ pg_dump -d latin1db -Fc -f a.backup
$ ./pg_restore -t pöö a.backup
(restores nothing)
pg_restore doesn't convert encodings when it matches the table name
given with -t option with the table names in the dump. Hence in above
example, where the dump is in LATIN1 encoding and the console uses a
UTF-8 locale, the table name is not matched even though there is a table
with that name in the dump.
Unfortunately I don't see any easy solution to this third issue :-(. We
don't have any infrastructure to do encoding conversions in the client.
I guess we could use iconv(3) if it's available, but I'm a bit reluctant
to start using that, given that we've managed to do with out client-side
conversions this far. Or we could do the conversion in the server using
"convert_from()", but that only works if pg_restore is connected to a
database. Perhaps it's best to just throw a warning in if -t is used and
the dump's encoding doesn't match the current locale.
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
Heikki Linnakangas <hlinnakangas@vmware.com> writes:
When client encoding is not specified explicitly with the -E option, or
PGCLIENTENCODING env variable, the dump is created in the server encoding.
Yeah, that's intentional as I recall.
However, pg_dump is special, because client encoding affects not only
the encoding used to speak to the server, but it also determines how the
resulting dump is encoded. If you have a UTF-8 server, and a LATIN1
console, there is no way to get a UTF-8 encoded dump of a single table
which has non-ASCII characters in its name. There is a good reason to
want to dump in the server encoding regardless of the encoding of the
client: that avoids the costly encoding conversion during the dump, and
very likely another conversion back on restore. (as a convenience, it
would be nice if you could specify "-E server" to mean "same as server
encoding")
There's a considerably more compelling reason than speed to default to
avoiding a conversion: doing a conversion carries significant risk of
outright failure, due to not being able to convert some data character
to the client character set.
The pg_dump -E option just sets client_encoding, but I think it would be
better for -E to only set the encoding used in the dump, and
PGCLIENTENCODING env variable (if set) was used to determine the
encoding of the command-line arguments. Opinions?
I think this is going to be a lot easier said than done, but feel free
to see if you can make it work. (As you point out, we don't have
any client-side encoding conversion infrastructure, but I don't see
how you're going to make this work without it.)
A second issue is whether we should divorce -E and PGCLIENTENCODING like
that, when they have always meant the same thing. You mentioned the
alternative of looking at pg_dump's locale environment to determine the
command line encoding --- would that be better?
regards, tom lane
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On 26.08.2013 18:59, Tom Lane wrote:
Heikki Linnakangas<hlinnakangas@vmware.com> writes:
The pg_dump -E option just sets client_encoding, but I think it would be
better for -E to only set the encoding used in the dump, and
PGCLIENTENCODING env variable (if set) was used to determine the
encoding of the command-line arguments. Opinions?I think this is going to be a lot easier said than done, but feel free
to see if you can make it work. (As you point out, we don't have
any client-side encoding conversion infrastructure, but I don't see
how you're going to make this work without it.)
First set client_encoding to PGCLIENTENCODING (ie. let libpq do its
default thing), and run the queries that fetch the OIDs of any matching
tables. Only after doing that, set client_encoding to that specified by
-E. That's actually pretty easy to do, as pg_dump already issues all the
queries that include user-given strings first.
There's one small wrinkle in that: if the dump fails because the server
sends an error, the error would come from the server in the -E encoding,
because that's used as the client_encoding after the initial queries.
Logically, the errors should be printed in PGCLIENTENCODING. But I think
we can live with that.
The pg_restore part is more difficult, as pg_restore needs to work
without a database connection at all. There the conversion has to be
done client-side.
A second issue is whether we should divorce -E and PGCLIENTENCODING like
that, when they have always meant the same thing. You mentioned the
alternative of looking at pg_dump's locale environment to determine the
command line encoding --- would that be better?
Hmm. I wasn't thinking of looking at the locale environment as an
alternative to the divorce, just as a way to determine the default for
the client encoding if PGCLIENTENCODING is not set.
It would be good for pg_dump to be consistent with other tools
(reindexdb etc.). If we say that LC_CTYPE determines the encoding of
command-line options, regardless of PGCLIENTENCODING, we should do the
same in other tools, too. And that would be weird for the other tools.
So no, I don't think we should do that.
My plan is to assume that the command-line options are in the "client
encoding", and the client encoding is determined by the following
things, in this order:
1. client_encoding setting given explicitly in the command line, as part
of the connection string.
2. PGCLIENTENCODING
3. LC_CTYPE
4. same as server_encoding
The encoding to be used in the pg_dump output is a different concept,
and there are reasons to *not* want the dump to be in the client
encoding. Hence using server_encoding for that would be a better default
than the current client encoding. This isn't so visible today, as
client_encoding defaults to server_encoding anyway, but it will be if we
set client_encoding='auto' (ie. look at LC_CTYPE) by default.
There are certainly cases where you'd want to use the client encoding as
the dump output encoding. For example if you pipe the pg_dump output to
some custom script that mangles it. Or if you just want to look at the
output, ie. if the output goes to a terminal. But for the usual case of
taking a backup, using the server encoding makes more sense. One option
is to check if the output is a tty (like psql does), and output the dump
in client or server encoding depending on that (if -E is not given).
So yeah, I think we should divorce -E and PGCLIENTENCODING. It feels
that using PGCLIENTENCODING to specify the output encoding was more
accidental than on purpose. Looking at the history, the implementation
has been that way forever, but the docs were adjusted by commit b524cb36
in 2005 to document that fact.
Here is a set of three patches that I've been working on:
0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patch
Separates pg_dump -E from PGCLIENTENCODING.
0002-Set-client_encoding-auto-in-all-client-utilities.patch
Set client_encoding='auto' in all the client utilities, like we already
did in psql. This fixes e.g "reindexdb -t" with a table name with
non-ASCII chars
0003-Convert-object-names-to-archive-encoding-before-matc.patch
Use iconv(3) in pg_restore to do encoding conversion in the client. This
involves a lot of autoconf changes that I'm not 100% sure about, other
than that it's pretty straightforward.
- Heikki
Attachments:
0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patchtext/x-diff; name=0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patchDownload
>From f722fee00a97ff1a7b4246176478fd97d360b716 Mon Sep 17 00:00:00 2001
From: Heikki Linnakangas <heikki.linnakangas@iki.fi>
Date: Mon, 26 Aug 2013 21:58:22 +0300
Subject: [PATCH 1/3] Divorce pg_dump -E option from PGCLIENTENCODING.
-E option now means the encoding to use in the dump output, and
PGCLIENTENCODING specifies what encoding object names given on the command
line options are in.
---
doc/src/sgml/ref/pg_dump.sgml | 20 +++++++--
src/bin/pg_dump/pg_dump.c | 99 ++++++++++++++++++++++++++++-------------
2 files changed, 84 insertions(+), 35 deletions(-)
diff --git a/doc/src/sgml/ref/pg_dump.sgml b/doc/src/sgml/ref/pg_dump.sgml
index 1ed5e4f..007bace 100644
--- a/doc/src/sgml/ref/pg_dump.sgml
+++ b/doc/src/sgml/ref/pg_dump.sgml
@@ -184,10 +184,24 @@ PostgreSQL documentation
<listitem>
<para>
Create the dump in the specified character set encoding. By default,
- the dump is created in the database encoding. (Another way to get the
- same result is to set the <envar>PGCLIENTENCODING</envar> environment
- variable to the desired dump encoding.)
+ the dump is created in the database encoding.
</para>
+
+ <note>
+ <para>
+ -E only affects the encoding used for the output. The client encoding
+ used to interpret any object names given on the command line is
+ determined by the <envar>PGCLIENTENCODING</envar> environment
+ variable.
+ </para>
+
+ <para>
+ Before <productname>PostgreSQL</productname> 9.3 setting
+ <envar>PGCLIENTENCODING</envar> had the same effect as the -E
+ option, and command-line options were always interpreted in the same
+ encoding as used in the output.
+ </para>
+ </note>
</listitem>
</varlistentry>
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index e1ef55f..2b41e00 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -143,8 +143,8 @@ static int serializable_deferrable = 0;
static void help(const char *progname);
-static void setup_connection(Archive *AH, const char *dumpencoding,
- char *use_role);
+static void setup_connection(Archive *AH, char *use_role);
+static void setup_dump_encoding(Archive *AH, const char *dumpencoding);
static ArchiveFormat parseArchiveFormat(const char *format, ArchiveMode *mode);
static void expand_schema_name_patterns(Archive *fout,
SimpleStringList *patterns,
@@ -635,7 +635,19 @@ main(int argc, char **argv)
* death.
*/
ConnectDatabase(fout, dbname, pghost, pgport, username, prompt_password);
- setup_connection(fout, dumpencoding, use_role);
+ setup_connection(fout, use_role);
+ /*
+ * Note that client_encoding is not set to the dump's output encoding,
+ * specified by -E, yet. We must expand the table/schema names given on
+ * the command line before we do that, because the command line arguments
+ * are assumed to *not* be in the encoding specified by the -E option, but
+ * whatever PGCLIENTENCODING is set to.
+ *
+ * Further note that the queries we run before setup_dump_encoding() is
+ * called must not return any strings that would be included in the dump,
+ * because they would be returned in different encoding than the dump.
+ * Oids are OK.
+ */
/*
* Disable security label support if server version < v9.1.x (prevents
@@ -720,6 +732,14 @@ main(int argc, char **argv)
/* non-matching exclusion patterns aren't an error */
/*
+ * We're now done with the queries using the table/schema names passed on
+ * the command-line, which are assumed to be in the client's true encoding
+ * (ie. PGCLIENTENCODING, not the encoding specified by -E). Switch
+ * client_encoding to the encoding to be used in the dump output.
+ */
+ setup_dump_encoding(fout, dumpencoding);
+
+ /*
* Dumping blobs is now default unless we saw an inclusion switch or -s
* ... but even if we did see one of these, -b turns it back on.
*/
@@ -915,37 +935,17 @@ help(const char *progname)
printf(_("Report bugs to <pgsql-bugs@postgresql.org>.\n"));
}
+/*
+ * Sets connection options that affect the dump output.
+ *
+ * Note: This does not set client_encoding. Call setup_dump_encoding() to set
+ * that, after running any queries that need to be run in in the clientÍ's real
+ * encoding.
+ */
static void
-setup_connection(Archive *AH, const char *dumpencoding, char *use_role)
+setup_connection(Archive *AH, char *use_role)
{
PGconn *conn = GetConnection(AH);
- const char *std_strings;
-
- /*
- * Set the client encoding if requested. If dumpencoding == NULL then
- * either it hasn't been requested or we're a cloned connection and then
- * this has already been set in CloneArchive according to the original
- * connection encoding.
- */
- if (dumpencoding)
- {
- if (PQsetClientEncoding(conn, dumpencoding) < 0)
- exit_horribly(NULL, "invalid client encoding \"%s\" specified\n",
- dumpencoding);
- }
-
- /*
- * Get the active encoding and the standard_conforming_strings setting, so
- * we know how to escape strings.
- */
- AH->encoding = PQclientEncoding(conn);
-
- std_strings = PQparameterStatus(conn, "standard_conforming_strings");
- AH->std_strings = (std_strings && strcmp(std_strings, "on") == 0);
-
- /* Set the role if requested */
- if (!use_role && AH->use_role)
- use_role = AH->use_role;
/* Set the role if requested */
if (use_role && AH->remoteVersion >= 80100)
@@ -1042,10 +1042,45 @@ setup_connection(Archive *AH, const char *dumpencoding, char *use_role)
}
}
+/*
+ * Sets client_encoding to the encoding the dump should be output in.
+ */
+static void
+setup_dump_encoding(Archive *AH, const char *dumpencoding)
+{
+ PGconn *conn = GetConnection(AH);
+ const char *std_strings;
+
+ /*
+ * If dumpenecoding is not specified, default to the server encoding.
+ */
+ if (!dumpencoding)
+ dumpencoding = PQparameterStatus(conn, "server_encoding");
+ if (dumpencoding)
+ {
+ if (PQsetClientEncoding(conn, dumpencoding) < 0)
+ {
+ exit_horribly(NULL, "invalid dump encoding \"%s\" specified: %s\n",
+ dumpencoding,
+ PQerrorMessage(conn));
+ }
+ }
+
+ /*
+ * Get the active encoding and the standard_conforming_strings setting, so
+ * we know how to escape strings.
+ */
+ AH->encoding = PQclientEncoding(conn);
+
+ std_strings = PQparameterStatus(conn, "standard_conforming_strings");
+ AH->std_strings = (std_strings && strcmp(std_strings, "on") == 0);
+}
+
static void
setupDumpWorker(Archive *AHX, RestoreOptions *ropt)
{
- setup_connection(AHX, NULL, NULL);
+ setup_connection(AHX, AHX->use_role);
+ setup_dump_encoding(AHX, pg_encoding_to_char(AHX->encoding));
}
static char *
--
1.7.10.4
0002-Set-client_encoding-auto-in-all-client-utilities.patchtext/x-diff; name=0002-Set-client_encoding-auto-in-all-client-utilities.patchDownload
>From e29c0c2aa7ec7eca370b877cdafc888d51b4939f Mon Sep 17 00:00:00 2001
From: Heikki Linnakangas <heikki.linnakangas@iki.fi>
Date: Mon, 26 Aug 2013 22:00:05 +0300
Subject: [PATCH 2/3] Set client_encoding='auto' in all client utilities.
We've been doing this for psql, when run in a terminal, since 9.1, and it
has worked pretty well, so expand to other utilities. Although the other
utilities don't have an interactive mode, client_encoding also affects the
e.g table names given on the command-line arguments are sent to the server.
---
doc/src/sgml/ref/pg_dump.sgml | 3 ++-
src/bin/pg_dump/pg_backup_db.c | 8 +++++---
src/bin/scripts/common.c | 8 +++++---
3 files changed, 12 insertions(+), 7 deletions(-)
diff --git a/doc/src/sgml/ref/pg_dump.sgml b/doc/src/sgml/ref/pg_dump.sgml
index 007bace..fc7c73f 100644
--- a/doc/src/sgml/ref/pg_dump.sgml
+++ b/doc/src/sgml/ref/pg_dump.sgml
@@ -192,7 +192,8 @@ PostgreSQL documentation
-E only affects the encoding used for the output. The client encoding
used to interpret any object names given on the command line is
determined by the <envar>PGCLIENTENCODING</envar> environment
- variable.
+ variable, or from the terminal's locale settings if not set
+ (<envar>LC_CTYPE</envar> environment variable on Unix systems).
</para>
<para>
diff --git a/src/bin/pg_dump/pg_backup_db.c b/src/bin/pg_dump/pg_backup_db.c
index cd2dded..75c82cd 100644
--- a/src/bin/pg_dump/pg_backup_db.c
+++ b/src/bin/pg_dump/pg_backup_db.c
@@ -240,7 +240,7 @@ ConnectDatabase(Archive *AHX,
*/
do
{
-#define PARAMS_ARRAY_SIZE 7
+#define PARAMS_ARRAY_SIZE 8
const char **keywords = pg_malloc(PARAMS_ARRAY_SIZE * sizeof(*keywords));
const char **values = pg_malloc(PARAMS_ARRAY_SIZE * sizeof(*values));
@@ -256,8 +256,10 @@ ConnectDatabase(Archive *AHX,
values[4] = dbname;
keywords[5] = "fallback_application_name";
values[5] = progname;
- keywords[6] = NULL;
- values[6] = NULL;
+ keywords[6] = "client_encoding";
+ values[6] = getenv("PGCLIENTENCODING") ? NULL : "auto";
+ keywords[7] = NULL;
+ values[7] = NULL;
new_pass = false;
AH->connection = PQconnectdbParams(keywords, values, true);
diff --git a/src/bin/scripts/common.c b/src/bin/scripts/common.c
index 4645bc1..bb95b44 100644
--- a/src/bin/scripts/common.c
+++ b/src/bin/scripts/common.c
@@ -107,7 +107,7 @@ connectDatabase(const char *dbname, const char *pghost, const char *pgport,
*/
do
{
-#define PARAMS_ARRAY_SIZE 7
+#define PARAMS_ARRAY_SIZE 8
const char **keywords = pg_malloc(PARAMS_ARRAY_SIZE * sizeof(*keywords));
const char **values = pg_malloc(PARAMS_ARRAY_SIZE * sizeof(*values));
@@ -123,8 +123,10 @@ connectDatabase(const char *dbname, const char *pghost, const char *pgport,
values[4] = dbname;
keywords[5] = "fallback_application_name";
values[5] = progname;
- keywords[6] = NULL;
- values[6] = NULL;
+ keywords[6] = "client_encoding";
+ values[6] = getenv("PGCLIENTENCODING") ? NULL : "auto";
+ keywords[7] = NULL;
+ values[7] = NULL;
new_pass = false;
conn = PQconnectdbParams(keywords, values, true);
--
1.7.10.4
0003-Convert-object-names-to-archive-encoding-before-matc.patchtext/x-diff; name=0003-Convert-object-names-to-archive-encoding-before-matc.patchDownload
>From 23ad1ef58c1f453f330cbbf15b64e529593c0344 Mon Sep 17 00:00:00 2001
From: Heikki Linnakangas <heikki.linnakangas@iki.fi>
Date: Tue, 27 Aug 2013 15:40:37 +0300
Subject: [PATCH 3/3] Convert object names to archive encoding before matching
in pg_restore.
If a table, function, or other object name is given on the pg_restore
command line, to only restore the named object, convert the name given on
the command line to the encoding the archive is in, before trying to match
the names. The command line options are assumed to be in the encoding
specified by PGCLIENTENCODING, if set, or the current locale (LC_CTYPE).
This is the same logic we use in client_encoding='auto' case, which is also
what we use in all the other client programs that come with PostgreSQL.
In pg_restore, we cannot just set client_encoding='auto' and let the server
deal with it, because pg_restore needs to work even without a database
connection, and with an archive in any encoding. Use iconv(3) to do the
conversion in the client-side.
This is the first time we use iconv(3), so add an autoconf check for it. If
iconv is not available, fall back to the old behavior and try to match the
strings even though we know they are in different encodings (that works OK
if the object name doesn't contain any non-ASCII characters).
The iconv.m4 file, and the lib-*.m4 files that it depends on, are copied
from the gettext distribution. They include the same "are we using GNU ld?"
check that we had in config/libtool.m4, so remove that.
---
aclocal.m4 | 5 +-
config/iconv.m4 | 268 +++++++++
config/lib-ld.m4 | 119 ++++
config/lib-link.m4 | 777 ++++++++++++++++++++++++
config/lib-prefix.m4 | 224 +++++++
config/libtool.m4 | 119 ----
configure | 1243 ++++++++++++++++++++++++++++++++++++---
configure.in | 4 +-
src/backend/utils/mb/encnames.c | 52 --
src/backend/utils/mb/mbutils.c | 23 +-
src/bin/pg_dump/pg_restore.c | 171 ++++++
src/include/pg_config.h.in | 6 +
src/include/port.h | 1 +
src/port/chklocale.c | 70 +++
14 files changed, 2799 insertions(+), 283 deletions(-)
create mode 100644 config/iconv.m4
create mode 100644 config/lib-ld.m4
create mode 100644 config/lib-link.m4
create mode 100644 config/lib-prefix.m4
delete mode 100644 config/libtool.m4
diff --git a/aclocal.m4 b/aclocal.m4
index eaf9800..a41a17f 100644
--- a/aclocal.m4
+++ b/aclocal.m4
@@ -5,8 +5,11 @@ m4_include([config/c-compiler.m4])
m4_include([config/c-library.m4])
m4_include([config/docbook.m4])
m4_include([config/general.m4])
-m4_include([config/libtool.m4])
m4_include([config/perl.m4])
m4_include([config/programs.m4])
m4_include([config/python.m4])
m4_include([config/tcl.m4])
+m4_include([config/lib-ld.m4])
+m4_include([config/lib-link.m4])
+m4_include([config/lib-prefix.m4])
+m4_include([config/iconv.m4])
diff --git a/config/iconv.m4 b/config/iconv.m4
new file mode 100644
index 0000000..a503646
--- /dev/null
+++ b/config/iconv.m4
@@ -0,0 +1,268 @@
+# iconv.m4 serial 18 (gettext-0.18.2)
+dnl Copyright (C) 2000-2002, 2007-2013 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+dnl From Bruno Haible.
+
+AC_DEFUN([AM_ICONV_LINKFLAGS_BODY],
+[
+ dnl Prerequisites of AC_LIB_LINKFLAGS_BODY.
+ AC_REQUIRE([AC_LIB_PREPARE_PREFIX])
+ AC_REQUIRE([AC_LIB_RPATH])
+
+ dnl Search for libiconv and define LIBICONV, LTLIBICONV and INCICONV
+ dnl accordingly.
+ AC_LIB_LINKFLAGS_BODY([iconv])
+])
+
+AC_DEFUN([AM_ICONV_LINK],
+[
+ dnl Some systems have iconv in libc, some have it in libiconv (OSF/1 and
+ dnl those with the standalone portable GNU libiconv installed).
+ AC_REQUIRE([AC_CANONICAL_HOST]) dnl for cross-compiles
+
+ dnl Search for libiconv and define LIBICONV, LTLIBICONV and INCICONV
+ dnl accordingly.
+ AC_REQUIRE([AM_ICONV_LINKFLAGS_BODY])
+
+ dnl Add $INCICONV to CPPFLAGS before performing the following checks,
+ dnl because if the user has installed libiconv and not disabled its use
+ dnl via --without-libiconv-prefix, he wants to use it. The first
+ dnl AC_LINK_IFELSE will then fail, the second AC_LINK_IFELSE will succeed.
+ am_save_CPPFLAGS="$CPPFLAGS"
+ AC_LIB_APPENDTOVAR([CPPFLAGS], [$INCICONV])
+
+ AC_CACHE_CHECK([for iconv], [am_cv_func_iconv], [
+ am_cv_func_iconv="no, consider installing GNU libiconv"
+ am_cv_lib_iconv=no
+ AC_LINK_IFELSE(
+ [AC_LANG_PROGRAM(
+ [[
+#include <stdlib.h>
+#include <iconv.h>
+ ]],
+ [[iconv_t cd = iconv_open("","");
+ iconv(cd,NULL,NULL,NULL,NULL);
+ iconv_close(cd);]])],
+ [am_cv_func_iconv=yes])
+ if test "$am_cv_func_iconv" != yes; then
+ am_save_LIBS="$LIBS"
+ LIBS="$LIBS $LIBICONV"
+ AC_LINK_IFELSE(
+ [AC_LANG_PROGRAM(
+ [[
+#include <stdlib.h>
+#include <iconv.h>
+ ]],
+ [[iconv_t cd = iconv_open("","");
+ iconv(cd,NULL,NULL,NULL,NULL);
+ iconv_close(cd);]])],
+ [am_cv_lib_iconv=yes]
+ [am_cv_func_iconv=yes])
+ LIBS="$am_save_LIBS"
+ fi
+ ])
+ if test "$am_cv_func_iconv" = yes; then
+ AC_CACHE_CHECK([for working iconv], [am_cv_func_iconv_works], [
+ dnl This tests against bugs in AIX 5.1, AIX 6.1..7.1, HP-UX 11.11,
+ dnl Solaris 10.
+ am_save_LIBS="$LIBS"
+ if test $am_cv_lib_iconv = yes; then
+ LIBS="$LIBS $LIBICONV"
+ fi
+ AC_RUN_IFELSE(
+ [AC_LANG_SOURCE([[
+#include <iconv.h>
+#include <string.h>
+int main ()
+{
+ int result = 0;
+ /* Test against AIX 5.1 bug: Failures are not distinguishable from successful
+ returns. */
+ {
+ iconv_t cd_utf8_to_88591 = iconv_open ("ISO8859-1", "UTF-8");
+ if (cd_utf8_to_88591 != (iconv_t)(-1))
+ {
+ static const char input[] = "\342\202\254"; /* EURO SIGN */
+ char buf[10];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_utf8_to_88591,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res == 0)
+ result |= 1;
+ iconv_close (cd_utf8_to_88591);
+ }
+ }
+ /* Test against Solaris 10 bug: Failures are not distinguishable from
+ successful returns. */
+ {
+ iconv_t cd_ascii_to_88591 = iconv_open ("ISO8859-1", "646");
+ if (cd_ascii_to_88591 != (iconv_t)(-1))
+ {
+ static const char input[] = "\263";
+ char buf[10];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_ascii_to_88591,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res == 0)
+ result |= 2;
+ iconv_close (cd_ascii_to_88591);
+ }
+ }
+ /* Test against AIX 6.1..7.1 bug: Buffer overrun. */
+ {
+ iconv_t cd_88591_to_utf8 = iconv_open ("UTF-8", "ISO-8859-1");
+ if (cd_88591_to_utf8 != (iconv_t)(-1))
+ {
+ static const char input[] = "\304";
+ static char buf[2] = { (char)0xDE, (char)0xAD };
+ const char *inptr = input;
+ size_t inbytesleft = 1;
+ char *outptr = buf;
+ size_t outbytesleft = 1;
+ size_t res = iconv (cd_88591_to_utf8,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res != (size_t)(-1) || outptr - buf > 1 || buf[1] != (char)0xAD)
+ result |= 4;
+ iconv_close (cd_88591_to_utf8);
+ }
+ }
+#if 0 /* This bug could be worked around by the caller. */
+ /* Test against HP-UX 11.11 bug: Positive return value instead of 0. */
+ {
+ iconv_t cd_88591_to_utf8 = iconv_open ("utf8", "iso88591");
+ if (cd_88591_to_utf8 != (iconv_t)(-1))
+ {
+ static const char input[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337";
+ char buf[50];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_88591_to_utf8,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if ((int)res > 0)
+ result |= 8;
+ iconv_close (cd_88591_to_utf8);
+ }
+ }
+#endif
+ /* Test against HP-UX 11.11 bug: No converter from EUC-JP to UTF-8 is
+ provided. */
+ if (/* Try standardized names. */
+ iconv_open ("UTF-8", "EUC-JP") == (iconv_t)(-1)
+ /* Try IRIX, OSF/1 names. */
+ && iconv_open ("UTF-8", "eucJP") == (iconv_t)(-1)
+ /* Try AIX names. */
+ && iconv_open ("UTF-8", "IBM-eucJP") == (iconv_t)(-1)
+ /* Try HP-UX names. */
+ && iconv_open ("utf8", "eucJP") == (iconv_t)(-1))
+ result |= 16;
+ return result;
+}]])],
+ [am_cv_func_iconv_works=yes],
+ [am_cv_func_iconv_works=no],
+ [
+changequote(,)dnl
+ case "$host_os" in
+ aix* | hpux*) am_cv_func_iconv_works="guessing no" ;;
+ *) am_cv_func_iconv_works="guessing yes" ;;
+ esac
+changequote([,])dnl
+ ])
+ LIBS="$am_save_LIBS"
+ ])
+ case "$am_cv_func_iconv_works" in
+ *no) am_func_iconv=no am_cv_lib_iconv=no ;;
+ *) am_func_iconv=yes ;;
+ esac
+ else
+ am_func_iconv=no am_cv_lib_iconv=no
+ fi
+ if test "$am_func_iconv" = yes; then
+ AC_DEFINE([HAVE_ICONV], [1],
+ [Define if you have the iconv() function and it works.])
+ fi
+ if test "$am_cv_lib_iconv" = yes; then
+ AC_MSG_CHECKING([how to link with libiconv])
+ AC_MSG_RESULT([$LIBICONV])
+ else
+ dnl If $LIBICONV didn't lead to a usable library, we don't need $INCICONV
+ dnl either.
+ CPPFLAGS="$am_save_CPPFLAGS"
+ LIBICONV=
+ LTLIBICONV=
+ fi
+ AC_SUBST([LIBICONV])
+ AC_SUBST([LTLIBICONV])
+])
+
+dnl Define AM_ICONV using AC_DEFUN_ONCE for Autoconf >= 2.64, in order to
+dnl avoid warnings like
+dnl "warning: AC_REQUIRE: `AM_ICONV' was expanded before it was required".
+dnl This is tricky because of the way 'aclocal' is implemented:
+dnl - It requires defining an auxiliary macro whose name ends in AC_DEFUN.
+dnl Otherwise aclocal's initial scan pass would miss the macro definition.
+dnl - It requires a line break inside the AC_DEFUN_ONCE and AC_DEFUN expansions.
+dnl Otherwise aclocal would emit many "Use of uninitialized value $1"
+dnl warnings.
+m4_define([gl_iconv_AC_DEFUN],
+ m4_version_prereq([2.64],
+ [[AC_DEFUN_ONCE(
+ [$1], [$2])]],
+ [m4_ifdef([gl_00GNULIB],
+ [[AC_DEFUN_ONCE(
+ [$1], [$2])]],
+ [[AC_DEFUN(
+ [$1], [$2])]])]))
+gl_iconv_AC_DEFUN([AM_ICONV],
+[
+ AM_ICONV_LINK
+ if test "$am_cv_func_iconv" = yes; then
+ AC_MSG_CHECKING([for iconv declaration])
+ AC_CACHE_VAL([am_cv_proto_iconv], [
+ AC_COMPILE_IFELSE(
+ [AC_LANG_PROGRAM(
+ [[
+#include <stdlib.h>
+#include <iconv.h>
+extern
+#ifdef __cplusplus
+"C"
+#endif
+#if defined(__STDC__) || defined(_MSC_VER) || defined(__cplusplus)
+size_t iconv (iconv_t cd, char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);
+#else
+size_t iconv();
+#endif
+ ]],
+ [[]])],
+ [am_cv_proto_iconv_arg1=""],
+ [am_cv_proto_iconv_arg1="const"])
+ am_cv_proto_iconv="extern size_t iconv (iconv_t cd, $am_cv_proto_iconv_arg1 char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);"])
+ am_cv_proto_iconv=`echo "[$]am_cv_proto_iconv" | tr -s ' ' | sed -e 's/( /(/'`
+ AC_MSG_RESULT([
+ $am_cv_proto_iconv])
+ AC_DEFINE_UNQUOTED([ICONV_CONST], [$am_cv_proto_iconv_arg1],
+ [Define as const if the declaration of iconv() needs const.])
+ dnl Also substitute ICONV_CONST in the gnulib generated <iconv.h>.
+ m4_ifdef([gl_ICONV_H_DEFAULTS],
+ [AC_REQUIRE([gl_ICONV_H_DEFAULTS])
+ if test -n "$am_cv_proto_iconv_arg1"; then
+ ICONV_CONST="const"
+ fi
+ ])
+ fi
+])
diff --git a/config/lib-ld.m4 b/config/lib-ld.m4
new file mode 100644
index 0000000..c145e47
--- /dev/null
+++ b/config/lib-ld.m4
@@ -0,0 +1,119 @@
+# lib-ld.m4 serial 6
+dnl Copyright (C) 1996-2003, 2009-2013 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+dnl Subroutines of libtool.m4,
+dnl with replacements s/_*LT_PATH/AC_LIB_PROG/ and s/lt_/acl_/ to avoid
+dnl collision with libtool.m4.
+
+dnl From libtool-2.4. Sets the variable with_gnu_ld to yes or no.
+AC_DEFUN([AC_LIB_PROG_LD_GNU],
+[AC_CACHE_CHECK([if the linker ($LD) is GNU ld], [acl_cv_prog_gnu_ld],
+[# I'd rather use --version here, but apparently some GNU lds only accept -v.
+case `$LD -v 2>&1 </dev/null` in
+*GNU* | *'with BFD'*)
+ acl_cv_prog_gnu_ld=yes
+ ;;
+*)
+ acl_cv_prog_gnu_ld=no
+ ;;
+esac])
+with_gnu_ld=$acl_cv_prog_gnu_ld
+])
+
+dnl From libtool-2.4. Sets the variable LD.
+AC_DEFUN([AC_LIB_PROG_LD],
+[AC_REQUIRE([AC_PROG_CC])dnl
+AC_REQUIRE([AC_CANONICAL_HOST])dnl
+
+AC_ARG_WITH([gnu-ld],
+ [AS_HELP_STRING([--with-gnu-ld],
+ [assume the C compiler uses GNU ld [default=no]])],
+ [test "$withval" = no || with_gnu_ld=yes],
+ [with_gnu_ld=no])dnl
+
+# Prepare PATH_SEPARATOR.
+# The user is always right.
+if test "${PATH_SEPARATOR+set}" != set; then
+ # Determine PATH_SEPARATOR by trying to find /bin/sh in a PATH which
+ # contains only /bin. Note that ksh looks also at the FPATH variable,
+ # so we have to set that as well for the test.
+ PATH_SEPARATOR=:
+ (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 \
+ && { (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 \
+ || PATH_SEPARATOR=';'
+ }
+fi
+
+ac_prog=ld
+if test "$GCC" = yes; then
+ # Check if gcc -print-prog-name=ld gives a path.
+ AC_MSG_CHECKING([for ld used by $CC])
+ case $host in
+ *-*-mingw*)
+ # gcc leaves a trailing carriage return which upsets mingw
+ ac_prog=`($CC -print-prog-name=ld) 2>&5 | tr -d '\015'` ;;
+ *)
+ ac_prog=`($CC -print-prog-name=ld) 2>&5` ;;
+ esac
+ case $ac_prog in
+ # Accept absolute paths.
+ [[\\/]]* | ?:[[\\/]]*)
+ re_direlt='/[[^/]][[^/]]*/\.\./'
+ # Canonicalize the pathname of ld
+ ac_prog=`echo "$ac_prog"| sed 's%\\\\%/%g'`
+ while echo "$ac_prog" | grep "$re_direlt" > /dev/null 2>&1; do
+ ac_prog=`echo $ac_prog| sed "s%$re_direlt%/%"`
+ done
+ test -z "$LD" && LD="$ac_prog"
+ ;;
+ "")
+ # If it fails, then pretend we aren't using GCC.
+ ac_prog=ld
+ ;;
+ *)
+ # If it is relative, then search for the first ld in PATH.
+ with_gnu_ld=unknown
+ ;;
+ esac
+elif test "$with_gnu_ld" = yes; then
+ AC_MSG_CHECKING([for GNU ld])
+else
+ AC_MSG_CHECKING([for non-GNU ld])
+fi
+AC_CACHE_VAL([acl_cv_path_LD],
+[if test -z "$LD"; then
+ acl_save_ifs="$IFS"; IFS=$PATH_SEPARATOR
+ for ac_dir in $PATH; do
+ IFS="$acl_save_ifs"
+ test -z "$ac_dir" && ac_dir=.
+ if test -f "$ac_dir/$ac_prog" || test -f "$ac_dir/$ac_prog$ac_exeext"; then
+ acl_cv_path_LD="$ac_dir/$ac_prog"
+ # Check to see if the program is GNU ld. I'd rather use --version,
+ # but apparently some variants of GNU ld only accept -v.
+ # Break only if it was the GNU/non-GNU ld that we prefer.
+ case `"$acl_cv_path_LD" -v 2>&1 </dev/null` in
+ *GNU* | *'with BFD'*)
+ test "$with_gnu_ld" != no && break
+ ;;
+ *)
+ test "$with_gnu_ld" != yes && break
+ ;;
+ esac
+ fi
+ done
+ IFS="$acl_save_ifs"
+else
+ acl_cv_path_LD="$LD" # Let the user override the test with a path.
+fi])
+LD="$acl_cv_path_LD"
+if test -n "$LD"; then
+ AC_MSG_RESULT([$LD])
+else
+ AC_MSG_RESULT([no])
+fi
+test -z "$LD" && AC_MSG_ERROR([no acceptable ld found in \$PATH])
+AC_LIB_PROG_LD_GNU
+])
diff --git a/config/lib-link.m4 b/config/lib-link.m4
new file mode 100644
index 0000000..073f040
--- /dev/null
+++ b/config/lib-link.m4
@@ -0,0 +1,777 @@
+# lib-link.m4 serial 26 (gettext-0.18.2)
+dnl Copyright (C) 2001-2013 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+dnl From Bruno Haible.
+
+AC_PREREQ([2.54])
+
+dnl AC_LIB_LINKFLAGS(name [, dependencies]) searches for libname and
+dnl the libraries corresponding to explicit and implicit dependencies.
+dnl Sets and AC_SUBSTs the LIB${NAME} and LTLIB${NAME} variables and
+dnl augments the CPPFLAGS variable.
+dnl Sets and AC_SUBSTs the LIB${NAME}_PREFIX variable to nonempty if libname
+dnl was found in ${LIB${NAME}_PREFIX}/$acl_libdirstem.
+AC_DEFUN([AC_LIB_LINKFLAGS],
+[
+ AC_REQUIRE([AC_LIB_PREPARE_PREFIX])
+ AC_REQUIRE([AC_LIB_RPATH])
+ pushdef([Name],[m4_translit([$1],[./+-], [____])])
+ pushdef([NAME],[m4_translit([$1],[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+ AC_CACHE_CHECK([how to link with lib[]$1], [ac_cv_lib[]Name[]_libs], [
+ AC_LIB_LINKFLAGS_BODY([$1], [$2])
+ ac_cv_lib[]Name[]_libs="$LIB[]NAME"
+ ac_cv_lib[]Name[]_ltlibs="$LTLIB[]NAME"
+ ac_cv_lib[]Name[]_cppflags="$INC[]NAME"
+ ac_cv_lib[]Name[]_prefix="$LIB[]NAME[]_PREFIX"
+ ])
+ LIB[]NAME="$ac_cv_lib[]Name[]_libs"
+ LTLIB[]NAME="$ac_cv_lib[]Name[]_ltlibs"
+ INC[]NAME="$ac_cv_lib[]Name[]_cppflags"
+ LIB[]NAME[]_PREFIX="$ac_cv_lib[]Name[]_prefix"
+ AC_LIB_APPENDTOVAR([CPPFLAGS], [$INC]NAME)
+ AC_SUBST([LIB]NAME)
+ AC_SUBST([LTLIB]NAME)
+ AC_SUBST([LIB]NAME[_PREFIX])
+ dnl Also set HAVE_LIB[]NAME so that AC_LIB_HAVE_LINKFLAGS can reuse the
+ dnl results of this search when this library appears as a dependency.
+ HAVE_LIB[]NAME=yes
+ popdef([NAME])
+ popdef([Name])
+])
+
+dnl AC_LIB_HAVE_LINKFLAGS(name, dependencies, includes, testcode, [missing-message])
+dnl searches for libname and the libraries corresponding to explicit and
+dnl implicit dependencies, together with the specified include files and
+dnl the ability to compile and link the specified testcode. The missing-message
+dnl defaults to 'no' and may contain additional hints for the user.
+dnl If found, it sets and AC_SUBSTs HAVE_LIB${NAME}=yes and the LIB${NAME}
+dnl and LTLIB${NAME} variables and augments the CPPFLAGS variable, and
+dnl #defines HAVE_LIB${NAME} to 1. Otherwise, it sets and AC_SUBSTs
+dnl HAVE_LIB${NAME}=no and LIB${NAME} and LTLIB${NAME} to empty.
+dnl Sets and AC_SUBSTs the LIB${NAME}_PREFIX variable to nonempty if libname
+dnl was found in ${LIB${NAME}_PREFIX}/$acl_libdirstem.
+AC_DEFUN([AC_LIB_HAVE_LINKFLAGS],
+[
+ AC_REQUIRE([AC_LIB_PREPARE_PREFIX])
+ AC_REQUIRE([AC_LIB_RPATH])
+ pushdef([Name],[m4_translit([$1],[./+-], [____])])
+ pushdef([NAME],[m4_translit([$1],[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+
+ dnl Search for lib[]Name and define LIB[]NAME, LTLIB[]NAME and INC[]NAME
+ dnl accordingly.
+ AC_LIB_LINKFLAGS_BODY([$1], [$2])
+
+ dnl Add $INC[]NAME to CPPFLAGS before performing the following checks,
+ dnl because if the user has installed lib[]Name and not disabled its use
+ dnl via --without-lib[]Name-prefix, he wants to use it.
+ ac_save_CPPFLAGS="$CPPFLAGS"
+ AC_LIB_APPENDTOVAR([CPPFLAGS], [$INC]NAME)
+
+ AC_CACHE_CHECK([for lib[]$1], [ac_cv_lib[]Name], [
+ ac_save_LIBS="$LIBS"
+ dnl If $LIB[]NAME contains some -l options, add it to the end of LIBS,
+ dnl because these -l options might require -L options that are present in
+ dnl LIBS. -l options benefit only from the -L options listed before it.
+ dnl Otherwise, add it to the front of LIBS, because it may be a static
+ dnl library that depends on another static library that is present in LIBS.
+ dnl Static libraries benefit only from the static libraries listed after
+ dnl it.
+ case " $LIB[]NAME" in
+ *" -l"*) LIBS="$LIBS $LIB[]NAME" ;;
+ *) LIBS="$LIB[]NAME $LIBS" ;;
+ esac
+ AC_LINK_IFELSE(
+ [AC_LANG_PROGRAM([[$3]], [[$4]])],
+ [ac_cv_lib[]Name=yes],
+ [ac_cv_lib[]Name='m4_if([$5], [], [no], [[$5]])'])
+ LIBS="$ac_save_LIBS"
+ ])
+ if test "$ac_cv_lib[]Name" = yes; then
+ HAVE_LIB[]NAME=yes
+ AC_DEFINE([HAVE_LIB]NAME, 1, [Define if you have the lib][$1 library.])
+ AC_MSG_CHECKING([how to link with lib[]$1])
+ AC_MSG_RESULT([$LIB[]NAME])
+ else
+ HAVE_LIB[]NAME=no
+ dnl If $LIB[]NAME didn't lead to a usable library, we don't need
+ dnl $INC[]NAME either.
+ CPPFLAGS="$ac_save_CPPFLAGS"
+ LIB[]NAME=
+ LTLIB[]NAME=
+ LIB[]NAME[]_PREFIX=
+ fi
+ AC_SUBST([HAVE_LIB]NAME)
+ AC_SUBST([LIB]NAME)
+ AC_SUBST([LTLIB]NAME)
+ AC_SUBST([LIB]NAME[_PREFIX])
+ popdef([NAME])
+ popdef([Name])
+])
+
+dnl Determine the platform dependent parameters needed to use rpath:
+dnl acl_libext,
+dnl acl_shlibext,
+dnl acl_libname_spec,
+dnl acl_library_names_spec,
+dnl acl_hardcode_libdir_flag_spec,
+dnl acl_hardcode_libdir_separator,
+dnl acl_hardcode_direct,
+dnl acl_hardcode_minus_L.
+AC_DEFUN([AC_LIB_RPATH],
+[
+ dnl Tell automake >= 1.10 to complain if config.rpath is missing.
+ m4_ifdef([AC_REQUIRE_AUX_FILE], [AC_REQUIRE_AUX_FILE([config.rpath])])
+ AC_REQUIRE([AC_PROG_CC]) dnl we use $CC, $GCC, $LDFLAGS
+ AC_REQUIRE([AC_LIB_PROG_LD]) dnl we use $LD, $with_gnu_ld
+ AC_REQUIRE([AC_CANONICAL_HOST]) dnl we use $host
+ AC_REQUIRE([AC_CONFIG_AUX_DIR_DEFAULT]) dnl we use $ac_aux_dir
+ AC_CACHE_CHECK([for shared library run path origin], [acl_cv_rpath], [
+ CC="$CC" GCC="$GCC" LDFLAGS="$LDFLAGS" LD="$LD" with_gnu_ld="$with_gnu_ld" \
+ ${CONFIG_SHELL-/bin/sh} "$ac_aux_dir/config.rpath" "$host" > conftest.sh
+ . ./conftest.sh
+ rm -f ./conftest.sh
+ acl_cv_rpath=done
+ ])
+ wl="$acl_cv_wl"
+ acl_libext="$acl_cv_libext"
+ acl_shlibext="$acl_cv_shlibext"
+ acl_libname_spec="$acl_cv_libname_spec"
+ acl_library_names_spec="$acl_cv_library_names_spec"
+ acl_hardcode_libdir_flag_spec="$acl_cv_hardcode_libdir_flag_spec"
+ acl_hardcode_libdir_separator="$acl_cv_hardcode_libdir_separator"
+ acl_hardcode_direct="$acl_cv_hardcode_direct"
+ acl_hardcode_minus_L="$acl_cv_hardcode_minus_L"
+ dnl Determine whether the user wants rpath handling at all.
+ AC_ARG_ENABLE([rpath],
+ [ --disable-rpath do not hardcode runtime library paths],
+ :, enable_rpath=yes)
+])
+
+dnl AC_LIB_FROMPACKAGE(name, package)
+dnl declares that libname comes from the given package. The configure file
+dnl will then not have a --with-libname-prefix option but a
+dnl --with-package-prefix option. Several libraries can come from the same
+dnl package. This declaration must occur before an AC_LIB_LINKFLAGS or similar
+dnl macro call that searches for libname.
+AC_DEFUN([AC_LIB_FROMPACKAGE],
+[
+ pushdef([NAME],[m4_translit([$1],[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+ define([acl_frompackage_]NAME, [$2])
+ popdef([NAME])
+ pushdef([PACK],[$2])
+ pushdef([PACKUP],[m4_translit(PACK,[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+ define([acl_libsinpackage_]PACKUP,
+ m4_ifdef([acl_libsinpackage_]PACKUP, [m4_defn([acl_libsinpackage_]PACKUP)[, ]],)[lib$1])
+ popdef([PACKUP])
+ popdef([PACK])
+])
+
+dnl AC_LIB_LINKFLAGS_BODY(name [, dependencies]) searches for libname and
+dnl the libraries corresponding to explicit and implicit dependencies.
+dnl Sets the LIB${NAME}, LTLIB${NAME} and INC${NAME} variables.
+dnl Also, sets the LIB${NAME}_PREFIX variable to nonempty if libname was found
+dnl in ${LIB${NAME}_PREFIX}/$acl_libdirstem.
+AC_DEFUN([AC_LIB_LINKFLAGS_BODY],
+[
+ AC_REQUIRE([AC_LIB_PREPARE_MULTILIB])
+ pushdef([NAME],[m4_translit([$1],[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+ pushdef([PACK],[m4_ifdef([acl_frompackage_]NAME, [acl_frompackage_]NAME, lib[$1])])
+ pushdef([PACKUP],[m4_translit(PACK,[abcdefghijklmnopqrstuvwxyz./+-],
+ [ABCDEFGHIJKLMNOPQRSTUVWXYZ____])])
+ pushdef([PACKLIBS],[m4_ifdef([acl_frompackage_]NAME, [acl_libsinpackage_]PACKUP, lib[$1])])
+ dnl Autoconf >= 2.61 supports dots in --with options.
+ pushdef([P_A_C_K],[m4_if(m4_version_compare(m4_defn([m4_PACKAGE_VERSION]),[2.61]),[-1],[m4_translit(PACK,[.],[_])],PACK)])
+ dnl By default, look in $includedir and $libdir.
+ use_additional=yes
+ AC_LIB_WITH_FINAL_PREFIX([
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+ ])
+ AC_ARG_WITH(P_A_C_K[-prefix],
+[[ --with-]]P_A_C_K[[-prefix[=DIR] search for ]PACKLIBS[ in DIR/include and DIR/lib
+ --without-]]P_A_C_K[[-prefix don't search for ]PACKLIBS[ in includedir and libdir]],
+[
+ if test "X$withval" = "Xno"; then
+ use_additional=no
+ else
+ if test "X$withval" = "X"; then
+ AC_LIB_WITH_FINAL_PREFIX([
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+ ])
+ else
+ additional_includedir="$withval/include"
+ additional_libdir="$withval/$acl_libdirstem"
+ if test "$acl_libdirstem2" != "$acl_libdirstem" \
+ && ! test -d "$withval/$acl_libdirstem"; then
+ additional_libdir="$withval/$acl_libdirstem2"
+ fi
+ fi
+ fi
+])
+ dnl Search the library and its dependencies in $additional_libdir and
+ dnl $LDFLAGS. Using breadth-first-seach.
+ LIB[]NAME=
+ LTLIB[]NAME=
+ INC[]NAME=
+ LIB[]NAME[]_PREFIX=
+ dnl HAVE_LIB${NAME} is an indicator that LIB${NAME}, LTLIB${NAME} have been
+ dnl computed. So it has to be reset here.
+ HAVE_LIB[]NAME=
+ rpathdirs=
+ ltrpathdirs=
+ names_already_handled=
+ names_next_round='$1 $2'
+ while test -n "$names_next_round"; do
+ names_this_round="$names_next_round"
+ names_next_round=
+ for name in $names_this_round; do
+ already_handled=
+ for n in $names_already_handled; do
+ if test "$n" = "$name"; then
+ already_handled=yes
+ break
+ fi
+ done
+ if test -z "$already_handled"; then
+ names_already_handled="$names_already_handled $name"
+ dnl See if it was already located by an earlier AC_LIB_LINKFLAGS
+ dnl or AC_LIB_HAVE_LINKFLAGS call.
+ uppername=`echo "$name" | sed -e 'y|abcdefghijklmnopqrstuvwxyz./+-|ABCDEFGHIJKLMNOPQRSTUVWXYZ____|'`
+ eval value=\"\$HAVE_LIB$uppername\"
+ if test -n "$value"; then
+ if test "$value" = yes; then
+ eval value=\"\$LIB$uppername\"
+ test -z "$value" || LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$value"
+ eval value=\"\$LTLIB$uppername\"
+ test -z "$value" || LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }$value"
+ else
+ dnl An earlier call to AC_LIB_HAVE_LINKFLAGS has determined
+ dnl that this library doesn't exist. So just drop it.
+ :
+ fi
+ else
+ dnl Search the library lib$name in $additional_libdir and $LDFLAGS
+ dnl and the already constructed $LIBNAME/$LTLIBNAME.
+ found_dir=
+ found_la=
+ found_so=
+ found_a=
+ eval libname=\"$acl_libname_spec\" # typically: libname=lib$name
+ if test -n "$acl_shlibext"; then
+ shrext=".$acl_shlibext" # typically: shrext=.so
+ else
+ shrext=
+ fi
+ if test $use_additional = yes; then
+ dir="$additional_libdir"
+ dnl The same code as in the loop below:
+ dnl First look for a shared library.
+ if test -n "$acl_shlibext"; then
+ if test -f "$dir/$libname$shrext"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext"
+ else
+ if test "$acl_library_names_spec" = '$libname$shrext$versuffix'; then
+ ver=`(cd "$dir" && \
+ for f in "$libname$shrext".*; do echo "$f"; done \
+ | sed -e "s,^$libname$shrext\\\\.,," \
+ | sort -t '.' -n -r -k1,1 -k2,2 -k3,3 -k4,4 -k5,5 \
+ | sed 1q ) 2>/dev/null`
+ if test -n "$ver" && test -f "$dir/$libname$shrext.$ver"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext.$ver"
+ fi
+ else
+ eval library_names=\"$acl_library_names_spec\"
+ for f in $library_names; do
+ if test -f "$dir/$f"; then
+ found_dir="$dir"
+ found_so="$dir/$f"
+ break
+ fi
+ done
+ fi
+ fi
+ fi
+ dnl Then look for a static library.
+ if test "X$found_dir" = "X"; then
+ if test -f "$dir/$libname.$acl_libext"; then
+ found_dir="$dir"
+ found_a="$dir/$libname.$acl_libext"
+ fi
+ fi
+ if test "X$found_dir" != "X"; then
+ if test -f "$dir/$libname.la"; then
+ found_la="$dir/$libname.la"
+ fi
+ fi
+ fi
+ if test "X$found_dir" = "X"; then
+ for x in $LDFLAGS $LTLIB[]NAME; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ case "$x" in
+ -L*)
+ dir=`echo "X$x" | sed -e 's/^X-L//'`
+ dnl First look for a shared library.
+ if test -n "$acl_shlibext"; then
+ if test -f "$dir/$libname$shrext"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext"
+ else
+ if test "$acl_library_names_spec" = '$libname$shrext$versuffix'; then
+ ver=`(cd "$dir" && \
+ for f in "$libname$shrext".*; do echo "$f"; done \
+ | sed -e "s,^$libname$shrext\\\\.,," \
+ | sort -t '.' -n -r -k1,1 -k2,2 -k3,3 -k4,4 -k5,5 \
+ | sed 1q ) 2>/dev/null`
+ if test -n "$ver" && test -f "$dir/$libname$shrext.$ver"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext.$ver"
+ fi
+ else
+ eval library_names=\"$acl_library_names_spec\"
+ for f in $library_names; do
+ if test -f "$dir/$f"; then
+ found_dir="$dir"
+ found_so="$dir/$f"
+ break
+ fi
+ done
+ fi
+ fi
+ fi
+ dnl Then look for a static library.
+ if test "X$found_dir" = "X"; then
+ if test -f "$dir/$libname.$acl_libext"; then
+ found_dir="$dir"
+ found_a="$dir/$libname.$acl_libext"
+ fi
+ fi
+ if test "X$found_dir" != "X"; then
+ if test -f "$dir/$libname.la"; then
+ found_la="$dir/$libname.la"
+ fi
+ fi
+ ;;
+ esac
+ if test "X$found_dir" != "X"; then
+ break
+ fi
+ done
+ fi
+ if test "X$found_dir" != "X"; then
+ dnl Found the library.
+ LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }-L$found_dir -l$name"
+ if test "X$found_so" != "X"; then
+ dnl Linking with a shared library. We attempt to hardcode its
+ dnl directory into the executable's runpath, unless it's the
+ dnl standard /usr/lib.
+ if test "$enable_rpath" = no \
+ || test "X$found_dir" = "X/usr/$acl_libdirstem" \
+ || test "X$found_dir" = "X/usr/$acl_libdirstem2"; then
+ dnl No hardcoding is needed.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$found_so"
+ else
+ dnl Use an explicit option to hardcode DIR into the resulting
+ dnl binary.
+ dnl Potentially add DIR to ltrpathdirs.
+ dnl The ltrpathdirs will be appended to $LTLIBNAME at the end.
+ haveit=
+ for x in $ltrpathdirs; do
+ if test "X$x" = "X$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ ltrpathdirs="$ltrpathdirs $found_dir"
+ fi
+ dnl The hardcoding into $LIBNAME is system dependent.
+ if test "$acl_hardcode_direct" = yes; then
+ dnl Using DIR/libNAME.so during linking hardcodes DIR into the
+ dnl resulting binary.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$found_so"
+ else
+ if test -n "$acl_hardcode_libdir_flag_spec" && test "$acl_hardcode_minus_L" = no; then
+ dnl Use an explicit option to hardcode DIR into the resulting
+ dnl binary.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$found_so"
+ dnl Potentially add DIR to rpathdirs.
+ dnl The rpathdirs will be appended to $LIBNAME at the end.
+ haveit=
+ for x in $rpathdirs; do
+ if test "X$x" = "X$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ rpathdirs="$rpathdirs $found_dir"
+ fi
+ else
+ dnl Rely on "-L$found_dir".
+ dnl But don't add it if it's already contained in the LDFLAGS
+ dnl or the already constructed $LIBNAME
+ haveit=
+ for x in $LDFLAGS $LIB[]NAME; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-L$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }-L$found_dir"
+ fi
+ if test "$acl_hardcode_minus_L" != no; then
+ dnl FIXME: Not sure whether we should use
+ dnl "-L$found_dir -l$name" or "-L$found_dir $found_so"
+ dnl here.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$found_so"
+ else
+ dnl We cannot use $acl_hardcode_runpath_var and LD_RUN_PATH
+ dnl here, because this doesn't fit in flags passed to the
+ dnl compiler. So give up. No hardcoding. This affects only
+ dnl very old systems.
+ dnl FIXME: Not sure whether we should use
+ dnl "-L$found_dir -l$name" or "-L$found_dir $found_so"
+ dnl here.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }-l$name"
+ fi
+ fi
+ fi
+ fi
+ else
+ if test "X$found_a" != "X"; then
+ dnl Linking with a static library.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$found_a"
+ else
+ dnl We shouldn't come here, but anyway it's good to have a
+ dnl fallback.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }-L$found_dir -l$name"
+ fi
+ fi
+ dnl Assume the include files are nearby.
+ additional_includedir=
+ case "$found_dir" in
+ */$acl_libdirstem | */$acl_libdirstem/)
+ basedir=`echo "X$found_dir" | sed -e 's,^X,,' -e "s,/$acl_libdirstem/"'*$,,'`
+ if test "$name" = '$1'; then
+ LIB[]NAME[]_PREFIX="$basedir"
+ fi
+ additional_includedir="$basedir/include"
+ ;;
+ */$acl_libdirstem2 | */$acl_libdirstem2/)
+ basedir=`echo "X$found_dir" | sed -e 's,^X,,' -e "s,/$acl_libdirstem2/"'*$,,'`
+ if test "$name" = '$1'; then
+ LIB[]NAME[]_PREFIX="$basedir"
+ fi
+ additional_includedir="$basedir/include"
+ ;;
+ esac
+ if test "X$additional_includedir" != "X"; then
+ dnl Potentially add $additional_includedir to $INCNAME.
+ dnl But don't add it
+ dnl 1. if it's the standard /usr/include,
+ dnl 2. if it's /usr/local/include and we are using GCC on Linux,
+ dnl 3. if it's already present in $CPPFLAGS or the already
+ dnl constructed $INCNAME,
+ dnl 4. if it doesn't exist as a directory.
+ if test "X$additional_includedir" != "X/usr/include"; then
+ haveit=
+ if test "X$additional_includedir" = "X/usr/local/include"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux* | gnu* | k*bsd*-gnu) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ for x in $CPPFLAGS $INC[]NAME; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-I$additional_includedir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_includedir"; then
+ dnl Really add $additional_includedir to $INCNAME.
+ INC[]NAME="${INC[]NAME}${INC[]NAME:+ }-I$additional_includedir"
+ fi
+ fi
+ fi
+ fi
+ fi
+ dnl Look for dependencies.
+ if test -n "$found_la"; then
+ dnl Read the .la file. It defines the variables
+ dnl dlname, library_names, old_library, dependency_libs, current,
+ dnl age, revision, installed, dlopen, dlpreopen, libdir.
+ save_libdir="$libdir"
+ case "$found_la" in
+ */* | *\\*) . "$found_la" ;;
+ *) . "./$found_la" ;;
+ esac
+ libdir="$save_libdir"
+ dnl We use only dependency_libs.
+ for dep in $dependency_libs; do
+ case "$dep" in
+ -L*)
+ additional_libdir=`echo "X$dep" | sed -e 's/^X-L//'`
+ dnl Potentially add $additional_libdir to $LIBNAME and $LTLIBNAME.
+ dnl But don't add it
+ dnl 1. if it's the standard /usr/lib,
+ dnl 2. if it's /usr/local/lib and we are using GCC on Linux,
+ dnl 3. if it's already present in $LDFLAGS or the already
+ dnl constructed $LIBNAME,
+ dnl 4. if it doesn't exist as a directory.
+ if test "X$additional_libdir" != "X/usr/$acl_libdirstem" \
+ && test "X$additional_libdir" != "X/usr/$acl_libdirstem2"; then
+ haveit=
+ if test "X$additional_libdir" = "X/usr/local/$acl_libdirstem" \
+ || test "X$additional_libdir" = "X/usr/local/$acl_libdirstem2"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux* | gnu* | k*bsd*-gnu) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ haveit=
+ for x in $LDFLAGS $LIB[]NAME; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-L$additional_libdir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_libdir"; then
+ dnl Really add $additional_libdir to $LIBNAME.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }-L$additional_libdir"
+ fi
+ fi
+ haveit=
+ for x in $LDFLAGS $LTLIB[]NAME; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-L$additional_libdir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_libdir"; then
+ dnl Really add $additional_libdir to $LTLIBNAME.
+ LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }-L$additional_libdir"
+ fi
+ fi
+ fi
+ fi
+ ;;
+ -R*)
+ dir=`echo "X$dep" | sed -e 's/^X-R//'`
+ if test "$enable_rpath" != no; then
+ dnl Potentially add DIR to rpathdirs.
+ dnl The rpathdirs will be appended to $LIBNAME at the end.
+ haveit=
+ for x in $rpathdirs; do
+ if test "X$x" = "X$dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ rpathdirs="$rpathdirs $dir"
+ fi
+ dnl Potentially add DIR to ltrpathdirs.
+ dnl The ltrpathdirs will be appended to $LTLIBNAME at the end.
+ haveit=
+ for x in $ltrpathdirs; do
+ if test "X$x" = "X$dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ ltrpathdirs="$ltrpathdirs $dir"
+ fi
+ fi
+ ;;
+ -l*)
+ dnl Handle this in the next round.
+ names_next_round="$names_next_round "`echo "X$dep" | sed -e 's/^X-l//'`
+ ;;
+ *.la)
+ dnl Handle this in the next round. Throw away the .la's
+ dnl directory; it is already contained in a preceding -L
+ dnl option.
+ names_next_round="$names_next_round "`echo "X$dep" | sed -e 's,^X.*/,,' -e 's,^lib,,' -e 's,\.la$,,'`
+ ;;
+ *)
+ dnl Most likely an immediate library name.
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$dep"
+ LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }$dep"
+ ;;
+ esac
+ done
+ fi
+ else
+ dnl Didn't find the library; assume it is in the system directories
+ dnl known to the linker and runtime loader. (All the system
+ dnl directories known to the linker should also be known to the
+ dnl runtime loader, otherwise the system is severely misconfigured.)
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }-l$name"
+ LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }-l$name"
+ fi
+ fi
+ fi
+ done
+ done
+ if test "X$rpathdirs" != "X"; then
+ if test -n "$acl_hardcode_libdir_separator"; then
+ dnl Weird platform: only the last -rpath option counts, the user must
+ dnl pass all path elements in one option. We can arrange that for a
+ dnl single library, but not when more than one $LIBNAMEs are used.
+ alldirs=
+ for found_dir in $rpathdirs; do
+ alldirs="${alldirs}${alldirs:+$acl_hardcode_libdir_separator}$found_dir"
+ done
+ dnl Note: acl_hardcode_libdir_flag_spec uses $libdir and $wl.
+ acl_save_libdir="$libdir"
+ libdir="$alldirs"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$flag"
+ else
+ dnl The -rpath options are cumulative.
+ for found_dir in $rpathdirs; do
+ acl_save_libdir="$libdir"
+ libdir="$found_dir"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ LIB[]NAME="${LIB[]NAME}${LIB[]NAME:+ }$flag"
+ done
+ fi
+ fi
+ if test "X$ltrpathdirs" != "X"; then
+ dnl When using libtool, the option that works for both libraries and
+ dnl executables is -R. The -R options are cumulative.
+ for found_dir in $ltrpathdirs; do
+ LTLIB[]NAME="${LTLIB[]NAME}${LTLIB[]NAME:+ }-R$found_dir"
+ done
+ fi
+ popdef([P_A_C_K])
+ popdef([PACKLIBS])
+ popdef([PACKUP])
+ popdef([PACK])
+ popdef([NAME])
+])
+
+dnl AC_LIB_APPENDTOVAR(VAR, CONTENTS) appends the elements of CONTENTS to VAR,
+dnl unless already present in VAR.
+dnl Works only for CPPFLAGS, not for LIB* variables because that sometimes
+dnl contains two or three consecutive elements that belong together.
+AC_DEFUN([AC_LIB_APPENDTOVAR],
+[
+ for element in [$2]; do
+ haveit=
+ for x in $[$1]; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X$element"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ [$1]="${[$1]}${[$1]:+ }$element"
+ fi
+ done
+])
+
+dnl For those cases where a variable contains several -L and -l options
+dnl referring to unknown libraries and directories, this macro determines the
+dnl necessary additional linker options for the runtime path.
+dnl AC_LIB_LINKFLAGS_FROM_LIBS([LDADDVAR], [LIBSVALUE], [USE-LIBTOOL])
+dnl sets LDADDVAR to linker options needed together with LIBSVALUE.
+dnl If USE-LIBTOOL evaluates to non-empty, linking with libtool is assumed,
+dnl otherwise linking without libtool is assumed.
+AC_DEFUN([AC_LIB_LINKFLAGS_FROM_LIBS],
+[
+ AC_REQUIRE([AC_LIB_RPATH])
+ AC_REQUIRE([AC_LIB_PREPARE_MULTILIB])
+ $1=
+ if test "$enable_rpath" != no; then
+ if test -n "$acl_hardcode_libdir_flag_spec" && test "$acl_hardcode_minus_L" = no; then
+ dnl Use an explicit option to hardcode directories into the resulting
+ dnl binary.
+ rpathdirs=
+ next=
+ for opt in $2; do
+ if test -n "$next"; then
+ dir="$next"
+ dnl No need to hardcode the standard /usr/lib.
+ if test "X$dir" != "X/usr/$acl_libdirstem" \
+ && test "X$dir" != "X/usr/$acl_libdirstem2"; then
+ rpathdirs="$rpathdirs $dir"
+ fi
+ next=
+ else
+ case $opt in
+ -L) next=yes ;;
+ -L*) dir=`echo "X$opt" | sed -e 's,^X-L,,'`
+ dnl No need to hardcode the standard /usr/lib.
+ if test "X$dir" != "X/usr/$acl_libdirstem" \
+ && test "X$dir" != "X/usr/$acl_libdirstem2"; then
+ rpathdirs="$rpathdirs $dir"
+ fi
+ next= ;;
+ *) next= ;;
+ esac
+ fi
+ done
+ if test "X$rpathdirs" != "X"; then
+ if test -n ""$3""; then
+ dnl libtool is used for linking. Use -R options.
+ for dir in $rpathdirs; do
+ $1="${$1}${$1:+ }-R$dir"
+ done
+ else
+ dnl The linker is used for linking directly.
+ if test -n "$acl_hardcode_libdir_separator"; then
+ dnl Weird platform: only the last -rpath option counts, the user
+ dnl must pass all path elements in one option.
+ alldirs=
+ for dir in $rpathdirs; do
+ alldirs="${alldirs}${alldirs:+$acl_hardcode_libdir_separator}$dir"
+ done
+ acl_save_libdir="$libdir"
+ libdir="$alldirs"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ $1="$flag"
+ else
+ dnl The -rpath options are cumulative.
+ for dir in $rpathdirs; do
+ acl_save_libdir="$libdir"
+ libdir="$dir"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ $1="${$1}${$1:+ }$flag"
+ done
+ fi
+ fi
+ fi
+ fi
+ fi
+ AC_SUBST([$1])
+])
diff --git a/config/lib-prefix.m4 b/config/lib-prefix.m4
new file mode 100644
index 0000000..60908e8
--- /dev/null
+++ b/config/lib-prefix.m4
@@ -0,0 +1,224 @@
+# lib-prefix.m4 serial 7 (gettext-0.18)
+dnl Copyright (C) 2001-2005, 2008-2013 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+dnl From Bruno Haible.
+
+dnl AC_LIB_ARG_WITH is synonymous to AC_ARG_WITH in autoconf-2.13, and
+dnl similar to AC_ARG_WITH in autoconf 2.52...2.57 except that is doesn't
+dnl require excessive bracketing.
+ifdef([AC_HELP_STRING],
+[AC_DEFUN([AC_LIB_ARG_WITH], [AC_ARG_WITH([$1],[[$2]],[$3],[$4])])],
+[AC_DEFUN([AC_][LIB_ARG_WITH], [AC_ARG_WITH([$1],[$2],[$3],[$4])])])
+
+dnl AC_LIB_PREFIX adds to the CPPFLAGS and LDFLAGS the flags that are needed
+dnl to access previously installed libraries. The basic assumption is that
+dnl a user will want packages to use other packages he previously installed
+dnl with the same --prefix option.
+dnl This macro is not needed if only AC_LIB_LINKFLAGS is used to locate
+dnl libraries, but is otherwise very convenient.
+AC_DEFUN([AC_LIB_PREFIX],
+[
+ AC_BEFORE([$0], [AC_LIB_LINKFLAGS])
+ AC_REQUIRE([AC_PROG_CC])
+ AC_REQUIRE([AC_CANONICAL_HOST])
+ AC_REQUIRE([AC_LIB_PREPARE_MULTILIB])
+ AC_REQUIRE([AC_LIB_PREPARE_PREFIX])
+ dnl By default, look in $includedir and $libdir.
+ use_additional=yes
+ AC_LIB_WITH_FINAL_PREFIX([
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+ ])
+ AC_LIB_ARG_WITH([lib-prefix],
+[ --with-lib-prefix[=DIR] search for libraries in DIR/include and DIR/lib
+ --without-lib-prefix don't search for libraries in includedir and libdir],
+[
+ if test "X$withval" = "Xno"; then
+ use_additional=no
+ else
+ if test "X$withval" = "X"; then
+ AC_LIB_WITH_FINAL_PREFIX([
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+ ])
+ else
+ additional_includedir="$withval/include"
+ additional_libdir="$withval/$acl_libdirstem"
+ fi
+ fi
+])
+ if test $use_additional = yes; then
+ dnl Potentially add $additional_includedir to $CPPFLAGS.
+ dnl But don't add it
+ dnl 1. if it's the standard /usr/include,
+ dnl 2. if it's already present in $CPPFLAGS,
+ dnl 3. if it's /usr/local/include and we are using GCC on Linux,
+ dnl 4. if it doesn't exist as a directory.
+ if test "X$additional_includedir" != "X/usr/include"; then
+ haveit=
+ for x in $CPPFLAGS; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-I$additional_includedir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test "X$additional_includedir" = "X/usr/local/include"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux* | gnu* | k*bsd*-gnu) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ if test -d "$additional_includedir"; then
+ dnl Really add $additional_includedir to $CPPFLAGS.
+ CPPFLAGS="${CPPFLAGS}${CPPFLAGS:+ }-I$additional_includedir"
+ fi
+ fi
+ fi
+ fi
+ dnl Potentially add $additional_libdir to $LDFLAGS.
+ dnl But don't add it
+ dnl 1. if it's the standard /usr/lib,
+ dnl 2. if it's already present in $LDFLAGS,
+ dnl 3. if it's /usr/local/lib and we are using GCC on Linux,
+ dnl 4. if it doesn't exist as a directory.
+ if test "X$additional_libdir" != "X/usr/$acl_libdirstem"; then
+ haveit=
+ for x in $LDFLAGS; do
+ AC_LIB_WITH_FINAL_PREFIX([eval x=\"$x\"])
+ if test "X$x" = "X-L$additional_libdir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test "X$additional_libdir" = "X/usr/local/$acl_libdirstem"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux*) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ if test -d "$additional_libdir"; then
+ dnl Really add $additional_libdir to $LDFLAGS.
+ LDFLAGS="${LDFLAGS}${LDFLAGS:+ }-L$additional_libdir"
+ fi
+ fi
+ fi
+ fi
+ fi
+])
+
+dnl AC_LIB_PREPARE_PREFIX creates variables acl_final_prefix,
+dnl acl_final_exec_prefix, containing the values to which $prefix and
+dnl $exec_prefix will expand at the end of the configure script.
+AC_DEFUN([AC_LIB_PREPARE_PREFIX],
+[
+ dnl Unfortunately, prefix and exec_prefix get only finally determined
+ dnl at the end of configure.
+ if test "X$prefix" = "XNONE"; then
+ acl_final_prefix="$ac_default_prefix"
+ else
+ acl_final_prefix="$prefix"
+ fi
+ if test "X$exec_prefix" = "XNONE"; then
+ acl_final_exec_prefix='${prefix}'
+ else
+ acl_final_exec_prefix="$exec_prefix"
+ fi
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ eval acl_final_exec_prefix=\"$acl_final_exec_prefix\"
+ prefix="$acl_save_prefix"
+])
+
+dnl AC_LIB_WITH_FINAL_PREFIX([statement]) evaluates statement, with the
+dnl variables prefix and exec_prefix bound to the values they will have
+dnl at the end of the configure script.
+AC_DEFUN([AC_LIB_WITH_FINAL_PREFIX],
+[
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ $1
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+])
+
+dnl AC_LIB_PREPARE_MULTILIB creates
+dnl - a variable acl_libdirstem, containing the basename of the libdir, either
+dnl "lib" or "lib64" or "lib/64",
+dnl - a variable acl_libdirstem2, as a secondary possible value for
+dnl acl_libdirstem, either the same as acl_libdirstem or "lib/sparcv9" or
+dnl "lib/amd64".
+AC_DEFUN([AC_LIB_PREPARE_MULTILIB],
+[
+ dnl There is no formal standard regarding lib and lib64.
+ dnl On glibc systems, the current practice is that on a system supporting
+ dnl 32-bit and 64-bit instruction sets or ABIs, 64-bit libraries go under
+ dnl $prefix/lib64 and 32-bit libraries go under $prefix/lib. We determine
+ dnl the compiler's default mode by looking at the compiler's library search
+ dnl path. If at least one of its elements ends in /lib64 or points to a
+ dnl directory whose absolute pathname ends in /lib64, we assume a 64-bit ABI.
+ dnl Otherwise we use the default, namely "lib".
+ dnl On Solaris systems, the current practice is that on a system supporting
+ dnl 32-bit and 64-bit instruction sets or ABIs, 64-bit libraries go under
+ dnl $prefix/lib/64 (which is a symlink to either $prefix/lib/sparcv9 or
+ dnl $prefix/lib/amd64) and 32-bit libraries go under $prefix/lib.
+ AC_REQUIRE([AC_CANONICAL_HOST])
+ acl_libdirstem=lib
+ acl_libdirstem2=
+ case "$host_os" in
+ solaris*)
+ dnl See Solaris 10 Software Developer Collection > Solaris 64-bit Developer's Guide > The Development Environment
+ dnl <http://docs.sun.com/app/docs/doc/816-5138/dev-env?l=en&a=view>.
+ dnl "Portable Makefiles should refer to any library directories using the 64 symbolic link."
+ dnl But we want to recognize the sparcv9 or amd64 subdirectory also if the
+ dnl symlink is missing, so we set acl_libdirstem2 too.
+ AC_CACHE_CHECK([for 64-bit host], [gl_cv_solaris_64bit],
+ [AC_EGREP_CPP([sixtyfour bits], [
+#ifdef _LP64
+sixtyfour bits
+#endif
+ ], [gl_cv_solaris_64bit=yes], [gl_cv_solaris_64bit=no])
+ ])
+ if test $gl_cv_solaris_64bit = yes; then
+ acl_libdirstem=lib/64
+ case "$host_cpu" in
+ sparc*) acl_libdirstem2=lib/sparcv9 ;;
+ i*86 | x86_64) acl_libdirstem2=lib/amd64 ;;
+ esac
+ fi
+ ;;
+ *)
+ searchpath=`(LC_ALL=C $CC -print-search-dirs) 2>/dev/null | sed -n -e 's,^libraries: ,,p' | sed -e 's,^=,,'`
+ if test -n "$searchpath"; then
+ acl_save_IFS="${IFS= }"; IFS=":"
+ for searchdir in $searchpath; do
+ if test -d "$searchdir"; then
+ case "$searchdir" in
+ */lib64/ | */lib64 ) acl_libdirstem=lib64 ;;
+ */../ | */.. )
+ # Better ignore directories of this form. They are misleading.
+ ;;
+ *) searchdir=`cd "$searchdir" && pwd`
+ case "$searchdir" in
+ */lib64 ) acl_libdirstem=lib64 ;;
+ esac ;;
+ esac
+ fi
+ done
+ IFS="$acl_save_IFS"
+ fi
+ ;;
+ esac
+ test -n "$acl_libdirstem2" || acl_libdirstem2="$acl_libdirstem"
+])
diff --git a/config/libtool.m4 b/config/libtool.m4
deleted file mode 100644
index f6e426d..0000000
--- a/config/libtool.m4
+++ /dev/null
@@ -1,119 +0,0 @@
-## libtool.m4 - Configure libtool for the host system. -*-Shell-script-*-
-## Copyright (C) 1996-1999,2000 Free Software Foundation, Inc.
-## Originally by Gordon Matzigkeit <gord@gnu.ai.mit.edu>, 1996
-##
-## This program is free software; you can redistribute it and/or modify
-## it under the terms of the GNU General Public License as published by
-## the Free Software Foundation; either version 2 of the License, or
-## (at your option) any later version.
-##
-## This program is distributed in the hope that it will be useful, but
-## WITHOUT ANY WARRANTY; without even the implied warranty of
-## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
-## General Public License for more details.
-##
-## You should have received a copy of the GNU General Public License
-## along with this program; if not, write to the Free Software
-## Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
-##
-## As a special exception to the GNU General Public License, if you
-## distribute this file as part of a program that contains a
-## configuration script generated by Autoconf, you may include it under
-## the same distribution terms that you use for the rest of that program.
-
-# No, PostgreSQL doesn't use libtool (yet), we just borrow stuff from it.
-# This file was taken on 2000-10-20 from the multi-language branch (since
-# that is the branch that PostgreSQL would most likely adopt anyway).
-# --petere
-
-# ... bunch of stuff removed here ...
-
-# PGAC_PROG_LD - find the path to the GNU or non-GNU linker
-AC_DEFUN([PGAC_PROG_LD],
-[AC_ARG_WITH(gnu-ld,
-[ --with-gnu-ld assume the C compiler uses GNU ld [[default=no]]],
-test "$withval" = no || with_gnu_ld=yes, with_gnu_ld=no)
-AC_REQUIRE([AC_PROG_CC])dnl
-AC_REQUIRE([AC_CANONICAL_HOST])dnl
-dnl ###not for PostgreSQL### AC_REQUIRE([AC_CANONICAL_BUILD])dnl
-ac_prog=ld
-if test "$GCC" = yes; then
- # Check if gcc -print-prog-name=ld gives a path.
- AC_MSG_CHECKING([for ld used by GCC])
- case $host in
- *-*-mingw*)
- # gcc leaves a trailing carriage return which upsets mingw
- ac_prog=`($CC -print-prog-name=ld) 2>&5 | tr -d '\015'` ;;
- *)
- ac_prog=`($CC -print-prog-name=ld) 2>&5` ;;
- esac
- case "$ac_prog" in
- # Accept absolute paths.
-changequote(,)dnl
- [\\/]* | [A-Za-z]:[\\/]*)
- re_direlt='/[^/][^/]*/\.\./'
-changequote([,])dnl
- # Canonicalize the path of ld
- ac_prog=`echo $ac_prog| sed 's%\\\\%/%g'`
- while echo $ac_prog | grep "$re_direlt" > /dev/null 2>&1; do
- ac_prog=`echo $ac_prog| sed "s%$re_direlt%/%"`
- done
- test -z "$LD" && LD="$ac_prog"
- ;;
- "")
- # If it fails, then pretend we aren't using GCC.
- ac_prog=ld
- ;;
- *)
- # If it is relative, then search for the first ld in PATH.
- with_gnu_ld=unknown
- ;;
- esac
-elif test "$with_gnu_ld" = yes; then
- AC_MSG_CHECKING([for GNU ld])
-else
- AC_MSG_CHECKING([for non-GNU ld])
-fi
-AC_CACHE_VAL(ac_cv_path_LD,
-[if test -z "$LD"; then
- IFS="${IFS= }"; ac_save_ifs="$IFS"; IFS="${IFS}${PATH_SEPARATOR-:}"
- for ac_dir in $PATH; do
- test -z "$ac_dir" && ac_dir=.
- if test -f "$ac_dir/$ac_prog" || test -f "$ac_dir/$ac_prog$ac_exeext"; then
- ac_cv_path_LD="$ac_dir/$ac_prog"
- # Check to see if the program is GNU ld. I'd rather use --version,
- # but apparently some GNU ld's only accept -v.
- # Break only if it was the GNU/non-GNU ld that we prefer.
- if "$ac_cv_path_LD" -v 2>&1 < /dev/null | egrep '(GNU|with BFD)' > /dev/null; then
- test "$with_gnu_ld" != no && break
- else
- test "$with_gnu_ld" != yes && break
- fi
- fi
- done
- IFS="$ac_save_ifs"
-else
- ac_cv_path_LD="$LD" # Let the user override the test with a path.
-fi])
-LD="$ac_cv_path_LD"
-if test -n "$LD"; then
- AC_MSG_RESULT($LD)
-else
- AC_MSG_RESULT(no)
-fi
-test -z "$LD" && AC_MSG_ERROR([no acceptable ld found in \$PATH])
-PGAC_PROG_LD_GNU
-])
-
-AC_DEFUN([PGAC_PROG_LD_GNU],
-[AC_CACHE_CHECK([if the linker ($LD) is GNU ld], ac_cv_prog_gnu_ld,
-[# I'd rather use --version here, but apparently some GNU ld's only accept -v.
-if $LD -v 2>&1 </dev/null | egrep '(GNU|with BFD)' 1>&5; then
- ac_cv_prog_gnu_ld=yes
-else
- ac_cv_prog_gnu_ld=no
-fi])
-with_gnu_ld=$ac_cv_prog_gnu_ld
-])
-
-# ... more stuff removed ...
diff --git a/configure b/configure
index d4a544d..51ef5da 100755
--- a/configure
+++ b/configure
@@ -657,6 +657,8 @@ TCL_LIB_FILE
TCL_INCLUDE_SPEC
TCL_CONFIG_SH
TCLSH
+LTLIBICONV
+LIBICONV
XGETTEXT
MSGMERGE
MSGFMT
@@ -843,10 +845,11 @@ with_libxml
with_libxslt
with_system_tzdata
with_zlib
-with_gnu_ld
enable_largefile
enable_float4_byval
enable_float8_byval
+with_gnu_ld
+with_libiconv_prefix
'
ac_precious_vars='build_alias
host_alias
@@ -1502,6 +1505,7 @@ Optional Features:
--disable-largefile omit support for large files
--disable-float4-byval disable float4 passed by value
--disable-float8-byval disable float8 passed by value
+ --disable-rpath do not hardcode runtime library paths
Optional Packages:
--with-PACKAGE[=ARG] use PACKAGE [ARG=yes]
@@ -1542,6 +1546,8 @@ Optional Packages:
use system time zone data in DIR
--without-zlib do not use Zlib
--with-gnu-ld assume the C compiler uses GNU ld [default=no]
+ --with-libiconv-prefix[=DIR] search for libiconv in DIR/include and DIR/lib
+ --without-libiconv-prefix don't search for libiconv in includedir and libdir
Some influential environment variables:
CC C compiler command
@@ -2010,7 +2016,6 @@ ac_compiler_gnu=$ac_cv_c_compiler_gnu
-
ac_aux_dir=
for ac_dir in config "$srcdir"/config; do
if test -f "$ac_dir/install-sh"; then
@@ -6218,105 +6223,24 @@ LDFLAGS="$LDFLAGS $LIBDIRS"
-
-# Check whether --with-gnu-ld was given.
-if test "${with_gnu_ld+set}" = set; then
- withval=$with_gnu_ld; test "$withval" = no || with_gnu_ld=yes
-else
- with_gnu_ld=no
-fi
-
-ac_prog=ld
-if test "$GCC" = yes; then
- # Check if gcc -print-prog-name=ld gives a path.
- { $as_echo "$as_me:$LINENO: checking for ld used by GCC" >&5
-$as_echo_n "checking for ld used by GCC... " >&6; }
- case $host in
- *-*-mingw*)
- # gcc leaves a trailing carriage return which upsets mingw
- ac_prog=`($CC -print-prog-name=ld) 2>&5 | tr -d '\015'` ;;
- *)
- ac_prog=`($CC -print-prog-name=ld) 2>&5` ;;
- esac
- case "$ac_prog" in
- # Accept absolute paths.
- [\\/]* | [A-Za-z]:[\\/]*)
- re_direlt='/[^/][^/]*/\.\./'
- # Canonicalize the path of ld
- ac_prog=`echo $ac_prog| sed 's%\\\\%/%g'`
- while echo $ac_prog | grep "$re_direlt" > /dev/null 2>&1; do
- ac_prog=`echo $ac_prog| sed "s%$re_direlt%/%"`
- done
- test -z "$LD" && LD="$ac_prog"
- ;;
- "")
- # If it fails, then pretend we aren't using GCC.
- ac_prog=ld
- ;;
- *)
- # If it is relative, then search for the first ld in PATH.
- with_gnu_ld=unknown
- ;;
- esac
-elif test "$with_gnu_ld" = yes; then
- { $as_echo "$as_me:$LINENO: checking for GNU ld" >&5
-$as_echo_n "checking for GNU ld... " >&6; }
-else
- { $as_echo "$as_me:$LINENO: checking for non-GNU ld" >&5
-$as_echo_n "checking for non-GNU ld... " >&6; }
-fi
-if test "${ac_cv_path_LD+set}" = set; then
- $as_echo_n "(cached) " >&6
-else
- if test -z "$LD"; then
- IFS="${IFS= }"; ac_save_ifs="$IFS"; IFS="${IFS}${PATH_SEPARATOR-:}"
- for ac_dir in $PATH; do
- test -z "$ac_dir" && ac_dir=.
- if test -f "$ac_dir/$ac_prog" || test -f "$ac_dir/$ac_prog$ac_exeext"; then
- ac_cv_path_LD="$ac_dir/$ac_prog"
- # Check to see if the program is GNU ld. I'd rather use --version,
- # but apparently some GNU ld's only accept -v.
- # Break only if it was the GNU/non-GNU ld that we prefer.
- if "$ac_cv_path_LD" -v 2>&1 < /dev/null | egrep '(GNU|with BFD)' > /dev/null; then
- test "$with_gnu_ld" != no && break
- else
- test "$with_gnu_ld" != yes && break
- fi
- fi
- done
- IFS="$ac_save_ifs"
-else
- ac_cv_path_LD="$LD" # Let the user override the test with a path.
-fi
-fi
-
-LD="$ac_cv_path_LD"
-if test -n "$LD"; then
- { $as_echo "$as_me:$LINENO: result: $LD" >&5
-$as_echo "$LD" >&6; }
-else
- { $as_echo "$as_me:$LINENO: result: no" >&5
-$as_echo "no" >&6; }
-fi
-test -z "$LD" && { { $as_echo "$as_me:$LINENO: error: no acceptable ld found in \$PATH" >&5
-$as_echo "$as_me: error: no acceptable ld found in \$PATH" >&2;}
- { (exit 1); exit 1; }; }
{ $as_echo "$as_me:$LINENO: checking if the linker ($LD) is GNU ld" >&5
$as_echo_n "checking if the linker ($LD) is GNU ld... " >&6; }
-if test "${ac_cv_prog_gnu_ld+set}" = set; then
+if test "${acl_cv_prog_gnu_ld+set}" = set; then
$as_echo_n "(cached) " >&6
else
- # I'd rather use --version here, but apparently some GNU ld's only accept -v.
-if $LD -v 2>&1 </dev/null | egrep '(GNU|with BFD)' 1>&5; then
- ac_cv_prog_gnu_ld=yes
-else
- ac_cv_prog_gnu_ld=no
-fi
+ # I'd rather use --version here, but apparently some GNU lds only accept -v.
+case `$LD -v 2>&1 </dev/null` in
+*GNU* | *'with BFD'*)
+ acl_cv_prog_gnu_ld=yes
+ ;;
+*)
+ acl_cv_prog_gnu_ld=no
+ ;;
+esac
fi
-{ $as_echo "$as_me:$LINENO: result: $ac_cv_prog_gnu_ld" >&5
-$as_echo "$ac_cv_prog_gnu_ld" >&6; }
-with_gnu_ld=$ac_cv_prog_gnu_ld
-
+{ $as_echo "$as_me:$LINENO: result: $acl_cv_prog_gnu_ld" >&5
+$as_echo "$acl_cv_prog_gnu_ld" >&6; }
+with_gnu_ld=$acl_cv_prog_gnu_ld
@@ -29259,6 +29183,1133 @@ done
fi
+
+ if test "X$prefix" = "XNONE"; then
+ acl_final_prefix="$ac_default_prefix"
+ else
+ acl_final_prefix="$prefix"
+ fi
+ if test "X$exec_prefix" = "XNONE"; then
+ acl_final_exec_prefix='${prefix}'
+ else
+ acl_final_exec_prefix="$exec_prefix"
+ fi
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ eval acl_final_exec_prefix=\"$acl_final_exec_prefix\"
+ prefix="$acl_save_prefix"
+
+
+
+# Check whether --with-gnu-ld was given.
+if test "${with_gnu_ld+set}" = set; then
+ withval=$with_gnu_ld; test "$withval" = no || with_gnu_ld=yes
+else
+ with_gnu_ld=no
+fi
+
+# Prepare PATH_SEPARATOR.
+# The user is always right.
+if test "${PATH_SEPARATOR+set}" != set; then
+ # Determine PATH_SEPARATOR by trying to find /bin/sh in a PATH which
+ # contains only /bin. Note that ksh looks also at the FPATH variable,
+ # so we have to set that as well for the test.
+ PATH_SEPARATOR=:
+ (PATH='/bin;/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 \
+ && { (PATH='/bin:/bin'; FPATH=$PATH; sh -c :) >/dev/null 2>&1 \
+ || PATH_SEPARATOR=';'
+ }
+fi
+
+ac_prog=ld
+if test "$GCC" = yes; then
+ # Check if gcc -print-prog-name=ld gives a path.
+ { $as_echo "$as_me:$LINENO: checking for ld used by $CC" >&5
+$as_echo_n "checking for ld used by $CC... " >&6; }
+ case $host in
+ *-*-mingw*)
+ # gcc leaves a trailing carriage return which upsets mingw
+ ac_prog=`($CC -print-prog-name=ld) 2>&5 | tr -d '\015'` ;;
+ *)
+ ac_prog=`($CC -print-prog-name=ld) 2>&5` ;;
+ esac
+ case $ac_prog in
+ # Accept absolute paths.
+ [\\/]* | ?:[\\/]*)
+ re_direlt='/[^/][^/]*/\.\./'
+ # Canonicalize the pathname of ld
+ ac_prog=`echo "$ac_prog"| sed 's%\\\\%/%g'`
+ while echo "$ac_prog" | grep "$re_direlt" > /dev/null 2>&1; do
+ ac_prog=`echo $ac_prog| sed "s%$re_direlt%/%"`
+ done
+ test -z "$LD" && LD="$ac_prog"
+ ;;
+ "")
+ # If it fails, then pretend we aren't using GCC.
+ ac_prog=ld
+ ;;
+ *)
+ # If it is relative, then search for the first ld in PATH.
+ with_gnu_ld=unknown
+ ;;
+ esac
+elif test "$with_gnu_ld" = yes; then
+ { $as_echo "$as_me:$LINENO: checking for GNU ld" >&5
+$as_echo_n "checking for GNU ld... " >&6; }
+else
+ { $as_echo "$as_me:$LINENO: checking for non-GNU ld" >&5
+$as_echo_n "checking for non-GNU ld... " >&6; }
+fi
+if test "${acl_cv_path_LD+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+ if test -z "$LD"; then
+ acl_save_ifs="$IFS"; IFS=$PATH_SEPARATOR
+ for ac_dir in $PATH; do
+ IFS="$acl_save_ifs"
+ test -z "$ac_dir" && ac_dir=.
+ if test -f "$ac_dir/$ac_prog" || test -f "$ac_dir/$ac_prog$ac_exeext"; then
+ acl_cv_path_LD="$ac_dir/$ac_prog"
+ # Check to see if the program is GNU ld. I'd rather use --version,
+ # but apparently some variants of GNU ld only accept -v.
+ # Break only if it was the GNU/non-GNU ld that we prefer.
+ case `"$acl_cv_path_LD" -v 2>&1 </dev/null` in
+ *GNU* | *'with BFD'*)
+ test "$with_gnu_ld" != no && break
+ ;;
+ *)
+ test "$with_gnu_ld" != yes && break
+ ;;
+ esac
+ fi
+ done
+ IFS="$acl_save_ifs"
+else
+ acl_cv_path_LD="$LD" # Let the user override the test with a path.
+fi
+fi
+
+LD="$acl_cv_path_LD"
+if test -n "$LD"; then
+ { $as_echo "$as_me:$LINENO: result: $LD" >&5
+$as_echo "$LD" >&6; }
+else
+ { $as_echo "$as_me:$LINENO: result: no" >&5
+$as_echo "no" >&6; }
+fi
+test -z "$LD" && { { $as_echo "$as_me:$LINENO: error: no acceptable ld found in \$PATH" >&5
+$as_echo "$as_me: error: no acceptable ld found in \$PATH" >&2;}
+ { (exit 1); exit 1; }; }
+{ $as_echo "$as_me:$LINENO: checking if the linker ($LD) is GNU ld" >&5
+$as_echo_n "checking if the linker ($LD) is GNU ld... " >&6; }
+if test "${acl_cv_prog_gnu_ld+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+ # I'd rather use --version here, but apparently some GNU lds only accept -v.
+case `$LD -v 2>&1 </dev/null` in
+*GNU* | *'with BFD'*)
+ acl_cv_prog_gnu_ld=yes
+ ;;
+*)
+ acl_cv_prog_gnu_ld=no
+ ;;
+esac
+fi
+{ $as_echo "$as_me:$LINENO: result: $acl_cv_prog_gnu_ld" >&5
+$as_echo "$acl_cv_prog_gnu_ld" >&6; }
+with_gnu_ld=$acl_cv_prog_gnu_ld
+
+
+
+
+ { $as_echo "$as_me:$LINENO: checking for shared library run path origin" >&5
+$as_echo_n "checking for shared library run path origin... " >&6; }
+if test "${acl_cv_rpath+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+
+ CC="$CC" GCC="$GCC" LDFLAGS="$LDFLAGS" LD="$LD" with_gnu_ld="$with_gnu_ld" \
+ ${CONFIG_SHELL-/bin/sh} "$ac_aux_dir/config.rpath" "$host" > conftest.sh
+ . ./conftest.sh
+ rm -f ./conftest.sh
+ acl_cv_rpath=done
+
+fi
+{ $as_echo "$as_me:$LINENO: result: $acl_cv_rpath" >&5
+$as_echo "$acl_cv_rpath" >&6; }
+ wl="$acl_cv_wl"
+ acl_libext="$acl_cv_libext"
+ acl_shlibext="$acl_cv_shlibext"
+ acl_libname_spec="$acl_cv_libname_spec"
+ acl_library_names_spec="$acl_cv_library_names_spec"
+ acl_hardcode_libdir_flag_spec="$acl_cv_hardcode_libdir_flag_spec"
+ acl_hardcode_libdir_separator="$acl_cv_hardcode_libdir_separator"
+ acl_hardcode_direct="$acl_cv_hardcode_direct"
+ acl_hardcode_minus_L="$acl_cv_hardcode_minus_L"
+ # Check whether --enable-rpath was given.
+if test "${enable_rpath+set}" = set; then
+ enableval=$enable_rpath; :
+else
+ enable_rpath=yes
+fi
+
+
+
+
+ acl_libdirstem=lib
+ acl_libdirstem2=
+ case "$host_os" in
+ solaris*)
+ { $as_echo "$as_me:$LINENO: checking for 64-bit host" >&5
+$as_echo_n "checking for 64-bit host... " >&6; }
+if test "${gl_cv_solaris_64bit+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+ cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h. */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h. */
+
+#ifdef _LP64
+sixtyfour bits
+#endif
+
+_ACEOF
+if (eval "$ac_cpp conftest.$ac_ext") 2>&5 |
+ $EGREP "sixtyfour bits" >/dev/null 2>&1; then
+ gl_cv_solaris_64bit=yes
+else
+ gl_cv_solaris_64bit=no
+fi
+rm -f conftest*
+
+
+fi
+{ $as_echo "$as_me:$LINENO: result: $gl_cv_solaris_64bit" >&5
+$as_echo "$gl_cv_solaris_64bit" >&6; }
+ if test $gl_cv_solaris_64bit = yes; then
+ acl_libdirstem=lib/64
+ case "$host_cpu" in
+ sparc*) acl_libdirstem2=lib/sparcv9 ;;
+ i*86 | x86_64) acl_libdirstem2=lib/amd64 ;;
+ esac
+ fi
+ ;;
+ *)
+ searchpath=`(LC_ALL=C $CC -print-search-dirs) 2>/dev/null | sed -n -e 's,^libraries: ,,p' | sed -e 's,^=,,'`
+ if test -n "$searchpath"; then
+ acl_save_IFS="${IFS= }"; IFS=":"
+ for searchdir in $searchpath; do
+ if test -d "$searchdir"; then
+ case "$searchdir" in
+ */lib64/ | */lib64 ) acl_libdirstem=lib64 ;;
+ */../ | */.. )
+ # Better ignore directories of this form. They are misleading.
+ ;;
+ *) searchdir=`cd "$searchdir" && pwd`
+ case "$searchdir" in
+ */lib64 ) acl_libdirstem=lib64 ;;
+ esac ;;
+ esac
+ fi
+ done
+ IFS="$acl_save_IFS"
+ fi
+ ;;
+ esac
+ test -n "$acl_libdirstem2" || acl_libdirstem2="$acl_libdirstem"
+
+
+
+
+
+
+
+
+
+
+
+
+ use_additional=yes
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+
+# Check whether --with-libiconv-prefix was given.
+if test "${with_libiconv_prefix+set}" = set; then
+ withval=$with_libiconv_prefix;
+ if test "X$withval" = "Xno"; then
+ use_additional=no
+ else
+ if test "X$withval" = "X"; then
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+
+ eval additional_includedir=\"$includedir\"
+ eval additional_libdir=\"$libdir\"
+
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ else
+ additional_includedir="$withval/include"
+ additional_libdir="$withval/$acl_libdirstem"
+ if test "$acl_libdirstem2" != "$acl_libdirstem" \
+ && ! test -d "$withval/$acl_libdirstem"; then
+ additional_libdir="$withval/$acl_libdirstem2"
+ fi
+ fi
+ fi
+
+fi
+
+ LIBICONV=
+ LTLIBICONV=
+ INCICONV=
+ LIBICONV_PREFIX=
+ HAVE_LIBICONV=
+ rpathdirs=
+ ltrpathdirs=
+ names_already_handled=
+ names_next_round='iconv '
+ while test -n "$names_next_round"; do
+ names_this_round="$names_next_round"
+ names_next_round=
+ for name in $names_this_round; do
+ already_handled=
+ for n in $names_already_handled; do
+ if test "$n" = "$name"; then
+ already_handled=yes
+ break
+ fi
+ done
+ if test -z "$already_handled"; then
+ names_already_handled="$names_already_handled $name"
+ uppername=`echo "$name" | sed -e 'y|abcdefghijklmnopqrstuvwxyz./+-|ABCDEFGHIJKLMNOPQRSTUVWXYZ____|'`
+ eval value=\"\$HAVE_LIB$uppername\"
+ if test -n "$value"; then
+ if test "$value" = yes; then
+ eval value=\"\$LIB$uppername\"
+ test -z "$value" || LIBICONV="${LIBICONV}${LIBICONV:+ }$value"
+ eval value=\"\$LTLIB$uppername\"
+ test -z "$value" || LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }$value"
+ else
+ :
+ fi
+ else
+ found_dir=
+ found_la=
+ found_so=
+ found_a=
+ eval libname=\"$acl_libname_spec\" # typically: libname=lib$name
+ if test -n "$acl_shlibext"; then
+ shrext=".$acl_shlibext" # typically: shrext=.so
+ else
+ shrext=
+ fi
+ if test $use_additional = yes; then
+ dir="$additional_libdir"
+ if test -n "$acl_shlibext"; then
+ if test -f "$dir/$libname$shrext"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext"
+ else
+ if test "$acl_library_names_spec" = '$libname$shrext$versuffix'; then
+ ver=`(cd "$dir" && \
+ for f in "$libname$shrext".*; do echo "$f"; done \
+ | sed -e "s,^$libname$shrext\\\\.,," \
+ | sort -t '.' -n -r -k1,1 -k2,2 -k3,3 -k4,4 -k5,5 \
+ | sed 1q ) 2>/dev/null`
+ if test -n "$ver" && test -f "$dir/$libname$shrext.$ver"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext.$ver"
+ fi
+ else
+ eval library_names=\"$acl_library_names_spec\"
+ for f in $library_names; do
+ if test -f "$dir/$f"; then
+ found_dir="$dir"
+ found_so="$dir/$f"
+ break
+ fi
+ done
+ fi
+ fi
+ fi
+ if test "X$found_dir" = "X"; then
+ if test -f "$dir/$libname.$acl_libext"; then
+ found_dir="$dir"
+ found_a="$dir/$libname.$acl_libext"
+ fi
+ fi
+ if test "X$found_dir" != "X"; then
+ if test -f "$dir/$libname.la"; then
+ found_la="$dir/$libname.la"
+ fi
+ fi
+ fi
+ if test "X$found_dir" = "X"; then
+ for x in $LDFLAGS $LTLIBICONV; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ case "$x" in
+ -L*)
+ dir=`echo "X$x" | sed -e 's/^X-L//'`
+ if test -n "$acl_shlibext"; then
+ if test -f "$dir/$libname$shrext"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext"
+ else
+ if test "$acl_library_names_spec" = '$libname$shrext$versuffix'; then
+ ver=`(cd "$dir" && \
+ for f in "$libname$shrext".*; do echo "$f"; done \
+ | sed -e "s,^$libname$shrext\\\\.,," \
+ | sort -t '.' -n -r -k1,1 -k2,2 -k3,3 -k4,4 -k5,5 \
+ | sed 1q ) 2>/dev/null`
+ if test -n "$ver" && test -f "$dir/$libname$shrext.$ver"; then
+ found_dir="$dir"
+ found_so="$dir/$libname$shrext.$ver"
+ fi
+ else
+ eval library_names=\"$acl_library_names_spec\"
+ for f in $library_names; do
+ if test -f "$dir/$f"; then
+ found_dir="$dir"
+ found_so="$dir/$f"
+ break
+ fi
+ done
+ fi
+ fi
+ fi
+ if test "X$found_dir" = "X"; then
+ if test -f "$dir/$libname.$acl_libext"; then
+ found_dir="$dir"
+ found_a="$dir/$libname.$acl_libext"
+ fi
+ fi
+ if test "X$found_dir" != "X"; then
+ if test -f "$dir/$libname.la"; then
+ found_la="$dir/$libname.la"
+ fi
+ fi
+ ;;
+ esac
+ if test "X$found_dir" != "X"; then
+ break
+ fi
+ done
+ fi
+ if test "X$found_dir" != "X"; then
+ LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }-L$found_dir -l$name"
+ if test "X$found_so" != "X"; then
+ if test "$enable_rpath" = no \
+ || test "X$found_dir" = "X/usr/$acl_libdirstem" \
+ || test "X$found_dir" = "X/usr/$acl_libdirstem2"; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$found_so"
+ else
+ haveit=
+ for x in $ltrpathdirs; do
+ if test "X$x" = "X$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ ltrpathdirs="$ltrpathdirs $found_dir"
+ fi
+ if test "$acl_hardcode_direct" = yes; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$found_so"
+ else
+ if test -n "$acl_hardcode_libdir_flag_spec" && test "$acl_hardcode_minus_L" = no; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$found_so"
+ haveit=
+ for x in $rpathdirs; do
+ if test "X$x" = "X$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ rpathdirs="$rpathdirs $found_dir"
+ fi
+ else
+ haveit=
+ for x in $LDFLAGS $LIBICONV; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ if test "X$x" = "X-L$found_dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }-L$found_dir"
+ fi
+ if test "$acl_hardcode_minus_L" != no; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$found_so"
+ else
+ LIBICONV="${LIBICONV}${LIBICONV:+ }-l$name"
+ fi
+ fi
+ fi
+ fi
+ else
+ if test "X$found_a" != "X"; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$found_a"
+ else
+ LIBICONV="${LIBICONV}${LIBICONV:+ }-L$found_dir -l$name"
+ fi
+ fi
+ additional_includedir=
+ case "$found_dir" in
+ */$acl_libdirstem | */$acl_libdirstem/)
+ basedir=`echo "X$found_dir" | sed -e 's,^X,,' -e "s,/$acl_libdirstem/"'*$,,'`
+ if test "$name" = 'iconv'; then
+ LIBICONV_PREFIX="$basedir"
+ fi
+ additional_includedir="$basedir/include"
+ ;;
+ */$acl_libdirstem2 | */$acl_libdirstem2/)
+ basedir=`echo "X$found_dir" | sed -e 's,^X,,' -e "s,/$acl_libdirstem2/"'*$,,'`
+ if test "$name" = 'iconv'; then
+ LIBICONV_PREFIX="$basedir"
+ fi
+ additional_includedir="$basedir/include"
+ ;;
+ esac
+ if test "X$additional_includedir" != "X"; then
+ if test "X$additional_includedir" != "X/usr/include"; then
+ haveit=
+ if test "X$additional_includedir" = "X/usr/local/include"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux* | gnu* | k*bsd*-gnu) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ for x in $CPPFLAGS $INCICONV; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ if test "X$x" = "X-I$additional_includedir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_includedir"; then
+ INCICONV="${INCICONV}${INCICONV:+ }-I$additional_includedir"
+ fi
+ fi
+ fi
+ fi
+ fi
+ if test -n "$found_la"; then
+ save_libdir="$libdir"
+ case "$found_la" in
+ */* | *\\*) . "$found_la" ;;
+ *) . "./$found_la" ;;
+ esac
+ libdir="$save_libdir"
+ for dep in $dependency_libs; do
+ case "$dep" in
+ -L*)
+ additional_libdir=`echo "X$dep" | sed -e 's/^X-L//'`
+ if test "X$additional_libdir" != "X/usr/$acl_libdirstem" \
+ && test "X$additional_libdir" != "X/usr/$acl_libdirstem2"; then
+ haveit=
+ if test "X$additional_libdir" = "X/usr/local/$acl_libdirstem" \
+ || test "X$additional_libdir" = "X/usr/local/$acl_libdirstem2"; then
+ if test -n "$GCC"; then
+ case $host_os in
+ linux* | gnu* | k*bsd*-gnu) haveit=yes;;
+ esac
+ fi
+ fi
+ if test -z "$haveit"; then
+ haveit=
+ for x in $LDFLAGS $LIBICONV; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ if test "X$x" = "X-L$additional_libdir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_libdir"; then
+ LIBICONV="${LIBICONV}${LIBICONV:+ }-L$additional_libdir"
+ fi
+ fi
+ haveit=
+ for x in $LDFLAGS $LTLIBICONV; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ if test "X$x" = "X-L$additional_libdir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ if test -d "$additional_libdir"; then
+ LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }-L$additional_libdir"
+ fi
+ fi
+ fi
+ fi
+ ;;
+ -R*)
+ dir=`echo "X$dep" | sed -e 's/^X-R//'`
+ if test "$enable_rpath" != no; then
+ haveit=
+ for x in $rpathdirs; do
+ if test "X$x" = "X$dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ rpathdirs="$rpathdirs $dir"
+ fi
+ haveit=
+ for x in $ltrpathdirs; do
+ if test "X$x" = "X$dir"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ ltrpathdirs="$ltrpathdirs $dir"
+ fi
+ fi
+ ;;
+ -l*)
+ names_next_round="$names_next_round "`echo "X$dep" | sed -e 's/^X-l//'`
+ ;;
+ *.la)
+ names_next_round="$names_next_round "`echo "X$dep" | sed -e 's,^X.*/,,' -e 's,^lib,,' -e 's,\.la$,,'`
+ ;;
+ *)
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$dep"
+ LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }$dep"
+ ;;
+ esac
+ done
+ fi
+ else
+ LIBICONV="${LIBICONV}${LIBICONV:+ }-l$name"
+ LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }-l$name"
+ fi
+ fi
+ fi
+ done
+ done
+ if test "X$rpathdirs" != "X"; then
+ if test -n "$acl_hardcode_libdir_separator"; then
+ alldirs=
+ for found_dir in $rpathdirs; do
+ alldirs="${alldirs}${alldirs:+$acl_hardcode_libdir_separator}$found_dir"
+ done
+ acl_save_libdir="$libdir"
+ libdir="$alldirs"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$flag"
+ else
+ for found_dir in $rpathdirs; do
+ acl_save_libdir="$libdir"
+ libdir="$found_dir"
+ eval flag=\"$acl_hardcode_libdir_flag_spec\"
+ libdir="$acl_save_libdir"
+ LIBICONV="${LIBICONV}${LIBICONV:+ }$flag"
+ done
+ fi
+ fi
+ if test "X$ltrpathdirs" != "X"; then
+ for found_dir in $ltrpathdirs; do
+ LTLIBICONV="${LTLIBICONV}${LTLIBICONV:+ }-R$found_dir"
+ done
+ fi
+
+
+
+
+
+
+
+
+
+
+
+
+ am_save_CPPFLAGS="$CPPFLAGS"
+
+ for element in $INCICONV; do
+ haveit=
+ for x in $CPPFLAGS; do
+
+ acl_save_prefix="$prefix"
+ prefix="$acl_final_prefix"
+ acl_save_exec_prefix="$exec_prefix"
+ exec_prefix="$acl_final_exec_prefix"
+ eval x=\"$x\"
+ exec_prefix="$acl_save_exec_prefix"
+ prefix="$acl_save_prefix"
+
+ if test "X$x" = "X$element"; then
+ haveit=yes
+ break
+ fi
+ done
+ if test -z "$haveit"; then
+ CPPFLAGS="${CPPFLAGS}${CPPFLAGS:+ }$element"
+ fi
+ done
+
+
+ { $as_echo "$as_me:$LINENO: checking for iconv" >&5
+$as_echo_n "checking for iconv... " >&6; }
+if test "${am_cv_func_iconv+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+
+ am_cv_func_iconv="no, consider installing GNU libiconv"
+ am_cv_lib_iconv=no
+ cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h. */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h. */
+
+#include <stdlib.h>
+#include <iconv.h>
+
+int
+main ()
+{
+iconv_t cd = iconv_open("","");
+ iconv(cd,NULL,NULL,NULL,NULL);
+ iconv_close(cd);
+ ;
+ return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (ac_try="$ac_link"
+case "(($ac_try" in
+ *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+ *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+ (eval "$ac_link") 2>conftest.er1
+ ac_status=$?
+ grep -v '^ *+' conftest.er1 >conftest.err
+ rm -f conftest.er1
+ cat conftest.err >&5
+ $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+ (exit $ac_status); } && {
+ test -z "$ac_c_werror_flag" ||
+ test ! -s conftest.err
+ } && test -s conftest$ac_exeext && {
+ test "$cross_compiling" = yes ||
+ $as_test_x conftest$ac_exeext
+ }; then
+ am_cv_func_iconv=yes
+else
+ $as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+
+fi
+
+rm -rf conftest.dSYM
+rm -f core conftest.err conftest.$ac_objext conftest_ipa8_conftest.oo \
+ conftest$ac_exeext conftest.$ac_ext
+ if test "$am_cv_func_iconv" != yes; then
+ am_save_LIBS="$LIBS"
+ LIBS="$LIBS $LIBICONV"
+ cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h. */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h. */
+
+#include <stdlib.h>
+#include <iconv.h>
+
+int
+main ()
+{
+iconv_t cd = iconv_open("","");
+ iconv(cd,NULL,NULL,NULL,NULL);
+ iconv_close(cd);
+ ;
+ return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (ac_try="$ac_link"
+case "(($ac_try" in
+ *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+ *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+ (eval "$ac_link") 2>conftest.er1
+ ac_status=$?
+ grep -v '^ *+' conftest.er1 >conftest.err
+ rm -f conftest.er1
+ cat conftest.err >&5
+ $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+ (exit $ac_status); } && {
+ test -z "$ac_c_werror_flag" ||
+ test ! -s conftest.err
+ } && test -s conftest$ac_exeext && {
+ test "$cross_compiling" = yes ||
+ $as_test_x conftest$ac_exeext
+ }; then
+ am_cv_lib_iconv=yes
+ am_cv_func_iconv=yes
+else
+ $as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+
+fi
+
+rm -rf conftest.dSYM
+rm -f core conftest.err conftest.$ac_objext conftest_ipa8_conftest.oo \
+ conftest$ac_exeext conftest.$ac_ext
+ LIBS="$am_save_LIBS"
+ fi
+
+fi
+{ $as_echo "$as_me:$LINENO: result: $am_cv_func_iconv" >&5
+$as_echo "$am_cv_func_iconv" >&6; }
+ if test "$am_cv_func_iconv" = yes; then
+ { $as_echo "$as_me:$LINENO: checking for working iconv" >&5
+$as_echo_n "checking for working iconv... " >&6; }
+if test "${am_cv_func_iconv_works+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+
+ am_save_LIBS="$LIBS"
+ if test $am_cv_lib_iconv = yes; then
+ LIBS="$LIBS $LIBICONV"
+ fi
+ if test "$cross_compiling" = yes; then
+
+ case "$host_os" in
+ aix* | hpux*) am_cv_func_iconv_works="guessing no" ;;
+ *) am_cv_func_iconv_works="guessing yes" ;;
+ esac
+
+else
+ cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h. */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h. */
+
+#include <iconv.h>
+#include <string.h>
+int main ()
+{
+ int result = 0;
+ /* Test against AIX 5.1 bug: Failures are not distinguishable from successful
+ returns. */
+ {
+ iconv_t cd_utf8_to_88591 = iconv_open ("ISO8859-1", "UTF-8");
+ if (cd_utf8_to_88591 != (iconv_t)(-1))
+ {
+ static const char input[] = "\342\202\254"; /* EURO SIGN */
+ char buf[10];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_utf8_to_88591,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res == 0)
+ result |= 1;
+ iconv_close (cd_utf8_to_88591);
+ }
+ }
+ /* Test against Solaris 10 bug: Failures are not distinguishable from
+ successful returns. */
+ {
+ iconv_t cd_ascii_to_88591 = iconv_open ("ISO8859-1", "646");
+ if (cd_ascii_to_88591 != (iconv_t)(-1))
+ {
+ static const char input[] = "\263";
+ char buf[10];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_ascii_to_88591,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res == 0)
+ result |= 2;
+ iconv_close (cd_ascii_to_88591);
+ }
+ }
+ /* Test against AIX 6.1..7.1 bug: Buffer overrun. */
+ {
+ iconv_t cd_88591_to_utf8 = iconv_open ("UTF-8", "ISO-8859-1");
+ if (cd_88591_to_utf8 != (iconv_t)(-1))
+ {
+ static const char input[] = "\304";
+ static char buf[2] = { (char)0xDE, (char)0xAD };
+ const char *inptr = input;
+ size_t inbytesleft = 1;
+ char *outptr = buf;
+ size_t outbytesleft = 1;
+ size_t res = iconv (cd_88591_to_utf8,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if (res != (size_t)(-1) || outptr - buf > 1 || buf[1] != (char)0xAD)
+ result |= 4;
+ iconv_close (cd_88591_to_utf8);
+ }
+ }
+#if 0 /* This bug could be worked around by the caller. */
+ /* Test against HP-UX 11.11 bug: Positive return value instead of 0. */
+ {
+ iconv_t cd_88591_to_utf8 = iconv_open ("utf8", "iso88591");
+ if (cd_88591_to_utf8 != (iconv_t)(-1))
+ {
+ static const char input[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337";
+ char buf[50];
+ const char *inptr = input;
+ size_t inbytesleft = strlen (input);
+ char *outptr = buf;
+ size_t outbytesleft = sizeof (buf);
+ size_t res = iconv (cd_88591_to_utf8,
+ (char **) &inptr, &inbytesleft,
+ &outptr, &outbytesleft);
+ if ((int)res > 0)
+ result |= 8;
+ iconv_close (cd_88591_to_utf8);
+ }
+ }
+#endif
+ /* Test against HP-UX 11.11 bug: No converter from EUC-JP to UTF-8 is
+ provided. */
+ if (/* Try standardized names. */
+ iconv_open ("UTF-8", "EUC-JP") == (iconv_t)(-1)
+ /* Try IRIX, OSF/1 names. */
+ && iconv_open ("UTF-8", "eucJP") == (iconv_t)(-1)
+ /* Try AIX names. */
+ && iconv_open ("UTF-8", "IBM-eucJP") == (iconv_t)(-1)
+ /* Try HP-UX names. */
+ && iconv_open ("utf8", "eucJP") == (iconv_t)(-1))
+ result |= 16;
+ return result;
+}
+_ACEOF
+rm -f conftest$ac_exeext
+if { (ac_try="$ac_link"
+case "(($ac_try" in
+ *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+ *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+ (eval "$ac_link") 2>&5
+ ac_status=$?
+ $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+ (exit $ac_status); } && { ac_try='./conftest$ac_exeext'
+ { (case "(($ac_try" in
+ *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+ *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+ (eval "$ac_try") 2>&5
+ ac_status=$?
+ $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+ (exit $ac_status); }; }; then
+ am_cv_func_iconv_works=yes
+else
+ $as_echo "$as_me: program exited with status $ac_status" >&5
+$as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+( exit $ac_status )
+am_cv_func_iconv_works=no
+fi
+rm -rf conftest.dSYM
+rm -f core *.core core.conftest.* gmon.out bb.out conftest$ac_exeext conftest.$ac_objext conftest.$ac_ext
+fi
+
+
+ LIBS="$am_save_LIBS"
+
+fi
+{ $as_echo "$as_me:$LINENO: result: $am_cv_func_iconv_works" >&5
+$as_echo "$am_cv_func_iconv_works" >&6; }
+ case "$am_cv_func_iconv_works" in
+ *no) am_func_iconv=no am_cv_lib_iconv=no ;;
+ *) am_func_iconv=yes ;;
+ esac
+ else
+ am_func_iconv=no am_cv_lib_iconv=no
+ fi
+ if test "$am_func_iconv" = yes; then
+
+cat >>confdefs.h <<\_ACEOF
+#define HAVE_ICONV 1
+_ACEOF
+
+ fi
+ if test "$am_cv_lib_iconv" = yes; then
+ { $as_echo "$as_me:$LINENO: checking how to link with libiconv" >&5
+$as_echo_n "checking how to link with libiconv... " >&6; }
+ { $as_echo "$as_me:$LINENO: result: $LIBICONV" >&5
+$as_echo "$LIBICONV" >&6; }
+ else
+ CPPFLAGS="$am_save_CPPFLAGS"
+ LIBICONV=
+ LTLIBICONV=
+ fi
+
+
+
+ if test "$am_cv_func_iconv" = yes; then
+ { $as_echo "$as_me:$LINENO: checking for iconv declaration" >&5
+$as_echo_n "checking for iconv declaration... " >&6; }
+ if test "${am_cv_proto_iconv+set}" = set; then
+ $as_echo_n "(cached) " >&6
+else
+
+ cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h. */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h. */
+
+#include <stdlib.h>
+#include <iconv.h>
+extern
+#ifdef __cplusplus
+"C"
+#endif
+#if defined(__STDC__) || defined(_MSC_VER) || defined(__cplusplus)
+size_t iconv (iconv_t cd, char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);
+#else
+size_t iconv();
+#endif
+
+int
+main ()
+{
+
+ ;
+ return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext
+if { (ac_try="$ac_compile"
+case "(($ac_try" in
+ *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
+ *) ac_try_echo=$ac_try;;
+esac
+eval ac_try_echo="\"\$as_me:$LINENO: $ac_try_echo\""
+$as_echo "$ac_try_echo") >&5
+ (eval "$ac_compile") 2>conftest.er1
+ ac_status=$?
+ grep -v '^ *+' conftest.er1 >conftest.err
+ rm -f conftest.er1
+ cat conftest.err >&5
+ $as_echo "$as_me:$LINENO: \$? = $ac_status" >&5
+ (exit $ac_status); } && {
+ test -z "$ac_c_werror_flag" ||
+ test ! -s conftest.err
+ } && test -s conftest.$ac_objext; then
+ am_cv_proto_iconv_arg1=""
+else
+ $as_echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+ am_cv_proto_iconv_arg1="const"
+fi
+
+rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
+ am_cv_proto_iconv="extern size_t iconv (iconv_t cd, $am_cv_proto_iconv_arg1 char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);"
+fi
+
+ am_cv_proto_iconv=`echo "$am_cv_proto_iconv" | tr -s ' ' | sed -e 's/( /(/'`
+ { $as_echo "$as_me:$LINENO: result:
+ $am_cv_proto_iconv" >&5
+$as_echo "
+ $am_cv_proto_iconv" >&6; }
+
+cat >>confdefs.h <<_ACEOF
+#define ICONV_CONST $am_cv_proto_iconv_arg1
+_ACEOF
+
+
+ fi
+
+
# Check for Tcl configuration script tclConfig.sh
if test "$with_tcl" = yes; then
for ac_prog in tclsh tcl tclsh8.5 tclsh85 tclsh8.4 tclsh84 tclsh8.3 tclsh83
diff --git a/configure.in b/configure.in
index fe25419..86a8cf7 100644
--- a/configure.in
+++ b/configure.in
@@ -790,7 +790,7 @@ LDFLAGS="$LDFLAGS $LIBDIRS"
AC_ARG_VAR(LDFLAGS_EX, [extra linker flags for linking executables only])
AC_ARG_VAR(LDFLAGS_SL, [extra linker flags for linking shared libraries only])
-PGAC_PROG_LD
+AC_LIB_PROG_LD_GNU
AC_SUBST(LD)
AC_SUBST(with_gnu_ld)
case $host_os in sysv5*)
@@ -1800,6 +1800,8 @@ if test "$enable_nls" = yes ; then
PGAC_CHECK_GETTEXT
fi
+AM_ICONV
+
# Check for Tcl configuration script tclConfig.sh
if test "$with_tcl" = yes; then
PGAC_PATH_TCLCONFIGSH([$with_tclconfig])
diff --git a/src/backend/utils/mb/encnames.c b/src/backend/utils/mb/encnames.c
index 772d4a5..82b130e 100644
--- a/src/backend/utils/mb/encnames.c
+++ b/src/backend/utils/mb/encnames.c
@@ -350,58 +350,6 @@ pg_enc2name pg_enc2name_tbl[] =
DEF_ENC2NAME(SHIFT_JIS_2004, 932)
};
-/* ----------
- * These are encoding names for gettext.
- *
- * This covers all encodings except MULE_INTERNAL, which is alien to gettext.
- * ----------
- */
-pg_enc2gettext pg_enc2gettext_tbl[] =
-{
- {PG_SQL_ASCII, "US-ASCII"},
- {PG_UTF8, "UTF-8"},
- {PG_LATIN1, "LATIN1"},
- {PG_LATIN2, "LATIN2"},
- {PG_LATIN3, "LATIN3"},
- {PG_LATIN4, "LATIN4"},
- {PG_ISO_8859_5, "ISO-8859-5"},
- {PG_ISO_8859_6, "ISO_8859-6"},
- {PG_ISO_8859_7, "ISO-8859-7"},
- {PG_ISO_8859_8, "ISO-8859-8"},
- {PG_LATIN5, "LATIN5"},
- {PG_LATIN6, "LATIN6"},
- {PG_LATIN7, "LATIN7"},
- {PG_LATIN8, "LATIN8"},
- {PG_LATIN9, "LATIN-9"},
- {PG_LATIN10, "LATIN10"},
- {PG_KOI8R, "KOI8-R"},
- {PG_KOI8U, "KOI8-U"},
- {PG_WIN1250, "CP1250"},
- {PG_WIN1251, "CP1251"},
- {PG_WIN1252, "CP1252"},
- {PG_WIN1253, "CP1253"},
- {PG_WIN1254, "CP1254"},
- {PG_WIN1255, "CP1255"},
- {PG_WIN1256, "CP1256"},
- {PG_WIN1257, "CP1257"},
- {PG_WIN1258, "CP1258"},
- {PG_WIN866, "CP866"},
- {PG_WIN874, "CP874"},
- {PG_EUC_CN, "EUC-CN"},
- {PG_EUC_JP, "EUC-JP"},
- {PG_EUC_KR, "EUC-KR"},
- {PG_EUC_TW, "EUC-TW"},
- {PG_EUC_JIS_2004, "EUC-JP"},
- {PG_SJIS, "SHIFT-JIS"},
- {PG_BIG5, "BIG5"},
- {PG_GBK, "GBK"},
- {PG_UHC, "UHC"},
- {PG_GB18030, "GB18030"},
- {PG_JOHAB, "JOHAB"},
- {PG_SHIFT_JIS_2004, "SHIFT_JISX0213"},
- {0, NULL}
-};
-
/* ----------
* Encoding checks, for error returns -1 else encoding id
diff --git a/src/backend/utils/mb/mbutils.c b/src/backend/utils/mb/mbutils.c
index 6d1cd8e..243c919 100644
--- a/src/backend/utils/mb/mbutils.c
+++ b/src/backend/utils/mb/mbutils.c
@@ -901,23 +901,18 @@ static bool
raw_pg_bind_textdomain_codeset(const char *domainname, int encoding)
{
bool elog_ok = (CurrentMemoryContext != NULL);
- int i;
+ const char *codeset;
- for (i = 0; pg_enc2gettext_tbl[i].name != NULL; i++)
+ codeset = pg_get_gettext_codeset_from_encoding(encoding);
+ if (codeset != NULL)
{
- if (pg_enc2gettext_tbl[i].encoding == encoding)
- {
- if (bind_textdomain_codeset(domainname,
- pg_enc2gettext_tbl[i].name) != NULL)
- return true;
-
- if (elog_ok)
- elog(LOG, "bind_textdomain_codeset failed");
- else
- write_stderr("bind_textdomain_codeset failed");
+ if (bind_textdomain_codeset(domainname, codeset) != NULL)
+ return true;
- break;
- }
+ if (elog_ok)
+ elog(LOG, "bind_textdomain_codeset failed");
+ else
+ write_stderr("bind_textdomain_codeset failed");
}
return false;
diff --git a/src/bin/pg_dump/pg_restore.c b/src/bin/pg_dump/pg_restore.c
index 985c826..b8123bd 100644
--- a/src/bin/pg_dump/pg_restore.c
+++ b/src/bin/pg_dump/pg_restore.c
@@ -61,8 +61,15 @@ extern int optind;
#include <locale.h>
#endif
+#ifdef HAVE_ICONV
+#include <iconv.h>
+#include "mb/pg_wchar.h"
+#endif
+
static void usage(const char *progname);
+static int get_command_line_encoding(void);
+static void convert_opts(RestoreOptions *opts, int dest, int src);
typedef struct option optType;
@@ -72,6 +79,7 @@ main(int argc, char **argv)
RestoreOptions *opts;
int c;
int exit_code;
+ int cli_encoding;
int numWorkers = 1;
Archive *AH;
char *inputFileSpec;
@@ -377,6 +385,18 @@ main(int argc, char **argv)
*/
on_exit_close_archive(AH);
+ /*
+ * If the archive's encoding is different from the terminal's, we have to
+ * convert the object names passed on the command line to the archive's
+ * encoding, so that they can be matched against names in the archive.
+ */
+ if (AH->encoding > 0)
+ {
+ cli_encoding = get_command_line_encoding();
+ if (cli_encoding > 0 && AH->encoding != get_command_line_encoding())
+ convert_opts(opts, AH->encoding, cli_encoding);
+ }
+
/* Let the archiver know how noisy to be */
AH->verbose = opts->verbose;
@@ -476,3 +496,154 @@ usage(const char *progname)
printf(_("\nIf no input file name is supplied, then standard input is used.\n\n"));
printf(_("Report bugs to <pgsql-bugs@postgresql.org>.\n"));
}
+
+/*
+ * Get the encoding the command-line arguments are in.
+ */
+static int
+get_command_line_encoding(void)
+{
+ int encoding;
+ char *s;
+
+ if ((s = getenv("PGCLIENTENCODING")) != NULL)
+ encoding = pg_char_to_encoding(s);
+ else
+ encoding = pg_get_encoding_from_locale(NULL, true);
+
+ return encoding;
+}
+
+#ifdef HAVE_ICONV
+/*
+ * A helper function to convert a string using iconv(3).
+ *
+ * Returns a pointer to a static buffer that's only valid until the next call
+ * of the function. The caller should not attempt to free it.
+ */
+static char *
+convert_encoding(iconv_t cd, char *from)
+{
+ char *inbuf;
+ size_t inbytesleft;
+ static int outbuflen = 0;
+ static char *outbuf = NULL;
+ char *outbuf_p;
+ size_t outbytesleft;
+ size_t nconv;
+
+ inbuf = from;
+ inbytesleft = strlen(inbuf);
+
+ if (outbuf == NULL)
+ {
+ outbuflen = 50;
+ outbuf = pg_malloc(outbuflen);
+ }
+ outbytesleft = outbuflen - 1;
+ outbuf_p = outbuf;
+
+ while (inbytesleft > 0)
+ {
+ nconv = iconv(cd, &inbuf, &inbytesleft, &outbuf_p, &outbytesleft);
+
+ if (nconv == (size_t) -1)
+ {
+ if (errno == E2BIG)
+ {
+ /* enlarge output buffer */
+ size_t outsz = outbuf_p - outbuf;
+ outbuflen *= 2;
+ outbuf = pg_realloc(outbuf, outbuflen);
+ outbuf_p = outbuf + outsz;
+ outbytesleft = outbuflen - outsz - 1;
+ }
+ else
+ {
+ /* Error. */
+ return NULL;
+ }
+ }
+ }
+ *outbuf_p = '\0';
+ return outbuf;
+}
+
+static void
+convert_string(iconv_t cd, char **str)
+{
+ char *newstr;
+
+ if (*str == NULL)
+ return;
+
+ newstr = convert_encoding(cd, *str);
+ if (newstr != NULL)
+ {
+ pg_free(*str);
+ *str = pg_strdup(newstr);
+ }
+}
+
+static void
+convert_simple_string_list(iconv_t cd, SimpleStringList *list)
+{
+ SimpleStringListCell *oldcell;
+ SimpleStringList newlist = { NULL, NULL };
+
+ for (oldcell = list->head; oldcell != NULL; oldcell = oldcell->next)
+ {
+ char *newstr = convert_encoding(cd, oldcell->val);
+
+ simple_string_list_append(&newlist, newstr ? newstr : oldcell->val);
+ }
+ memcpy(list, &newlist, sizeof(SimpleStringList));
+}
+#endif /* HAVE_ICONV */
+
+/*
+ * Convert any object names in RestoreOptions from 'src' encoding to 'dest'.
+ */
+static void
+convert_opts(RestoreOptions *opts, int dest, int src)
+{
+#ifdef HAVE_ICONV
+ const char *dest_charset;
+ const char *src_charset;
+ iconv_t cd;
+
+ src_charset = pg_get_gettext_codeset_from_encoding(src);
+ if (src_charset == NULL)
+ {
+ fprintf(stderr, _("could not transcode cli arguments\n"));
+ return;
+ }
+
+ dest_charset = pg_get_gettext_codeset_from_encoding(dest);
+ if (dest_charset == NULL)
+ {
+ fprintf(stderr, _("could not transcode cli arguments\n"));
+ return;
+ }
+
+ cd = iconv_open(dest_charset, src_charset);
+
+ /* superuser username? username? use_role? */
+ convert_string(cd, &opts->schemaNames);
+ convert_string(cd, &opts->functionNames);
+ convert_string(cd, &opts->indexNames);
+ convert_string(cd, &opts->triggerNames);
+ convert_simple_string_list(cd, &opts->tableNames);
+
+ iconv_close(cd);
+#else
+ /*
+ * Without iconv(3), We cannot convert encodings in the client-side.
+ * Take the command-line arguments as is. even though we know they are in
+ * different encoding than the archive. It will still work for object
+ * names that contain only ASCII characters, and that's the best we can
+ * do.
+ */
+ return;
+#endif /* HAVE_ICONV */
+}
diff --git a/src/include/pg_config.h.in b/src/include/pg_config.h.in
index 033127b..90b69f9 100644
--- a/src/include/pg_config.h.in
+++ b/src/include/pg_config.h.in
@@ -224,6 +224,9 @@
/* Define to 1 if you have the `history_truncate_file' function. */
#undef HAVE_HISTORY_TRUNCATE_FILE
+/* Define if you have the iconv() function and it works. */
+#undef HAVE_ICONV
+
/* Define to 1 if you have the <ieeefp.h> header file. */
#undef HAVE_IEEEFP_H
@@ -653,6 +656,9 @@
/* Define to 1 if your compiler understands __VA_ARGS__ in macros. */
#undef HAVE__VA_ARGS
+/* Define as const if the declaration of iconv() needs const. */
+#undef ICONV_CONST
+
/* Define to the appropriate snprintf format for 64-bit ints. */
#undef INT64_FORMAT
diff --git a/src/include/port.h b/src/include/port.h
index 5ef4b0a..2a45fe6 100644
--- a/src/include/port.h
+++ b/src/include/port.h
@@ -451,6 +451,7 @@ extern void qsort_arg(void *base, size_t nel, size_t elsize,
/* port/chklocale.c */
extern int pg_get_encoding_from_locale(const char *ctype, bool write_message);
+extern const char *pg_get_gettext_codeset_from_encoding(int encoding);
#if defined(WIN32) && !defined(FRONTEND)
extern int pg_codepage_to_encoding(UINT cp);
diff --git a/src/port/chklocale.c b/src/port/chklocale.c
index 8b8862f..d9388d8 100644
--- a/src/port/chklocale.c
+++ b/src/port/chklocale.c
@@ -187,6 +187,57 @@ static const struct encoding_match encoding_match_list[] = {
{PG_SQL_ASCII, NULL} /* end marker */
};
+/*
+ * These are encoding names for gettext and iconv.
+ *
+ * This covers all encodings except MULE_INTERNAL, which is alien to gettext.
+ */
+pg_enc2gettext pg_enc2gettext_tbl[] =
+{
+ {PG_SQL_ASCII, "US-ASCII"},
+ {PG_UTF8, "UTF-8"},
+ {PG_LATIN1, "LATIN1"},
+ {PG_LATIN2, "LATIN2"},
+ {PG_LATIN3, "LATIN3"},
+ {PG_LATIN4, "LATIN4"},
+ {PG_ISO_8859_5, "ISO-8859-5"},
+ {PG_ISO_8859_6, "ISO_8859-6"},
+ {PG_ISO_8859_7, "ISO-8859-7"},
+ {PG_ISO_8859_8, "ISO-8859-8"},
+ {PG_LATIN5, "LATIN5"},
+ {PG_LATIN6, "LATIN6"},
+ {PG_LATIN7, "LATIN7"},
+ {PG_LATIN8, "LATIN8"},
+ {PG_LATIN9, "LATIN-9"},
+ {PG_LATIN10, "LATIN10"},
+ {PG_KOI8R, "KOI8-R"},
+ {PG_KOI8U, "KOI8-U"},
+ {PG_WIN1250, "CP1250"},
+ {PG_WIN1251, "CP1251"},
+ {PG_WIN1252, "CP1252"},
+ {PG_WIN1253, "CP1253"},
+ {PG_WIN1254, "CP1254"},
+ {PG_WIN1255, "CP1255"},
+ {PG_WIN1256, "CP1256"},
+ {PG_WIN1257, "CP1257"},
+ {PG_WIN1258, "CP1258"},
+ {PG_WIN866, "CP866"},
+ {PG_WIN874, "CP874"},
+ {PG_EUC_CN, "EUC-CN"},
+ {PG_EUC_JP, "EUC-JP"},
+ {PG_EUC_KR, "EUC-KR"},
+ {PG_EUC_TW, "EUC-TW"},
+ {PG_EUC_JIS_2004, "EUC-JP"},
+ {PG_SJIS, "SHIFT-JIS"},
+ {PG_BIG5, "BIG5"},
+ {PG_GBK, "GBK"},
+ {PG_UHC, "UHC"},
+ {PG_GB18030, "GB18030"},
+ {PG_JOHAB, "JOHAB"},
+ {PG_SHIFT_JIS_2004, "SHIFT_JISX0213"},
+ {0, NULL}
+};
+
#ifdef WIN32
/*
* On Windows, use CP<code page number> instead of the nl_langinfo() result
@@ -392,6 +443,25 @@ pg_get_encoding_from_locale(const char *ctype, bool write_message)
free(sys);
return -1;
}
+
+/*
+ * Returns the gettext/iconv codeset name for a PostgreSQL encoding
+ * (or NULL if it cannot be determined).
+ */
+const char *
+pg_get_gettext_codeset_from_encoding(int encoding)
+{
+ int i;
+
+ for (i = 0; pg_enc2gettext_tbl[i].name != NULL; i++)
+ {
+ if (pg_enc2gettext_tbl[i].encoding == encoding)
+ return pg_enc2gettext_tbl[i].name;
+ }
+
+ return NULL;
+}
+
#else /* (HAVE_LANGINFO_H && CODESET) || WIN32 */
/*
--
1.7.10.4
Re: pg_dump/restore encoding woes
On 08/27/2013 10:36 AM, Heikki Linnakangas wrote:
0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patch
Separates pg_dump -E from PGCLIENTENCODING.
Wouldn't it be better to do this another way? Separating these two will
be confusing, to say the least, as well as inconsistent with what os
done elsewhere. Why not provide a new option that specifically allows a
client encoding that only applies during the query phase?
cheers
andrew
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On 27.08.2013 18:03, Andrew Dunstan wrote:
On 08/27/2013 10:36 AM, Heikki Linnakangas wrote:
0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patch
Separates pg_dump -E from PGCLIENTENCODING.
Wouldn't it be better to do this another way? Separating these two will
be confusing, to say the least, as well as inconsistent with what os
done elsewhere.
What would it be inconsistent with? There is no -E option in other
client tools, pg_dump is unique in that. initdb does have a -E option,
but that *is* separate from PGCLIENTENCODING, so if anything the current
situation is inconsistent.
Why not provide a new option that specifically allows a
client encoding that only applies during the query phase?
Well, that would be different from all the other client programs. And
doesn't seem any less confusing to me.
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On 08/27/2013 11:14 AM, Heikki Linnakangas wrote:
On 27.08.2013 18:03, Andrew Dunstan wrote:
On 08/27/2013 10:36 AM, Heikki Linnakangas wrote:
0001-Divorce-pg_dump-E-option-from-PGCLIENTENCODING.patch
Separates pg_dump -E from PGCLIENTENCODING.
Wouldn't it be better to do this another way? Separating these two will
be confusing, to say the least, as well as inconsistent with what os
done elsewhere.What would it be inconsistent with? There is no -E option in other
client tools, pg_dump is unique in that. initdb does have a -E option,
but that *is* separate from PGCLIENTENCODING, so if anything the
current situation is inconsistent.
Yeah, you're right, I was probably thinking of initdb, although it
doesn't so much separate these as ignore PGCLIENTENCODING completely.
I guess I'm mainly concerned that we're going to make one of these do
something different, and it will be hard to remember which is which, at
least for me (brain cells are no doubt dying at an ever increasing rate
as I approach my seventh decade.)
cheers
andrew
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On Tue, 2013-08-27 at 17:36 +0300, Heikki Linnakangas wrote:
Here is a set of three patches that I've been working on:
There is a compiler warning:
pg_backup_db.c:243:0: warning: "PARAMS_ARRAY_SIZE" redefined [enabled by default]
pg_backup_db.c:139:0: note: this is the location of the previous definition
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On 27 August 2013 20:06, Heikki Linnakangas <hlinnakangas@vmware.com> wrote:
On 26.08.2013 18:59, Tom Lane wrote:
Heikki Linnakangas<hlinnakangas@**vmware.com <hlinnakangas@vmware.com>>
writes:The pg_dump -E option just sets client_encoding, but I think it would be
better for -E to only set the encoding used in the dump, and
PGCLIENTENCODING env variable (if set) was used to determine the
encoding of the command-line arguments. Opinions?I think this is going to be a lot easier said than done, but feel free
to see if you can make it work. (As you point out, we don't have
any client-side encoding conversion infrastructure, but I don't see
how you're going to make this work without it.)First set client_encoding to PGCLIENTENCODING (ie. let libpq do its
default thing), and run the queries that fetch the OIDs of any matching
tables. Only after doing that, set client_encoding to that specified by -E.
That's actually pretty easy to do, as pg_dump already issues all the
queries that include user-given strings first.There's one small wrinkle in that: if the dump fails because the server
sends an error, the error would come from the server in the -E encoding,
because that's used as the client_encoding after the initial queries.
Logically, the errors should be printed in PGCLIENTENCODING. But I think we
can live with that.The pg_restore part is more difficult, as pg_restore needs to work without
a database connection at all. There the conversion has to be done
client-side.A second issue is whether we should divorce -E and PGCLIENTENCODING like
that, when they have always meant the same thing. You mentioned the
alternative of looking at pg_dump's locale environment to determine the
command line encoding --- would that be better?Hmm. I wasn't thinking of looking at the locale environment as an
alternative to the divorce, just as a way to determine the default for the
client encoding if PGCLIENTENCODING is not set.It would be good for pg_dump to be consistent with other tools (reindexdb
etc.). If we say that LC_CTYPE determines the encoding of command-line
options, regardless of PGCLIENTENCODING, we should do the same in other
tools, too. And that would be weird for the other tools. So no, I don't
think we should do that.My plan is to assume that the command-line options are in the "client
encoding", and the client encoding is determined by the following things,
in this order:1. client_encoding setting given explicitly in the command line, as part
of the connection string.
2. PGCLIENTENCODING
3. LC_CTYPE
4. same as server_encodingThe encoding to be used in the pg_dump output is a different concept, and
there are reasons to *not* want the dump to be in the client encoding.
Hence using server_encoding for that would be a better default than the
current client encoding. This isn't so visible today, as client_encoding
defaults to server_encoding anyway, but it will be if we set
client_encoding='auto' (ie. look at LC_CTYPE) by default.There are certainly cases where you'd want to use the client encoding as
the dump output encoding. For example if you pipe the pg_dump output to
some custom script that mangles it. Or if you just want to look at the
output, ie. if the output goes to a terminal. But for the usual case of
taking a backup, using the server encoding makes more sense. One option is
to check if the output is a tty (like psql does), and output the dump in
client or server encoding depending on that (if -E is not given).So yeah, I think we should divorce -E and PGCLIENTENCODING. It feels that
using PGCLIENTENCODING to specify the output encoding was more accidental
than on purpose. Looking at the history, the implementation has been that
way forever, but the docs were adjusted by commit b524cb36 in 2005 to
document that fact.Here is a set of three patches that I've been working on:
0001-Divorce-pg_dump-E-option-**from-PGCLIENTENCODING.patch
Separates pg_dump -E from PGCLIENTENCODING.
Since AH->encoding is now used only to represent dump encoding, we should
rename it as AH->dump_encoding.
-----
The standard_conforming_strings parameter is currently set in
setup_connection. You have moved it in setup_dump_encoding(). Is it in any
way related to encoding ? If not, I think we should keep it in
setup_connection().
-----
If a user supplies pg_dump --role=ROLENAME, we do the SET-ROLE in
setup_connection. The role name is in client encoding. In case of parallel
pg_dump, the setup_connection() in the parent process does it right, but in
the worker processes, the client_encoding is already set to the
dump_encoding when -E is given (parent has already updated it's
client_encoding to dump_encoding and I think the child inherits
client_encoding from parent), and then in the child when SET-ROLE is called
through setup_connection, it does not work because role name is in client
encoding, not dump encoding.
$ pg_dump -t "pöö" -E LATIN1 --role=uöö postgres
# Above succeeds
$ `which pg_dump` -t "pöö" -E LATIN1 -j 5 -f dumpdir -Fd postgres
--role=uöö
pg_dump: [archiver (db)] query failed: ERROR: role "uöö" does not exist
pg_dump: [parallel archiver] query was: SET ROLE "uöö"
pg_dump: [archiver (db)] query failed: ERROR: role "uöö" does not exist
pg_dump: [archiver (db)] query failed: ERROR: role "uöö" does not exist
pg_dump: [archiver (db)] query failed: ERROR: role "uöö" does not exist
pg_dump: [archiver (db)] query failed: ERROR: role "uöö" does not exist
0002-Set-client_encoding-auto-**in-all-client-utilities.patch
Set client_encoding='auto' in all the client utilities, like we already
did in psql. This fixes e.g "reindexdb -t" with a table name with non-ASCII
chars
This patch looks good, I don't have any issues with this.
0003-Convert-object-names-to-**archive-encoding-before-matc.**patch
Use iconv(3) in pg_restore to do encoding conversion in the client. This
involves a lot of autoconf changes that I'm not 100% sure about, other than
that it's pretty straightforward.
I haven't looked into this one yet.
Show quoted text
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Re: pg_dump/restore encoding woes
On 25 September 2013 12:49, Amit Khandekar
<amit.khandekar@enterprisedb.com>wrote:
0003-Convert-object-names-to-**archive-encoding-before-matc.**patch
Use iconv(3) in pg_restore to do encoding conversion in the client. This
involves a lot of autoconf changes that I'm not 100% sure about, other than
that it's pretty straightforward.I haven't looked into this one yet.
I have verified that the *.m4 files that you have included match the ones
available through gettext. In general, the iconv-related config changes
look good.
---------------
I wanted to compare the new configure file that is included in the patch
with one that I tried generating myself from the new configure.in file
modified by your patch. It gives me :
Autoconf version 2.63 is required.
My autoconf is 2.68. Do we have to use 2.63 only ?
---------------
pg_restore.c:
cli_encoding can be declared in the if block.
---------------
pg_restore.c:
get_command_line_encoding() is called twice here :
cli_encoding = get_command_line_encoding();
if (cli_encoding > 0 && AH->encoding != get_command_line_encoding())
---------------
pg_restore.c:
You actually intended to use convert_simple_string_list() rather than
convert_string() below:
convert_string(cd, &opts->schemaNames);
convert_string(cd, &opts->functionNames);
convert_string(cd, &opts->indexNames);
convert_string(cd, &opts->triggerNames);
Show quoted text
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers