Schemas not available for pl/pgsql %TYPE....

Sean Chittenden <sean@chittenden.org> writes:

Call me crazy, but shouldn't the following work? :~|

Sure should. Want to fix plpgsql's parser?

regards, tom lane

Sean Chittenden <sean@chittenden.org> writes:

Call me crazy, but shouldn't the following work? :~|

Sure should. Want to fix plpgsql's parser?

Why not: I've never been one to avoid strapping on 4tons in rocks and
jumping into the deep end. ::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

--
Sean Chittenden

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

regards, tom lane

Tom Lane wrote:

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

Added to TODO:

o Make PL/PgSQL %TYPE schema-aware

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies
on the number of names to give it some idea of what to look for, and
those rules are probably all toast now. Please come up with a
sketch of what you think the behavior should be before you start
hacking code.

Not a problem there. I walked around the code for a bit, made a few
hacks to see how things are working, and I can tell you strait up that
if you'd like this by 7.3, it won't be happening from me. <:~) I'm
stretched kinda thin as is and don't think I'll be able to get this
working correctly with time to test by release. I can send you the
patch I've got for the lexer, but that was chump. What I was going to
do could be totally wrong, but...

* Change the lexer to recognize schema.table.column%TYPE as a token
and was going to create parse_tripwordtype() that'd look up the
table and column in the appropriate schema and would return the
appropriate type.

If I were lazy, I'd just unshift the schema off of the token and
return what comes back from parse_dblwordtype(), but that doesn't
strike me as correct for something that's performance sensitive.
Beyond doing that, I'm at a loss. :-/ Thoughts? -sc

--
Sean Chittenden

Sean Chittenden wrote:

Not a problem there. I walked around the code for a bit, made a few
hacks to see how things are working, and I can tell you strait up that
if you'd like this by 7.3, it won't be happening from me. <:~) I'm
stretched kinda thin as is and don't think I'll be able to get this
working correctly with time to test by release. I can send you the
patch I've got for the lexer, but that was chump.

If you want to send me what you've done so far, I'll take a look and see
if I can figure it out. I think this is probably a must do item for 7.3.

Any further guidance or thoughts?

Thanks,

Joe

Tom Lane wrote:

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

Attached is a diff -c format proposal to fix this. I've also attached a short
test script. Seems to work OK and passes all regression tests.

Here's a breakdown of how I understand plpgsql's "Special word rules" -- I
think it illustrates the behavior reasonably well. New functions added by this
patch are plpgsql_parse_tripwordtype and plpgsql_parse_dblwordrowtype:

============================================================================
Identifiers (represents) parsing function
----------------------------------------------------------------------------
identifier plpgsql_parse_word
tg_argv
T_LABEL (label)
T_VARIABLE (variable)
T_RECORD (record)
T_ROW (row)
----------------------------------------------------------------------------
identifier.identifier plpgsql_parse_dblword
T_LABEL
T_VARIABLE (label.variable)
T_RECORD (label.record)
T_ROW (label.row)
T_RECORD
T_VARIABLE (record.variable)
T_ROW
T_VARIABLE (row.variable)
----------------------------------------------------------------------------
identifier.identifier.identifier plpgsql_parse_tripword
T_LABEL
T_RECORD
T_VARIABLE (label.record.variable)
T_ROW
T_VARIABLE (label.row.variable)
----------------------------------------------------------------------------
identifier%TYPE plpgsql_parse_wordtype
T_VARIABLE
T_DTYPE (variable%TYPE)
T_DTYPE (typname%TYPE)
----------------------------------------------------------------------------
identifier.identifier%TYPE plpgsql_parse_dblwordtype
T_LABEL
T_VARIABLE
T_DTYPE (label.variable%TYPE)
T_DTYPE (relname.attname%TYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier.identifier%TYPE plpgsql_parse_tripwordtype
T_DTYPE (nspname.relname.attname%TYPE)
----------------------------------------------------------------------------
identifier%ROWTYPE plpgsql_parse_wordrowtype
T_DTYPE (relname%ROWTYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier%ROWTYPE plpgsql_parse_dblwordrowtype
T_DTYPE (nspname.relname%ROWTYPE)

============================================================================
Parameters - parallels the above
----------------------------------------------------------------------------
$# plpgsql_parse_word
$#.identifier plpgsql_parse_dblword
$#.identifier.identifier plpgsql_parse_tripword
$#%TYPE plpgsql_parse_wordtype
$#.identifier%TYPE plpgsql_parse_dblwordtype
$#.identifier.identifier%TYPE plpgsql_parse_tripwordtype
$#%ROWTYPE plpgsql_parse_wordrowtype
$#.identifier%ROWTYPE plpgsql_parse_dblwordrowtype

Comments?

Thanks,

Joe

Your patch has been added to the PostgreSQL unapplied patches list at:

http://candle.pha.pa.us/cgi-bin/pgpatches

I will try to apply it within the next 48 hours.

---------------------------------------------------------------------------

Joe Conway wrote:

Tom Lane wrote:

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

Attached is a diff -c format proposal to fix this. I've also attached a short
test script. Seems to work OK and passes all regression tests.

Here's a breakdown of how I understand plpgsql's "Special word rules" -- I
think it illustrates the behavior reasonably well. New functions added by this
patch are plpgsql_parse_tripwordtype and plpgsql_parse_dblwordrowtype:

============================================================================
Identifiers (represents) parsing function
----------------------------------------------------------------------------
identifier plpgsql_parse_word
tg_argv
T_LABEL (label)
T_VARIABLE (variable)
T_RECORD (record)
T_ROW (row)
----------------------------------------------------------------------------
identifier.identifier plpgsql_parse_dblword
T_LABEL
T_VARIABLE (label.variable)
T_RECORD (label.record)
T_ROW (label.row)
T_RECORD
T_VARIABLE (record.variable)
T_ROW
T_VARIABLE (row.variable)
----------------------------------------------------------------------------
identifier.identifier.identifier plpgsql_parse_tripword
T_LABEL
T_RECORD
T_VARIABLE (label.record.variable)
T_ROW
T_VARIABLE (label.row.variable)
----------------------------------------------------------------------------
identifier%TYPE plpgsql_parse_wordtype
T_VARIABLE
T_DTYPE (variable%TYPE)
T_DTYPE (typname%TYPE)
----------------------------------------------------------------------------
identifier.identifier%TYPE plpgsql_parse_dblwordtype
T_LABEL
T_VARIABLE
T_DTYPE (label.variable%TYPE)
T_DTYPE (relname.attname%TYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier.identifier%TYPE plpgsql_parse_tripwordtype
T_DTYPE (nspname.relname.attname%TYPE)
----------------------------------------------------------------------------
identifier%ROWTYPE plpgsql_parse_wordrowtype
T_DTYPE (relname%ROWTYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier%ROWTYPE plpgsql_parse_dblwordrowtype
T_DTYPE (nspname.relname%ROWTYPE)

============================================================================
Parameters - parallels the above
----------------------------------------------------------------------------
$# plpgsql_parse_word
$#.identifier plpgsql_parse_dblword
$#.identifier.identifier plpgsql_parse_tripword
$#%TYPE plpgsql_parse_wordtype
$#.identifier%TYPE plpgsql_parse_dblwordtype
$#.identifier.identifier%TYPE plpgsql_parse_tripwordtype
$#%ROWTYPE plpgsql_parse_wordrowtype
$#.identifier%ROWTYPE plpgsql_parse_dblwordrowtype

Comments?

Thanks,

Joe

Index: src/pl/plpgsql/src/pl_comp.c
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/pl_comp.c,v
retrieving revision 1.51
diff -c -r1.51 pl_comp.c
*** src/pl/plpgsql/src/pl_comp.c	4 Sep 2002 20:31:47 -0000	1.51
--- src/pl/plpgsql/src/pl_comp.c	9 Sep 2002 04:22:24 -0000
***************
*** 1092,1097 ****
--- 1092,1217 ----
return T_DTYPE;
}

+ /* ----------
+  * plpgsql_parse_tripwordtype		Same lookup for word.word.word%TYPE
+  * ----------
+  */
+ #define TYPE_JUNK_LEN	5
+ 
+ int
+ plpgsql_parse_tripwordtype(char *word)
+ {
+ 	Oid			classOid;
+ 	HeapTuple	classtup;
+ 	Form_pg_class classStruct;
+ 	HeapTuple	attrtup;
+ 	Form_pg_attribute attrStruct;
+ 	HeapTuple	typetup;
+ 	Form_pg_type typeStruct;
+ 	PLpgSQL_type *typ;
+ 	char	   *cp[2];
+ 	int			qualified_att_len;
+ 	int			numdots = 0;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	qualified_att_len = strlen(word) - TYPE_JUNK_LEN;
+ 	Assert(word[qualified_att_len] == '%');
+ 
+ 	for (i = 0; i < qualified_att_len; i++)
+ 	{
+ 		if (word[i] == '.' && ++numdots == 2)
+ 		{
+ 			cp[0] = (char *) palloc((i + 1) * sizeof(char));
+ 			memset(cp[0], 0, (i + 1) * sizeof(char));
+ 			memcpy(cp[0], word, i * sizeof(char));
+ 
+ 			/* qualified_att_len - one based position + 1 (null terminator) */
+ 			cp[1] = (char *) palloc((qualified_att_len - i) * sizeof(char));
+ 			memset(cp[1], 0, (qualified_att_len - i) * sizeof(char));
+ 			memcpy(cp[1], &word[i + 1], (qualified_att_len - i - 1) * sizeof(char));
+ 
+ 			break;
+ 		}
+ 	}
+ 
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp[0], "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	classtup = SearchSysCache(RELOID,
+ 							  ObjectIdGetDatum(classOid),
+ 							  0, 0, 0);
+ 	if (!HeapTupleIsValid(classtup))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * It must be a relation, sequence, view, or type
+ 	 */
+ 	classStruct = (Form_pg_class) GETSTRUCT(classtup);
+ 	if (classStruct->relkind != RELKIND_RELATION &&
+ 		classStruct->relkind != RELKIND_SEQUENCE &&
+ 		classStruct->relkind != RELKIND_VIEW &&
+ 		classStruct->relkind != RELKIND_COMPOSITE_TYPE)
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * Fetch the named table field and it's type
+ 	 */
+ 	attrtup = SearchSysCacheAttName(classOid, cp[1]);
+ 	if (!HeapTupleIsValid(attrtup))
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	attrStruct = (Form_pg_attribute) GETSTRUCT(attrtup);
+ 
+ 	typetup = SearchSysCache(TYPEOID,
+ 							 ObjectIdGetDatum(attrStruct->atttypid),
+ 							 0, 0, 0);
+ 	if (!HeapTupleIsValid(typetup))
+ 		elog(ERROR, "cache lookup for type %u of %s.%s failed",
+ 			 attrStruct->atttypid, cp[0], cp[1]);
+ 	typeStruct = (Form_pg_type) GETSTRUCT(typetup);
+ 
+ 	/*
+ 	 * Found that - build a compiler type struct and return it
+ 	 */
+ 	typ = (PLpgSQL_type *) malloc(sizeof(PLpgSQL_type));
+ 
+ 	typ->typname = strdup(NameStr(typeStruct->typname));
+ 	typ->typoid = attrStruct->atttypid;
+ 	perm_fmgr_info(typeStruct->typinput, &(typ->typinput));
+ 	typ->typelem = typeStruct->typelem;
+ 	typ->typbyval = typeStruct->typbyval;
+ 	typ->typlen = typeStruct->typlen;
+ 	typ->atttypmod = attrStruct->atttypmod;
+ 
+ 	plpgsql_yylval.dtype = typ;
+ 
+ 	ReleaseSysCache(classtup);
+ 	ReleaseSysCache(attrtup);
+ 	ReleaseSysCache(typetup);
+ 	pfree(cp[0]);
+ 	pfree(cp[1]);
+ 	return T_DTYPE;
+ }

/* ----------
* plpgsql_parse_wordrowtype		Scanner found word%ROWTYPE.
***************
*** 1125,1130 ****
--- 1245,1290 ----

pfree(cp[0]);
pfree(cp[1]);
+ 
+ 	return T_ROW;
+ }
+ 
+ /* ----------
+  * plpgsql_parse_dblwordrowtype		Scanner found word.word%ROWTYPE.
+  *			So word must be namespace qualified a table name.
+  * ----------
+  */
+ #define ROWTYPE_JUNK_LEN	8
+ 
+ int
+ plpgsql_parse_dblwordrowtype(char *word)
+ {
+ 	Oid			classOid;
+ 	char	   *cp;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	/* We convert %rowtype to .rowtype momentarily to keep converter happy */
+ 	i = strlen(word) - ROWTYPE_JUNK_LEN;
+ 	Assert(word[i] == '%');
+ 
+ 	cp = (char *) palloc((i + 1) * sizeof(char));
+ 	memset(cp, 0, (i + 1) * sizeof(char));
+ 	memcpy(cp, word, i * sizeof(char));
+ 
+ 	/* Lookup the relation */
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp, "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 		elog(ERROR, "%s: no such class", cp);
+ 
+ 	/*
+ 	 * Build and return the complete row definition
+ 	 */
+ 	plpgsql_yylval.row = build_rowtype(classOid);
+ 
+ 	pfree(cp);

return T_ROW;
}
Index: src/pl/plpgsql/src/plpgsql.h
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/plpgsql.h,v
retrieving revision 1.27
diff -c -r1.27 plpgsql.h
*** src/pl/plpgsql/src/plpgsql.h	4 Sep 2002 20:31:47 -0000	1.27
--- src/pl/plpgsql/src/plpgsql.h	9 Sep 2002 04:21:37 -0000
***************
*** 568,574 ****
--- 568,576 ----
extern int	plpgsql_parse_tripword(char *word);
extern int	plpgsql_parse_wordtype(char *word);
extern int	plpgsql_parse_dblwordtype(char *word);
+ extern int	plpgsql_parse_tripwordtype(char *word);
extern int	plpgsql_parse_wordrowtype(char *word);
+ extern int	plpgsql_parse_dblwordrowtype(char *word);
extern PLpgSQL_type *plpgsql_parse_datatype(char *string);
extern void plpgsql_adddatum(PLpgSQL_datum * new);
extern int	plpgsql_add_initdatums(int **varnos);
Index: src/pl/plpgsql/src/scan.l
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/scan.l,v
retrieving revision 1.22
diff -c -r1.22 scan.l
*** src/pl/plpgsql/src/scan.l	30 Aug 2002 00:28:41 -0000	1.22
--- src/pl/plpgsql/src/scan.l	9 Sep 2002 04:23:49 -0000
***************
*** 170,183 ****
--- 170,187 ----
{identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
{identifier}{space}*%TYPE		{ return plpgsql_parse_wordtype(yytext);	}
{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ {identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_wordrowtype(yytext);	}
+ {identifier}{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

\${digit}+						{ return plpgsql_parse_word(yytext);	}
\${digit}+{space}*\.{space}*{identifier}	{ return plpgsql_parse_dblword(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
\${digit}+{space}*%TYPE			{ return plpgsql_parse_wordtype(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ \${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
\${digit}+{space}*%ROWTYPE		{ return plpgsql_parse_wordrowtype(yytext);	}
+ \${digit}+{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

{digit}+ { return T_NUMBER; }

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS TEXT AS '
DECLARE
col_name pg_catalog.pg_attribute.attname%TYPE;
BEGIN
col_name := ''uga'';
RETURN col_name;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute.attname%TYPE AS '
DECLARE
rec pg_catalog.pg_attribute.attname%TYPE;
BEGIN
SELECT INTO rec pg_catalog.pg_attribute.attname FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

Patch applied. Thanks.

---------------------------------------------------------------------------

Joe Conway wrote:

Tom Lane wrote:

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

Attached is a diff -c format proposal to fix this. I've also attached a short
test script. Seems to work OK and passes all regression tests.

Here's a breakdown of how I understand plpgsql's "Special word rules" -- I
think it illustrates the behavior reasonably well. New functions added by this
patch are plpgsql_parse_tripwordtype and plpgsql_parse_dblwordrowtype:

============================================================================
Identifiers (represents) parsing function
----------------------------------------------------------------------------
identifier plpgsql_parse_word
tg_argv
T_LABEL (label)
T_VARIABLE (variable)
T_RECORD (record)
T_ROW (row)
----------------------------------------------------------------------------
identifier.identifier plpgsql_parse_dblword
T_LABEL
T_VARIABLE (label.variable)
T_RECORD (label.record)
T_ROW (label.row)
T_RECORD
T_VARIABLE (record.variable)
T_ROW
T_VARIABLE (row.variable)
----------------------------------------------------------------------------
identifier.identifier.identifier plpgsql_parse_tripword
T_LABEL
T_RECORD
T_VARIABLE (label.record.variable)
T_ROW
T_VARIABLE (label.row.variable)
----------------------------------------------------------------------------
identifier%TYPE plpgsql_parse_wordtype
T_VARIABLE
T_DTYPE (variable%TYPE)
T_DTYPE (typname%TYPE)
----------------------------------------------------------------------------
identifier.identifier%TYPE plpgsql_parse_dblwordtype
T_LABEL
T_VARIABLE
T_DTYPE (label.variable%TYPE)
T_DTYPE (relname.attname%TYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier.identifier%TYPE plpgsql_parse_tripwordtype
T_DTYPE (nspname.relname.attname%TYPE)
----------------------------------------------------------------------------
identifier%ROWTYPE plpgsql_parse_wordrowtype
T_DTYPE (relname%ROWTYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier%ROWTYPE plpgsql_parse_dblwordrowtype
T_DTYPE (nspname.relname%ROWTYPE)

============================================================================
Parameters - parallels the above
----------------------------------------------------------------------------
$# plpgsql_parse_word
$#.identifier plpgsql_parse_dblword
$#.identifier.identifier plpgsql_parse_tripword
$#%TYPE plpgsql_parse_wordtype
$#.identifier%TYPE plpgsql_parse_dblwordtype
$#.identifier.identifier%TYPE plpgsql_parse_tripwordtype
$#%ROWTYPE plpgsql_parse_wordrowtype
$#.identifier%ROWTYPE plpgsql_parse_dblwordrowtype

Comments?

Thanks,

Joe

Index: src/pl/plpgsql/src/pl_comp.c
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/pl_comp.c,v
retrieving revision 1.51
diff -c -r1.51 pl_comp.c
*** src/pl/plpgsql/src/pl_comp.c	4 Sep 2002 20:31:47 -0000	1.51
--- src/pl/plpgsql/src/pl_comp.c	9 Sep 2002 04:22:24 -0000
***************
*** 1092,1097 ****
--- 1092,1217 ----
return T_DTYPE;
}

+ /* ----------
+  * plpgsql_parse_tripwordtype		Same lookup for word.word.word%TYPE
+  * ----------
+  */
+ #define TYPE_JUNK_LEN	5
+ 
+ int
+ plpgsql_parse_tripwordtype(char *word)
+ {
+ 	Oid			classOid;
+ 	HeapTuple	classtup;
+ 	Form_pg_class classStruct;
+ 	HeapTuple	attrtup;
+ 	Form_pg_attribute attrStruct;
+ 	HeapTuple	typetup;
+ 	Form_pg_type typeStruct;
+ 	PLpgSQL_type *typ;
+ 	char	   *cp[2];
+ 	int			qualified_att_len;
+ 	int			numdots = 0;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	qualified_att_len = strlen(word) - TYPE_JUNK_LEN;
+ 	Assert(word[qualified_att_len] == '%');
+ 
+ 	for (i = 0; i < qualified_att_len; i++)
+ 	{
+ 		if (word[i] == '.' && ++numdots == 2)
+ 		{
+ 			cp[0] = (char *) palloc((i + 1) * sizeof(char));
+ 			memset(cp[0], 0, (i + 1) * sizeof(char));
+ 			memcpy(cp[0], word, i * sizeof(char));
+ 
+ 			/* qualified_att_len - one based position + 1 (null terminator) */
+ 			cp[1] = (char *) palloc((qualified_att_len - i) * sizeof(char));
+ 			memset(cp[1], 0, (qualified_att_len - i) * sizeof(char));
+ 			memcpy(cp[1], &word[i + 1], (qualified_att_len - i - 1) * sizeof(char));
+ 
+ 			break;
+ 		}
+ 	}
+ 
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp[0], "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	classtup = SearchSysCache(RELOID,
+ 							  ObjectIdGetDatum(classOid),
+ 							  0, 0, 0);
+ 	if (!HeapTupleIsValid(classtup))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * It must be a relation, sequence, view, or type
+ 	 */
+ 	classStruct = (Form_pg_class) GETSTRUCT(classtup);
+ 	if (classStruct->relkind != RELKIND_RELATION &&
+ 		classStruct->relkind != RELKIND_SEQUENCE &&
+ 		classStruct->relkind != RELKIND_VIEW &&
+ 		classStruct->relkind != RELKIND_COMPOSITE_TYPE)
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * Fetch the named table field and it's type
+ 	 */
+ 	attrtup = SearchSysCacheAttName(classOid, cp[1]);
+ 	if (!HeapTupleIsValid(attrtup))
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	attrStruct = (Form_pg_attribute) GETSTRUCT(attrtup);
+ 
+ 	typetup = SearchSysCache(TYPEOID,
+ 							 ObjectIdGetDatum(attrStruct->atttypid),
+ 							 0, 0, 0);
+ 	if (!HeapTupleIsValid(typetup))
+ 		elog(ERROR, "cache lookup for type %u of %s.%s failed",
+ 			 attrStruct->atttypid, cp[0], cp[1]);
+ 	typeStruct = (Form_pg_type) GETSTRUCT(typetup);
+ 
+ 	/*
+ 	 * Found that - build a compiler type struct and return it
+ 	 */
+ 	typ = (PLpgSQL_type *) malloc(sizeof(PLpgSQL_type));
+ 
+ 	typ->typname = strdup(NameStr(typeStruct->typname));
+ 	typ->typoid = attrStruct->atttypid;
+ 	perm_fmgr_info(typeStruct->typinput, &(typ->typinput));
+ 	typ->typelem = typeStruct->typelem;
+ 	typ->typbyval = typeStruct->typbyval;
+ 	typ->typlen = typeStruct->typlen;
+ 	typ->atttypmod = attrStruct->atttypmod;
+ 
+ 	plpgsql_yylval.dtype = typ;
+ 
+ 	ReleaseSysCache(classtup);
+ 	ReleaseSysCache(attrtup);
+ 	ReleaseSysCache(typetup);
+ 	pfree(cp[0]);
+ 	pfree(cp[1]);
+ 	return T_DTYPE;
+ }

/* ----------
* plpgsql_parse_wordrowtype		Scanner found word%ROWTYPE.
***************
*** 1125,1130 ****
--- 1245,1290 ----

pfree(cp[0]);
pfree(cp[1]);
+ 
+ 	return T_ROW;
+ }
+ 
+ /* ----------
+  * plpgsql_parse_dblwordrowtype		Scanner found word.word%ROWTYPE.
+  *			So word must be namespace qualified a table name.
+  * ----------
+  */
+ #define ROWTYPE_JUNK_LEN	8
+ 
+ int
+ plpgsql_parse_dblwordrowtype(char *word)
+ {
+ 	Oid			classOid;
+ 	char	   *cp;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	/* We convert %rowtype to .rowtype momentarily to keep converter happy */
+ 	i = strlen(word) - ROWTYPE_JUNK_LEN;
+ 	Assert(word[i] == '%');
+ 
+ 	cp = (char *) palloc((i + 1) * sizeof(char));
+ 	memset(cp, 0, (i + 1) * sizeof(char));
+ 	memcpy(cp, word, i * sizeof(char));
+ 
+ 	/* Lookup the relation */
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp, "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 		elog(ERROR, "%s: no such class", cp);
+ 
+ 	/*
+ 	 * Build and return the complete row definition
+ 	 */
+ 	plpgsql_yylval.row = build_rowtype(classOid);
+ 
+ 	pfree(cp);

return T_ROW;
}
Index: src/pl/plpgsql/src/plpgsql.h
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/plpgsql.h,v
retrieving revision 1.27
diff -c -r1.27 plpgsql.h
*** src/pl/plpgsql/src/plpgsql.h	4 Sep 2002 20:31:47 -0000	1.27
--- src/pl/plpgsql/src/plpgsql.h	9 Sep 2002 04:21:37 -0000
***************
*** 568,574 ****
--- 568,576 ----
extern int	plpgsql_parse_tripword(char *word);
extern int	plpgsql_parse_wordtype(char *word);
extern int	plpgsql_parse_dblwordtype(char *word);
+ extern int	plpgsql_parse_tripwordtype(char *word);
extern int	plpgsql_parse_wordrowtype(char *word);
+ extern int	plpgsql_parse_dblwordrowtype(char *word);
extern PLpgSQL_type *plpgsql_parse_datatype(char *string);
extern void plpgsql_adddatum(PLpgSQL_datum * new);
extern int	plpgsql_add_initdatums(int **varnos);
Index: src/pl/plpgsql/src/scan.l
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/scan.l,v
retrieving revision 1.22
diff -c -r1.22 scan.l
*** src/pl/plpgsql/src/scan.l	30 Aug 2002 00:28:41 -0000	1.22
--- src/pl/plpgsql/src/scan.l	9 Sep 2002 04:23:49 -0000
***************
*** 170,183 ****
--- 170,187 ----
{identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
{identifier}{space}*%TYPE		{ return plpgsql_parse_wordtype(yytext);	}
{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ {identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_wordrowtype(yytext);	}
+ {identifier}{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

\${digit}+						{ return plpgsql_parse_word(yytext);	}
\${digit}+{space}*\.{space}*{identifier}	{ return plpgsql_parse_dblword(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
\${digit}+{space}*%TYPE			{ return plpgsql_parse_wordtype(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ \${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
\${digit}+{space}*%ROWTYPE		{ return plpgsql_parse_wordrowtype(yytext);	}
+ \${digit}+{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

{digit}+ { return T_NUMBER; }

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS TEXT AS '
DECLARE
col_name pg_catalog.pg_attribute.attname%TYPE;
BEGIN
col_name := ''uga'';
RETURN col_name;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute.attname%TYPE AS '
DECLARE
rec pg_catalog.pg_attribute.attname%TYPE;
BEGIN
SELECT INTO rec pg_catalog.pg_attribute.attname FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

Does anyone know if such effort is also required to pl/python to become
"schema aware"?

Regards,

Greg Copeland

Does pl/python even have a DECLARE section that can mimick the data type
of an existing table column?

---------------------------------------------------------------------------

Greg Copeland wrote:
-- Start of PGP signed section.

Does anyone know if such effort is also required to pl/python to become
"schema aware"?

Regards,

Greg Copeland

On Wed, 2002-09-11 at 19:24, Bruce Momjian wrote:

Patch applied. Thanks.

---------------------------------------------------------------------------

Joe Conway wrote:

Tom Lane wrote:

Sean Chittenden <sean@chittenden.org> writes:

::sigh:: Is it me or does it look like all
of pl/pgsql is schema un-aware (ie, all of the declarations). -sc

Yeah. The group of routines parse_word, parse_dblword, etc that are
called by the lexer certainly all need work. There are some
definitional issues to think about, too --- plpgsql presently relies on
the number of names to give it some idea of what to look for, and those
rules are probably all toast now. Please come up with a sketch of what
you think the behavior should be before you start hacking code.

Attached is a diff -c format proposal to fix this. I've also attached a short
test script. Seems to work OK and passes all regression tests.

Here's a breakdown of how I understand plpgsql's "Special word rules" -- I
think it illustrates the behavior reasonably well. New functions added by this
patch are plpgsql_parse_tripwordtype and plpgsql_parse_dblwordrowtype:

============================================================================
Identifiers (represents) parsing function
----------------------------------------------------------------------------
identifier plpgsql_parse_word
tg_argv
T_LABEL (label)
T_VARIABLE (variable)
T_RECORD (record)
T_ROW (row)
----------------------------------------------------------------------------
identifier.identifier plpgsql_parse_dblword
T_LABEL
T_VARIABLE (label.variable)
T_RECORD (label.record)
T_ROW (label.row)
T_RECORD
T_VARIABLE (record.variable)
T_ROW
T_VARIABLE (row.variable)
----------------------------------------------------------------------------
identifier.identifier.identifier plpgsql_parse_tripword
T_LABEL
T_RECORD
T_VARIABLE (label.record.variable)
T_ROW
T_VARIABLE (label.row.variable)
----------------------------------------------------------------------------
identifier%TYPE plpgsql_parse_wordtype
T_VARIABLE
T_DTYPE (variable%TYPE)
T_DTYPE (typname%TYPE)
----------------------------------------------------------------------------
identifier.identifier%TYPE plpgsql_parse_dblwordtype
T_LABEL
T_VARIABLE
T_DTYPE (label.variable%TYPE)
T_DTYPE (relname.attname%TYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier.identifier%TYPE plpgsql_parse_tripwordtype
T_DTYPE (nspname.relname.attname%TYPE)
----------------------------------------------------------------------------
identifier%ROWTYPE plpgsql_parse_wordrowtype
T_DTYPE (relname%ROWTYPE)
----------------------------------------------------------------------------
<new>
identifier.identifier%ROWTYPE plpgsql_parse_dblwordrowtype
T_DTYPE (nspname.relname%ROWTYPE)

============================================================================
Parameters - parallels the above
----------------------------------------------------------------------------
$# plpgsql_parse_word
$#.identifier plpgsql_parse_dblword
$#.identifier.identifier plpgsql_parse_tripword
$#%TYPE plpgsql_parse_wordtype
$#.identifier%TYPE plpgsql_parse_dblwordtype
$#.identifier.identifier%TYPE plpgsql_parse_tripwordtype
$#%ROWTYPE plpgsql_parse_wordrowtype
$#.identifier%ROWTYPE plpgsql_parse_dblwordrowtype

Comments?

Thanks,

Joe

Index: src/pl/plpgsql/src/pl_comp.c
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/pl_comp.c,v
retrieving revision 1.51
diff -c -r1.51 pl_comp.c
*** src/pl/plpgsql/src/pl_comp.c	4 Sep 2002 20:31:47 -0000	1.51
--- src/pl/plpgsql/src/pl_comp.c	9 Sep 2002 04:22:24 -0000
***************
*** 1092,1097 ****
--- 1092,1217 ----
return T_DTYPE;
}

+ /* ----------
+  * plpgsql_parse_tripwordtype		Same lookup for word.word.word%TYPE
+  * ----------
+  */
+ #define TYPE_JUNK_LEN	5
+ 
+ int
+ plpgsql_parse_tripwordtype(char *word)
+ {
+ 	Oid			classOid;
+ 	HeapTuple	classtup;
+ 	Form_pg_class classStruct;
+ 	HeapTuple	attrtup;
+ 	Form_pg_attribute attrStruct;
+ 	HeapTuple	typetup;
+ 	Form_pg_type typeStruct;
+ 	PLpgSQL_type *typ;
+ 	char	   *cp[2];
+ 	int			qualified_att_len;
+ 	int			numdots = 0;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	qualified_att_len = strlen(word) - TYPE_JUNK_LEN;
+ 	Assert(word[qualified_att_len] == '%');
+ 
+ 	for (i = 0; i < qualified_att_len; i++)
+ 	{
+ 		if (word[i] == '.' && ++numdots == 2)
+ 		{
+ 			cp[0] = (char *) palloc((i + 1) * sizeof(char));
+ 			memset(cp[0], 0, (i + 1) * sizeof(char));
+ 			memcpy(cp[0], word, i * sizeof(char));
+ 
+ 			/* qualified_att_len - one based position + 1 (null terminator) */
+ 			cp[1] = (char *) palloc((qualified_att_len - i) * sizeof(char));
+ 			memset(cp[1], 0, (qualified_att_len - i) * sizeof(char));
+ 			memcpy(cp[1], &word[i + 1], (qualified_att_len - i - 1) * sizeof(char));
+ 
+ 			break;
+ 		}
+ 	}
+ 
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp[0], "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	classtup = SearchSysCache(RELOID,
+ 							  ObjectIdGetDatum(classOid),
+ 							  0, 0, 0);
+ 	if (!HeapTupleIsValid(classtup))
+ 	{
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * It must be a relation, sequence, view, or type
+ 	 */
+ 	classStruct = (Form_pg_class) GETSTRUCT(classtup);
+ 	if (classStruct->relkind != RELKIND_RELATION &&
+ 		classStruct->relkind != RELKIND_SEQUENCE &&
+ 		classStruct->relkind != RELKIND_VIEW &&
+ 		classStruct->relkind != RELKIND_COMPOSITE_TYPE)
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 
+ 	/*
+ 	 * Fetch the named table field and it's type
+ 	 */
+ 	attrtup = SearchSysCacheAttName(classOid, cp[1]);
+ 	if (!HeapTupleIsValid(attrtup))
+ 	{
+ 		ReleaseSysCache(classtup);
+ 		pfree(cp[0]);
+ 		pfree(cp[1]);
+ 		return T_ERROR;
+ 	}
+ 	attrStruct = (Form_pg_attribute) GETSTRUCT(attrtup);
+ 
+ 	typetup = SearchSysCache(TYPEOID,
+ 							 ObjectIdGetDatum(attrStruct->atttypid),
+ 							 0, 0, 0);
+ 	if (!HeapTupleIsValid(typetup))
+ 		elog(ERROR, "cache lookup for type %u of %s.%s failed",
+ 			 attrStruct->atttypid, cp[0], cp[1]);
+ 	typeStruct = (Form_pg_type) GETSTRUCT(typetup);
+ 
+ 	/*
+ 	 * Found that - build a compiler type struct and return it
+ 	 */
+ 	typ = (PLpgSQL_type *) malloc(sizeof(PLpgSQL_type));
+ 
+ 	typ->typname = strdup(NameStr(typeStruct->typname));
+ 	typ->typoid = attrStruct->atttypid;
+ 	perm_fmgr_info(typeStruct->typinput, &(typ->typinput));
+ 	typ->typelem = typeStruct->typelem;
+ 	typ->typbyval = typeStruct->typbyval;
+ 	typ->typlen = typeStruct->typlen;
+ 	typ->atttypmod = attrStruct->atttypmod;
+ 
+ 	plpgsql_yylval.dtype = typ;
+ 
+ 	ReleaseSysCache(classtup);
+ 	ReleaseSysCache(attrtup);
+ 	ReleaseSysCache(typetup);
+ 	pfree(cp[0]);
+ 	pfree(cp[1]);
+ 	return T_DTYPE;
+ }

/* ----------
* plpgsql_parse_wordrowtype		Scanner found word%ROWTYPE.
***************
*** 1125,1130 ****
--- 1245,1290 ----

pfree(cp[0]);
pfree(cp[1]);
+ 
+ 	return T_ROW;
+ }
+ 
+ /* ----------
+  * plpgsql_parse_dblwordrowtype		Scanner found word.word%ROWTYPE.
+  *			So word must be namespace qualified a table name.
+  * ----------
+  */
+ #define ROWTYPE_JUNK_LEN	8
+ 
+ int
+ plpgsql_parse_dblwordrowtype(char *word)
+ {
+ 	Oid			classOid;
+ 	char	   *cp;
+ 	int			i;
+ 	RangeVar   *relvar;
+ 
+ 	/* Do case conversion and word separation */
+ 	/* We convert %rowtype to .rowtype momentarily to keep converter happy */
+ 	i = strlen(word) - ROWTYPE_JUNK_LEN;
+ 	Assert(word[i] == '%');
+ 
+ 	cp = (char *) palloc((i + 1) * sizeof(char));
+ 	memset(cp, 0, (i + 1) * sizeof(char));
+ 	memcpy(cp, word, i * sizeof(char));
+ 
+ 	/* Lookup the relation */
+ 	relvar = makeRangeVarFromNameList(stringToQualifiedNameList(cp, "plpgsql_parse_dblwordtype"));
+ 	classOid = RangeVarGetRelid(relvar, true);
+ 	if (!OidIsValid(classOid))
+ 		elog(ERROR, "%s: no such class", cp);
+ 
+ 	/*
+ 	 * Build and return the complete row definition
+ 	 */
+ 	plpgsql_yylval.row = build_rowtype(classOid);
+ 
+ 	pfree(cp);

return T_ROW;
}
Index: src/pl/plpgsql/src/plpgsql.h
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/plpgsql.h,v
retrieving revision 1.27
diff -c -r1.27 plpgsql.h
*** src/pl/plpgsql/src/plpgsql.h	4 Sep 2002 20:31:47 -0000	1.27
--- src/pl/plpgsql/src/plpgsql.h	9 Sep 2002 04:21:37 -0000
***************
*** 568,574 ****
--- 568,576 ----
extern int	plpgsql_parse_tripword(char *word);
extern int	plpgsql_parse_wordtype(char *word);
extern int	plpgsql_parse_dblwordtype(char *word);
+ extern int	plpgsql_parse_tripwordtype(char *word);
extern int	plpgsql_parse_wordrowtype(char *word);
+ extern int	plpgsql_parse_dblwordrowtype(char *word);
extern PLpgSQL_type *plpgsql_parse_datatype(char *string);
extern void plpgsql_adddatum(PLpgSQL_datum * new);
extern int	plpgsql_add_initdatums(int **varnos);
Index: src/pl/plpgsql/src/scan.l
===================================================================
RCS file: /opt/src/cvs/pgsql-server/src/pl/plpgsql/src/scan.l,v
retrieving revision 1.22
diff -c -r1.22 scan.l
*** src/pl/plpgsql/src/scan.l	30 Aug 2002 00:28:41 -0000	1.22
--- src/pl/plpgsql/src/scan.l	9 Sep 2002 04:23:49 -0000
***************
*** 170,183 ****
--- 170,187 ----
{identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
{identifier}{space}*%TYPE		{ return plpgsql_parse_wordtype(yytext);	}
{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ {identifier}{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_wordrowtype(yytext);	}
+ {identifier}{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

\${digit}+						{ return plpgsql_parse_word(yytext);	}
\${digit}+{space}*\.{space}*{identifier}	{ return plpgsql_parse_dblword(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}	{ return plpgsql_parse_tripword(yytext); }
\${digit}+{space}*%TYPE			{ return plpgsql_parse_wordtype(yytext);	}
\${digit}+{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_dblwordtype(yytext); }
+ \${digit}+{space}*\.{space}*{identifier}{space}*\.{space}*{identifier}{space}*%TYPE	{ return plpgsql_parse_tripwordtype(yytext); }
\${digit}+{space}*%ROWTYPE		{ return plpgsql_parse_wordrowtype(yytext);	}
+ \${digit}+{space}*\.{space}*{identifier}{space}*%ROWTYPE	{ return plpgsql_parse_dblwordrowtype(yytext);	}

{digit}+ { return T_NUMBER; }

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS TEXT AS '
DECLARE
col_name pg_catalog.pg_attribute.attname%TYPE;
BEGIN
col_name := ''uga'';
RETURN col_name;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

-- nspname.relname.attname%TYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute.attname%TYPE AS '
DECLARE
rec pg_catalog.pg_attribute.attname%TYPE;
BEGIN
SELECT INTO rec pg_catalog.pg_attribute.attname FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT t();

-- nspname.relname%ROWTYPE
DROP FUNCTION t();
CREATE OR REPLACE FUNCTION t() RETURNS pg_catalog.pg_attribute AS '
DECLARE
rec pg_catalog.pg_attribute%ROWTYPE;
BEGIN
SELECT INTO rec * FROM pg_catalog.pg_attribute WHERE attrelid = 1247 AND attname = ''typname'';
RETURN rec;
END;
' LANGUAGE 'plpgsql';
SELECT * FROM t();

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly

-- 
Bruce Momjian                        |  http://candle.pha.pa.us
pgman@candle.pha.pa.us               |  (610) 359-1001
+  If your life is a hard drive,     |  13 Roberts Road
+  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

-- End of PGP section, PGP failed!

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073