Solve the 'Turkish problem' with undesirable locale behavior for case

conversion of basic ASCII letters.  Remove all uses of strcasecmp and
strncasecmp in favor of new functions pg_strcasecmp and pg_strncasecmp;
remove most but not all direct uses of toupper and tolower in favor of
pg_toupper and pg_tolower.  These functions use the same notions of
case folding already developed for identifier case conversion.  I left
the straight locale-based folding in place for situations where we are
just manipulating user data and not trying to match it to built-in
strings --- for example, the SQL upper() function is still locale
dependent.  Perhaps this will prove not to be what's wanted, but at
the moment we can initdb and pass regression tests in Turkish locale.

7 files changed, 80 insertions, 82 deletions

diff --git a/src/backend/commands/aggregatecmds.c b/src/backend/commands/aggregatecmds.c
index 3dfacf7d18..d587245ef9 100644
--- a/src/backend/commands/aggregatecmds.c
+++ b/src/backend/commands/aggregatecmds.c
@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/aggregatecmds.c,v 1.16 2003/11/29 19:51:47 pgsql Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/aggregatecmds.c,v 1.17 2004/05/07 00:24:57 tgl Exp $
  *
  * DESCRIPTION
  *	  The "DefineFoo" routines take the parse tree and pick out the
@@ -75,21 +75,21 @@ DefineAggregate(List *names, List *parameters)
 		 * sfunc1, stype1, and initcond1 are accepted as obsolete
 		 * spellings for sfunc, stype, initcond.
 		 */
-		if (strcasecmp(defel->defname, "sfunc") == 0)
+		if (pg_strcasecmp(defel->defname, "sfunc") == 0)
 			transfuncName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "sfunc1") == 0)
+		else if (pg_strcasecmp(defel->defname, "sfunc1") == 0)
 			transfuncName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "finalfunc") == 0)
+		else if (pg_strcasecmp(defel->defname, "finalfunc") == 0)
 			finalfuncName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "basetype") == 0)
+		else if (pg_strcasecmp(defel->defname, "basetype") == 0)
 			baseType = defGetTypeName(defel);
-		else if (strcasecmp(defel->defname, "stype") == 0)
+		else if (pg_strcasecmp(defel->defname, "stype") == 0)
 			transType = defGetTypeName(defel);
-		else if (strcasecmp(defel->defname, "stype1") == 0)
+		else if (pg_strcasecmp(defel->defname, "stype1") == 0)
 			transType = defGetTypeName(defel);
-		else if (strcasecmp(defel->defname, "initcond") == 0)
+		else if (pg_strcasecmp(defel->defname, "initcond") == 0)
 			initval = defGetString(defel);
-		else if (strcasecmp(defel->defname, "initcond1") == 0)
+		else if (pg_strcasecmp(defel->defname, "initcond1") == 0)
 			initval = defGetString(defel);
 		else
 			ereport(WARNING,
@@ -124,7 +124,7 @@ DefineAggregate(List *names, List *parameters)
 	 * be able to store values of the transtype.  However, we can allow
 	 * polymorphic transtype in some cases (AggregateCreate will check).
 	 */
-	if (strcasecmp(TypeNameToString(baseType), "ANY") == 0)
+	if (pg_strcasecmp(TypeNameToString(baseType), "ANY") == 0)
 		baseTypeId = ANYOID;
 	else
 		baseTypeId = typenameTypeId(baseType);
diff --git a/src/backend/commands/define.c b/src/backend/commands/define.c
index fc24c2c30f..19f14879a0 100644
--- a/src/backend/commands/define.c
+++ b/src/backend/commands/define.c
@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/define.c,v 1.86 2004/02/21 00:34:52 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/define.c,v 1.87 2004/05/07 00:24:57 tgl Exp $
  *
  * DESCRIPTION
  *	  The "DefineFoo" routines take the parse tree and pick out the
@@ -240,12 +240,12 @@ defGetTypeLength(DefElem *def)
 							def->defname)));
 			break;
 		case T_String:
-			if (strcasecmp(strVal(def->arg), "variable") == 0)
+			if (pg_strcasecmp(strVal(def->arg), "variable") == 0)
 				return -1;		/* variable length */
 			break;
 		case T_TypeName:
 			/* cope if grammar chooses to believe "variable" is a typename */
-			if (strcasecmp(TypeNameToString((TypeName *) def->arg),
+			if (pg_strcasecmp(TypeNameToString((TypeName *) def->arg),
 						   "variable") == 0)
 				return -1;		/* variable length */
 			break;
diff --git a/src/backend/commands/functioncmds.c b/src/backend/commands/functioncmds.c
index c91b31ed6f..a0a9c58240 100644
--- a/src/backend/commands/functioncmds.c
+++ b/src/backend/commands/functioncmds.c
@@ -10,7 +10,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/functioncmds.c,v 1.44 2004/02/21 00:34:52 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/functioncmds.c,v 1.45 2004/05/07 00:24:57 tgl Exp $
  *
  * DESCRIPTION
  *	  These routines take the parse tree and pick out the
@@ -328,9 +328,9 @@ compute_attributes_with_style(List *parameters, bool *isStrict_p, char *volatili
 	{
 		DefElem    *param = (DefElem *) lfirst(pl);
 
-		if (strcasecmp(param->defname, "isstrict") == 0)
+		if (pg_strcasecmp(param->defname, "isstrict") == 0)
 			*isStrict_p = true;
-		else if (strcasecmp(param->defname, "iscachable") == 0)
+		else if (pg_strcasecmp(param->defname, "iscachable") == 0)
 		{
 			/* obsolete spelling of isImmutable */
 			*volatility_p = PROVOLATILE_IMMUTABLE;
diff --git a/src/backend/commands/operatorcmds.c b/src/backend/commands/operatorcmds.c
index 657e289fbb..2736a31f3c 100644
--- a/src/backend/commands/operatorcmds.c
+++ b/src/backend/commands/operatorcmds.c
@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/operatorcmds.c,v 1.13 2003/11/29 19:51:47 pgsql Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/operatorcmds.c,v 1.14 2004/05/07 00:24:57 tgl Exp $
  *
  * DESCRIPTION
  *	  The "DefineFoo" routines take the parse tree and pick out the
@@ -97,7 +97,7 @@ DefineOperator(List *names, List *parameters)
 	{
 		DefElem    *defel = (DefElem *) lfirst(pl);
 
-		if (strcasecmp(defel->defname, "leftarg") == 0)
+		if (pg_strcasecmp(defel->defname, "leftarg") == 0)
 		{
 			typeName1 = defGetTypeName(defel);
 			if (typeName1->setof)
@@ -105,7 +105,7 @@ DefineOperator(List *names, List *parameters)
 						(errcode(ERRCODE_INVALID_FUNCTION_DEFINITION),
 				errmsg("setof type not allowed for operator argument")));
 		}
-		else if (strcasecmp(defel->defname, "rightarg") == 0)
+		else if (pg_strcasecmp(defel->defname, "rightarg") == 0)
 		{
 			typeName2 = defGetTypeName(defel);
 			if (typeName2->setof)
@@ -113,27 +113,27 @@ DefineOperator(List *names, List *parameters)
 						(errcode(ERRCODE_INVALID_FUNCTION_DEFINITION),
 				errmsg("setof type not allowed for operator argument")));
 		}
-		else if (strcasecmp(defel->defname, "procedure") == 0)
+		else if (pg_strcasecmp(defel->defname, "procedure") == 0)
 			functionName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "commutator") == 0)
+		else if (pg_strcasecmp(defel->defname, "commutator") == 0)
 			commutatorName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "negator") == 0)
+		else if (pg_strcasecmp(defel->defname, "negator") == 0)
 			negatorName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "restrict") == 0)
+		else if (pg_strcasecmp(defel->defname, "restrict") == 0)
 			restrictionName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "join") == 0)
+		else if (pg_strcasecmp(defel->defname, "join") == 0)
 			joinName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "hashes") == 0)
+		else if (pg_strcasecmp(defel->defname, "hashes") == 0)
 			canHash = TRUE;
-		else if (strcasecmp(defel->defname, "merges") == 0)
+		else if (pg_strcasecmp(defel->defname, "merges") == 0)
 			canMerge = TRUE;
-		else if (strcasecmp(defel->defname, "sort1") == 0)
+		else if (pg_strcasecmp(defel->defname, "sort1") == 0)
 			leftSortName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "sort2") == 0)
+		else if (pg_strcasecmp(defel->defname, "sort2") == 0)
 			rightSortName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "ltcmp") == 0)
+		else if (pg_strcasecmp(defel->defname, "ltcmp") == 0)
 			ltCompareName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "gtcmp") == 0)
+		else if (pg_strcasecmp(defel->defname, "gtcmp") == 0)
 			gtCompareName = defGetQualifiedName(defel);
 		else
 			ereport(WARNING,
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index f19645a826..d74e1e7ab3 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/tablecmds.c,v 1.104 2004/05/06 16:10:57 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/tablecmds.c,v 1.105 2004/05/07 00:24:57 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -3126,13 +3126,13 @@ ATExecSetStorage(Relation rel, const char *colName, Node *newValue)
 	Assert(IsA(newValue, String));
 	storagemode = strVal(newValue);
 
-	if (strcasecmp(storagemode, "plain") == 0)
+	if (pg_strcasecmp(storagemode, "plain") == 0)
 		newstorage = 'p';
-	else if (strcasecmp(storagemode, "external") == 0)
+	else if (pg_strcasecmp(storagemode, "external") == 0)
 		newstorage = 'e';
-	else if (strcasecmp(storagemode, "extended") == 0)
+	else if (pg_strcasecmp(storagemode, "extended") == 0)
 		newstorage = 'x';
-	else if (strcasecmp(storagemode, "main") == 0)
+	else if (pg_strcasecmp(storagemode, "main") == 0)
 		newstorage = 'm';
 	else
 	{
diff --git a/src/backend/commands/typecmds.c b/src/backend/commands/typecmds.c
index d76e048a3e..411ad725bc 100644
--- a/src/backend/commands/typecmds.c
+++ b/src/backend/commands/typecmds.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/typecmds.c,v 1.54 2004/05/05 17:06:56 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/typecmds.c,v 1.55 2004/05/07 00:24:57 tgl Exp $
  *
  * DESCRIPTION
  *	  The "DefineFoo" routines take the parse tree and pick out the
@@ -142,28 +142,28 @@ DefineType(List *names, List *parameters)
 	{
 		DefElem    *defel = (DefElem *) lfirst(pl);
 
-		if (strcasecmp(defel->defname, "internallength") == 0)
+		if (pg_strcasecmp(defel->defname, "internallength") == 0)
 			internalLength = defGetTypeLength(defel);
-		else if (strcasecmp(defel->defname, "externallength") == 0)
+		else if (pg_strcasecmp(defel->defname, "externallength") == 0)
 			;					/* ignored -- remove after 7.3 */
-		else if (strcasecmp(defel->defname, "input") == 0)
+		else if (pg_strcasecmp(defel->defname, "input") == 0)
 			inputName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "output") == 0)
+		else if (pg_strcasecmp(defel->defname, "output") == 0)
 			outputName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "receive") == 0)
+		else if (pg_strcasecmp(defel->defname, "receive") == 0)
 			receiveName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "send") == 0)
+		else if (pg_strcasecmp(defel->defname, "send") == 0)
 			sendName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "analyze") == 0 ||
-				 strcasecmp(defel->defname, "analyse") == 0)
+		else if (pg_strcasecmp(defel->defname, "analyze") == 0 ||
+				 pg_strcasecmp(defel->defname, "analyse") == 0)
 			analyzeName = defGetQualifiedName(defel);
-		else if (strcasecmp(defel->defname, "delimiter") == 0)
+		else if (pg_strcasecmp(defel->defname, "delimiter") == 0)
 		{
 			char	   *p = defGetString(defel);
 
 			delimiter = p[0];
 		}
-		else if (strcasecmp(defel->defname, "element") == 0)
+		else if (pg_strcasecmp(defel->defname, "element") == 0)
 		{
 			elemType = typenameTypeId(defGetTypeName(defel));
 			/* disallow arrays of pseudotypes */
@@ -173,11 +173,11 @@ DefineType(List *names, List *parameters)
 						 errmsg("array element type cannot be %s",
 								format_type_be(elemType))));
 		}
-		else if (strcasecmp(defel->defname, "default") == 0)
+		else if (pg_strcasecmp(defel->defname, "default") == 0)
 			defaultValue = defGetString(defel);
-		else if (strcasecmp(defel->defname, "passedbyvalue") == 0)
+		else if (pg_strcasecmp(defel->defname, "passedbyvalue") == 0)
 			byValue = true;
-		else if (strcasecmp(defel->defname, "alignment") == 0)
+		else if (pg_strcasecmp(defel->defname, "alignment") == 0)
 		{
 			char	   *a = defGetString(defel);
 
@@ -187,35 +187,35 @@ DefineType(List *names, List *parameters)
 			 * recognize translated type names as well as the nominal
 			 * form.
 			 */
-			if (strcasecmp(a, "double") == 0 ||
-				strcasecmp(a, "float8") == 0 ||
-				strcasecmp(a, "pg_catalog.float8") == 0)
+			if (pg_strcasecmp(a, "double") == 0 ||
+				pg_strcasecmp(a, "float8") == 0 ||
+				pg_strcasecmp(a, "pg_catalog.float8") == 0)
 				alignment = 'd';
-			else if (strcasecmp(a, "int4") == 0 ||
-					 strcasecmp(a, "pg_catalog.int4") == 0)
+			else if (pg_strcasecmp(a, "int4") == 0 ||
+					 pg_strcasecmp(a, "pg_catalog.int4") == 0)
 				alignment = 'i';
-			else if (strcasecmp(a, "int2") == 0 ||
-					 strcasecmp(a, "pg_catalog.int2") == 0)
+			else if (pg_strcasecmp(a, "int2") == 0 ||
+					 pg_strcasecmp(a, "pg_catalog.int2") == 0)
 				alignment = 's';
-			else if (strcasecmp(a, "char") == 0 ||
-					 strcasecmp(a, "pg_catalog.bpchar") == 0)
+			else if (pg_strcasecmp(a, "char") == 0 ||
+					 pg_strcasecmp(a, "pg_catalog.bpchar") == 0)
 				alignment = 'c';
 			else
 				ereport(ERROR,
 						(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
 						 errmsg("alignment \"%s\" not recognized", a)));
 		}
-		else if (strcasecmp(defel->defname, "storage") == 0)
+		else if (pg_strcasecmp(defel->defname, "storage") == 0)
 		{
 			char	   *a = defGetString(defel);
 
-			if (strcasecmp(a, "plain") == 0)
+			if (pg_strcasecmp(a, "plain") == 0)
 				storage = 'p';
-			else if (strcasecmp(a, "external") == 0)
+			else if (pg_strcasecmp(a, "external") == 0)
 				storage = 'e';
-			else if (strcasecmp(a, "extended") == 0)
+			else if (pg_strcasecmp(a, "extended") == 0)
 				storage = 'x';
-			else if (strcasecmp(a, "main") == 0)
+			else if (pg_strcasecmp(a, "main") == 0)
 				storage = 'm';
 			else
 				ereport(ERROR,
diff --git a/src/backend/commands/variable.c b/src/backend/commands/variable.c
index 14c408e001..f83ce51bac 100644
--- a/src/backend/commands/variable.c
+++ b/src/backend/commands/variable.c
@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/commands/variable.c,v 1.93 2004/01/19 19:04:40 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/commands/variable.c,v 1.94 2004/05/07 00:24:57 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -23,6 +23,7 @@
 #include "catalog/pg_shadow.h"
 #include "commands/variable.h"
 #include "miscadmin.h"
+#include "parser/scansup.h"
 #include "utils/builtins.h"
 #include "utils/guc.h"
 #include "utils/syscache.h"
@@ -82,22 +83,22 @@ assign_datestyle(const char *value, bool doit, GucSource source)
 
 		/* Ugh. Somebody ought to write a table driven version -- mjl */
 
-		if (strcasecmp(tok, "ISO") == 0)
+		if (pg_strcasecmp(tok, "ISO") == 0)
 		{
 			newDateStyle = USE_ISO_DATES;
 			scnt++;
 		}
-		else if (strcasecmp(tok, "SQL") == 0)
+		else if (pg_strcasecmp(tok, "SQL") == 0)
 		{
 			newDateStyle = USE_SQL_DATES;
 			scnt++;
 		}
-		else if (strncasecmp(tok, "POSTGRES", 8) == 0)
+		else if (pg_strncasecmp(tok, "POSTGRES", 8) == 0)
 		{
 			newDateStyle = USE_POSTGRES_DATES;
 			scnt++;
 		}
-		else if (strcasecmp(tok, "GERMAN") == 0)
+		else if (pg_strcasecmp(tok, "GERMAN") == 0)
 		{
 			newDateStyle = USE_GERMAN_DATES;
 			scnt++;
@@ -105,25 +106,25 @@ assign_datestyle(const char *value, bool doit, GucSource source)
 			if (ocnt == 0)
 				newDateOrder = DATEORDER_DMY;
 		}
-		else if (strcasecmp(tok, "YMD") == 0)
+		else if (pg_strcasecmp(tok, "YMD") == 0)
 		{
 			newDateOrder = DATEORDER_YMD;
 			ocnt++;
 		}
-		else if (strcasecmp(tok, "DMY") == 0 ||
-				 strncasecmp(tok, "EURO", 4) == 0)
+		else if (pg_strcasecmp(tok, "DMY") == 0 ||
+				 pg_strncasecmp(tok, "EURO", 4) == 0)
 		{
 			newDateOrder = DATEORDER_DMY;
 			ocnt++;
 		}
-		else if (strcasecmp(tok, "MDY") == 0 ||
-				 strcasecmp(tok, "US") == 0 ||
-				 strncasecmp(tok, "NONEURO", 7) == 0)
+		else if (pg_strcasecmp(tok, "MDY") == 0 ||
+				 pg_strcasecmp(tok, "US") == 0 ||
+				 pg_strncasecmp(tok, "NONEURO", 7) == 0)
 		{
 			newDateOrder = DATEORDER_MDY;
 			ocnt++;
 		}
-		else if (strcasecmp(tok, "DEFAULT") == 0)
+		else if (pg_strcasecmp(tok, "DEFAULT") == 0)
 		{
 			/*
 			 * Easiest way to get the current DEFAULT state is to fetch
@@ -321,8 +322,7 @@ clear_tz(void)
 static bool
 tzset_succeeded(const char *tz)
 {
-	char		tztmp[TZBUF_LEN];
-	char	   *cp;
+	char	   *tztmp;
 	int			tzval;
 
 	/*
@@ -339,9 +339,7 @@ tzset_succeeded(const char *tz)
 	 * Check for known spellings of "UTC".	Note we must downcase the
 	 * input before passing it to DecodePosixTimezone().
 	 */
-	StrNCpy(tztmp, tz, sizeof(tztmp));
-	for (cp = tztmp; *cp; cp++)
-		*cp = tolower((unsigned char) *cp);
+	tztmp = downcase_truncate_identifier(tz, strlen(tz), false);
 	if (DecodePosixTimezone(tztmp, &tzval) == 0)
 		if (tzval == 0)
 			return true;
@@ -410,7 +408,7 @@ assign_timezone(const char *value, bool doit, GucSource source)
 	/*
 	 * Check for INTERVAL 'foo'
 	 */
-	if (strncasecmp(value, "interval", 8) == 0)
+	if (pg_strncasecmp(value, "interval", 8) == 0)
 	{
 		const char *valueptr = value;
 		char	   *val;
@@ -474,7 +472,7 @@ assign_timezone(const char *value, bool doit, GucSource source)
 				HasCTZSet = true;
 			}
 		}
-		else if (strcasecmp(value, "UNKNOWN") == 0)
+		else if (pg_strcasecmp(value, "UNKNOWN") == 0)
 		{
 			/*
 			 * UNKNOWN is the value shown as the "default" for TimeZone in