ICU: fix the use of locale names in Windows for ICU functions 3

Return the function _create_locale because although it returns different locales
for the same locale names, these locales have the same Language Code Identifier
(LCID). Therefore use LCID to get the canonical name of the ICU locale.

Return the ability to use language and country strings in locale names e.g.
"Russian_Russia[.encoding]" or "English_United States[.encoding]".

Author: Marina Polyakova

src/backend/commands/dbcommands.c

@@ -152,10 +152,6 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	char	   *icu_wincollate = NULL;
 	char	   *langtag = NULL;
 	const char *collate;
-#ifdef WIN32
-	char	   *dbcollate_original;
-	char	   *dbctype_original;
-#endif
 
 	/* Extract options from the statement node tree */
 	foreach(option, stmt->options)
@@ -398,11 +394,6 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 
 	/* Check that the chosen locales are valid, and get canonical spellings */
 
-#ifdef WIN32
-	dbcollate_original = dbcollate;
-	dbctype_original = dbctype;
-#endif
-
 	if (!check_locale(LC_CTYPE, dbctype, &canonname, '\0'))
 		ereport(ERROR,
 				(errcode(ERRCODE_WRONG_OBJECT_TYPE),
@@ -439,15 +430,6 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 	/* check lc_collate and lc_ctype for icu if we need it */
 	if (dbcollprovider == COLLPROVIDER_ICU)
 	{
-#ifdef WIN32
-		/* set the libc canonnames for dbcollate and dbctype */
-
-		check_winlocale(dbcollate_original, &canonname, NULL);
-		dbcollate = canonname;
-
-		check_winlocale(dbctype_original, &canonname, NULL);
-		dbctype = canonname;
-#endif
 		if (!check_locale(LC_COLLATE, dbcollate, NULL, dbcollprovider))
 			ereport(ERROR,
 					(errcode(ERRCODE_WRONG_OBJECT_TYPE),
@@ -467,20 +449,19 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
 #ifdef USE_ICU
 	if (dbcollprovider == COLLPROVIDER_ICU)
 	{
+		collate = (const char *) dbcollate;
 #ifdef WIN32
-		if (!locale_is_c(dbcollate))
+		if (!locale_is_c(collate))
 		{
-			check_winlocale(dbcollate, NULL, &icu_wincollate);
+			icu_wincollate = check_icu_winlocale(collate);
 			collate = (const char *) icu_wincollate;
 		}
-#else							/* not WIN32 */
-		collate = (const char *) dbcollate;
-#endif							/* not WIN32 */
+#endif							/* WIN32 */
 		langtag = get_icu_language_tag(collate);
 		collate = get_icu_collate(collate, langtag);
 	}
 	else
-#endif
+#endif							/* USE_ICU */
 	{
 		/* COLLPROVIDER_LIBC */
 		collate = (const char *) dbcollate;

src/backend/utils/adt/pg_locale.c

@@ -1836,9 +1836,6 @@ char2wchar(wchar_t *to, size_t tolen, const char *from, size_t fromlen,
  * If locale is "" return the environment value from setlocale().
  *
  * Otherwise return a malloc'd copy of locale if it is not NULL.
- *
- * In Windows, check that the name is not long (for example,
- * "English_United States[.encoding]").
  */
 static char *
 check_icu_locale(const char *locale)
@@ -1847,19 +1844,20 @@ check_icu_locale(const char *locale)
 	char	   *winlocale = NULL;
 	char	   *result;
 
+	/* Windows locales can be in the format ".codepage" */
+	if (locale && (strlen(locale) == 0 || locale[0] == '.'))
+	{
+		check_locale(LC_COLLATE, locale, &canonname, COLLPROVIDER_LIBC);
+		locale = (const char *) canonname;
+	}
+
 #ifdef WIN32
 	if (!locale_is_c(locale))
 	{
-		check_winlocale(locale, NULL, &winlocale);
+		winlocale = check_icu_winlocale(locale);
 		locale = (const char *) winlocale;
 	}
-#else							/* not WIN32 */
-	if (locale && strlen(locale) == 0)
-	{
-		check_locale(LC_COLLATE, locale, &canonname, COLLPROVIDER_LIBC);
-		locale = (const char *) canonname;
-	}
-#endif							/* not WIN32 */
+#endif
 
 	result = locale ? pstrdup(locale) : NULL;
 

src/backend/utils/init/postinit.c

@@ -442,15 +442,15 @@ CheckMyDatabase(const char *name, bool am_superuser)
 #ifdef WIN32
 		if (!locale_is_c(collcollate))
 		{
-			check_winlocale(collcollate, NULL, &wincollate);
+			wincollate = check_icu_winlocale(collcollate);
 			collcollate = (const char *) wincollate;
 		}
 #endif							/* WIN32 */
 		langtag = get_icu_language_tag(collcollate);
 		collcollate = get_icu_collate(collcollate, langtag);
 	}
 	else
-#endif
+#endif							/* USE_ICU */
 	{
 		/* COLLPROVIDER_LIBC */
 		collcollate = (const char *) collate;

src/bin/initdb/initdb.c

@@ -2357,10 +2357,6 @@ static void
 setlocales(void)
 {
 	char	   *canonname;
-#ifdef WIN32
-	char	   *lc_collate_original;
-	char	   *lc_ctype_original;
-#endif
 
 	/*
 	 * Set up the collation provider if possible and canonicalize the locale
@@ -2409,11 +2405,6 @@ setlocales(void)
 	 * current environment
 	 */
 
-#ifdef WIN32
-	lc_collate_original = pstrdup(lc_collate);
-	lc_ctype_original = pstrdup(lc_ctype);
-#endif
-
 	check_locale_name(LC_CTYPE, lc_ctype, &canonname, '\0');
 	lc_ctype = canonname;
 
@@ -2442,23 +2433,6 @@ setlocales(void)
 	/* check lc_collate and lc_ctype for icu if we need it */
 	if (collprovider == COLLPROVIDER_ICU)
 	{
-#ifdef WIN32
-		/* set the libc canonnames for lc_collate and lc_ctype */
-
-		check_winlocale(lc_collate_original, &canonname, NULL);
-		if (canonname == NULL && lc_collate_original != NULL)
-			exit(1);			/* check_winlocale printed the error */
-
-		pfree(lc_collate);
-		lc_collate = canonname;
-
-		check_winlocale(lc_ctype_original, &canonname, NULL);
-		if (canonname == NULL && lc_ctype_original != NULL)
-			exit(1);			/* check_winlocale printed the error */
-
-		pfree(lc_ctype);
-		lc_ctype = canonname;
-#endif
 		check_locale_name(LC_COLLATE, lc_collate, NULL, collprovider);
 		if (strcmp(lc_collate, lc_ctype) != 0)
 		{
@@ -2485,11 +2459,6 @@ setlocales(void)
 #endif
 
 	set_collation_version();
-
-#ifdef WIN32
-	free(lc_collate_original);
-	free(lc_ctype_original);
-#endif
 }
 
 /*
@@ -3505,9 +3474,6 @@ main(int argc, char *argv[])
  *
  * Otherwise return a malloc'd copy of locale if it is not NULL.
  *
- * In Windows, check that the name is not long (for example,
- * "English_United States[.encoding]").
- *
  * This should match the backend's check_icu_locale() function.
  */
 static char *
@@ -3517,23 +3483,24 @@ check_icu_locale_name(const char *locale)
 	char	   *winlocale = NULL;
 	char	   *result;
 
+	/* Windows locales can be in the format ".codepage" */
+	if (locale && (strlen(locale) == 0 || locale[0] == '.'))
+	{
+		check_locale_name(LC_COLLATE, locale, &canonname, COLLPROVIDER_LIBC);
+		locale = (const char *) canonname;
+	}
+
 #ifdef WIN32
 	if (!locale_is_c(locale))
 	{
-		check_winlocale(locale, NULL, &winlocale);
+		winlocale = check_icu_winlocale(locale);
 
 		if (winlocale == NULL && locale != NULL)
-			exit(1);			/* check_winlocale printed the error */
+			exit(1);			/* check_icu_winlocale printed the error */
 		else
 			locale = winlocale;
 	}
-#else							/* not WIN32 */
-	if (locale && strlen(locale) == 0)
-	{
-		check_locale_name(LC_COLLATE, locale, &canonname, COLLPROVIDER_LIBC);
-		locale = (const char *) canonname;
-	}
-#endif							/* not WIN32 */
+#endif
 
 	result = locale ? pstrdup(locale) : NULL;
 
@@ -3562,19 +3529,19 @@ set_collation_version(void)
 	if (collprovider == COLLPROVIDER_ICU)
 	{
 #ifdef USE_ICU
+		collate = (const char *) lc_collate;
+
 #ifdef WIN32
-		if (!locale_is_c(lc_collate))
+		if (!locale_is_c(collate))
 		{
-			check_winlocale(lc_collate, NULL, &wincollate);
+			wincollate = check_icu_winlocale(collate);
 
-			if (wincollate == NULL && lc_collate != NULL)
-				exit(1);		/* check_winlocale printed the error */
+			if (wincollate == NULL && collate != NULL)
+				exit(1);		/* check_icu_winlocale printed the error */
 			else
 				collate = (const char *) wincollate;
 		}
-#else							/* not WIN32 */
-		collate = (const char *) lc_collate;
-#endif							/* not WIN32 */
+#endif							/* WIN32 */
 
 		langtag = get_icu_language_tag(collate);
 		if (!langtag)
@@ -3585,13 +3552,13 @@ set_collation_version(void)
 			exit(1);
 		}
 		collate = get_icu_collate(collate, langtag);
-#else
+#else							/* not USE_ICU */
 		fprintf(stderr,
 				_("%s: ICU is not supported in this build\n"
 				  "You need to rebuild PostgreSQL using --with-icu.\n"),
 				progname);
 		exit(1);
-#endif
+#endif							/* not USE_ICU */
 	}
 	else
 	{

src/include/port.h

@@ -491,8 +491,7 @@ extern UCollator *open_collator(const char *collate);
 extern char * get_icu_language_tag(const char *localename);
 extern const char *get_icu_collate(const char *locale, const char *langtag);
 #ifdef WIN32
-extern void check_winlocale(const char *winlocale, char **libc_canononame,
-							char **icu_canonname);
+extern char * check_icu_winlocale(const char *winlocale);
 #endif							/* WIN32 */
 #endif							/* USE_ICU */