Introduce "builtin" collation provider.

New provider for collations, like "libc" or "icu", but without any
external dependency.

Initially, the only locale supported by the builtin provider is "C",
which is identical to the libc provider's "C" locale. The libc
provider's "C" locale has always been treated as a special case that
uses an internal implementation, without using libc at all -- so the
new builtin provider uses the same implementation.

The builtin provider's locale is independent of the server environment
variables LC_COLLATE and LC_CTYPE. Using the builtin provider, the
database collation locale can be "C" while LC_COLLATE and LC_CTYPE are
set to "en_US", which is impossible with the libc provider.

By offering a new builtin provider, it clarifies that the semantics of
a collation using this provider will never depend on libc, and makes
it easier to document the behavior.

Discussion: https://postgr.es/m/[email protected]
Discussion: https://postgr.es/m/[email protected]
Discussion: https://postgr.es/m/ff4c2f2f9c8fc7ca27c1c24ae37ecaeaeaff6b53.camel%40j-davis.com
Reviewed-by: Daniel Vérité, Peter Eisentraut, Jeremy Schneider

1 files changed, 37 insertions, 16 deletions

diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index de58002a5d4..8d53ef4a1fc 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -145,7 +145,9 @@ static char *lc_numeric = NULL;
 static char *lc_time = NULL;
 static char *lc_messages = NULL;
 static char locale_provider = COLLPROVIDER_LIBC;
+static bool builtin_locale_specified = false;
 static char *datlocale = NULL;
+static bool icu_locale_specified = false;
 static char *icu_rules = NULL;
 static const char *default_text_search_config = NULL;
 static char *username = NULL;
@@ -2368,7 +2370,7 @@ setlocales(void)
 			lc_monetary = locale;
 		if (!lc_messages)
 			lc_messages = locale;
-		if (!datlocale && locale_provider == COLLPROVIDER_ICU)
+		if (!datlocale && locale_provider != COLLPROVIDER_LIBC)
 			datlocale = locale;
 	}
 
@@ -2395,14 +2397,20 @@ setlocales(void)
 	lc_messages = canonname;
 #endif
 
-	if (locale_provider == COLLPROVIDER_ICU)
+	if (locale_provider != COLLPROVIDER_LIBC && datlocale == NULL)
+		pg_fatal("locale must be specified if provider is %s",
+				 collprovider_name(locale_provider));
+
+	if (locale_provider == COLLPROVIDER_BUILTIN)
+	{
+		if (strcmp(datlocale, "C") != 0)
+			pg_fatal("invalid locale name \"%s\" for builtin provider",
+					 datlocale);
+	}
+	else if (locale_provider == COLLPROVIDER_ICU)
 	{
 		char	   *langtag;
 
-		/* acquire default locale from the environment, if not specified */
-		if (datlocale == NULL)
-			pg_fatal("ICU locale must be specified");
-
 		/* canonicalize to a language tag */
 		langtag = icu_language_tag(datlocale);
 		printf(_("Using language tag \"%s\" for ICU locale \"%s\".\n"),
@@ -2447,7 +2455,8 @@ usage(const char *progname)
 			 "                            set default locale in the respective category for\n"
 			 "                            new databases (default taken from environment)\n"));
 	printf(_("      --no-locale           equivalent to --locale=C\n"));
-	printf(_("      --locale-provider={libc|icu}\n"
+	printf(_("      --builtin-locale=LOCALE   set builtin locale name for new databases\n"));
+	printf(_("      --locale-provider={builtin|libc|icu}\n"
 			 "                            set default locale provider for new databases\n"));
 	printf(_("      --pwfile=FILE         read password for the new superuser from file\n"));
 	printf(_("  -T, --text-search-config=CFG\n"
@@ -2609,9 +2618,9 @@ setup_locale_encoding(void)
 	else
 	{
 		printf(_("The database cluster will be initialized with this locale configuration:\n"));
-		printf(_("  provider:    %s\n"), collprovider_name(locale_provider));
-		if (datlocale)
-			printf(_("  ICU locale:  %s\n"), datlocale);
+		printf(_("  default collation provider:  %s\n"), collprovider_name(locale_provider));
+		if (locale_provider != COLLPROVIDER_LIBC)
+			printf(_("  default collation locale:    %s\n"), datlocale);
 		printf(_("  LC_COLLATE:  %s\n"
 				 "  LC_CTYPE:    %s\n"
 				 "  LC_MESSAGES: %s\n"
@@ -3104,9 +3113,10 @@ main(int argc, char *argv[])
 		{"allow-group-access", no_argument, NULL, 'g'},
 		{"discard-caches", no_argument, NULL, 14},
 		{"locale-provider", required_argument, NULL, 15},
-		{"icu-locale", required_argument, NULL, 16},
-		{"icu-rules", required_argument, NULL, 17},
-		{"sync-method", required_argument, NULL, 18},
+		{"builtin-locale", required_argument, NULL, 16},
+		{"icu-locale", required_argument, NULL, 17},
+		{"icu-rules", required_argument, NULL, 18},
+		{"sync-method", required_argument, NULL, 19},
 		{NULL, 0, NULL, 0}
 	};
 
@@ -3274,7 +3284,9 @@ main(int argc, char *argv[])
 										 "-c debug_discard_caches=1");
 				break;
 			case 15:
-				if (strcmp(optarg, "icu") == 0)
+				if (strcmp(optarg, "builtin") == 0)
+					locale_provider = COLLPROVIDER_BUILTIN;
+				else if (strcmp(optarg, "icu") == 0)
 					locale_provider = COLLPROVIDER_ICU;
 				else if (strcmp(optarg, "libc") == 0)
 					locale_provider = COLLPROVIDER_LIBC;
@@ -3283,11 +3295,16 @@ main(int argc, char *argv[])
 				break;
 			case 16:
 				datlocale = pg_strdup(optarg);
+				builtin_locale_specified = true;
 				break;
 			case 17:
-				icu_rules = pg_strdup(optarg);
+				datlocale = pg_strdup(optarg);
+				icu_locale_specified = true;
 				break;
 			case 18:
+				icu_rules = pg_strdup(optarg);
+				break;
+			case 19:
 				if (!parse_sync_method(optarg, &sync_method))
 					exit(1);
 				break;
@@ -3317,7 +3334,11 @@ main(int argc, char *argv[])
 		exit(1);
 	}
 
-	if (datlocale && locale_provider != COLLPROVIDER_ICU)
+	if (builtin_locale_specified && locale_provider != COLLPROVIDER_BUILTIN)
+		pg_fatal("%s cannot be specified unless locale provider \"%s\" is chosen",
+				 "--builtin-locale", "builtin");
+
+	if (icu_locale_specified && locale_provider != COLLPROVIDER_ICU)
 		pg_fatal("%s cannot be specified unless locale provider \"%s\" is chosen",
 				 "--icu-locale", "icu");