implement fail-safe static locales for newlocale

this frees applications which need to make temporary use of the C
locale (via uselocale) from the possibility that newlocale might fail.

the C.UTF-8 locale is also provided as a static locale. presently they
behave the same, but this may change in the future.

3 files changed, 46 insertions, 13 deletions

diff --git a/src/locale/freelocale.c b/src/locale/freelocale.c
index ee3f029a..c2ae1a31 100644
--- a/src/locale/freelocale.c
+++ b/src/locale/freelocale.c
@@ -2,9 +2,11 @@
 #include "locale_impl.h"
 #include "libc.h"
 
+int __loc_is_allocated(locale_t);
+
 void freelocale(locale_t l)
 {
-	free(l);
+	if (__loc_is_allocated(l)) free(l);
 }
 
 weak_alias(freelocale, __freelocale);
diff --git a/src/locale/locale_map.c b/src/locale/locale_map.c
index 30aa7fcc..4346bb02 100644
--- a/src/locale/locale_map.c
+++ b/src/locale/locale_map.c
@@ -26,7 +26,7 @@ static const char envvars[][12] = {
 
 static const uint32_t empty_mo[] = { 0x950412de, 0, -1, -1, -1 };
 
-static const struct __locale_map c_dot_utf8 = {
+const struct __locale_map __c_dot_utf8 = {
 	.map = empty_mo,
 	.map_size = sizeof empty_mo,
 	.name = "C.UTF-8"
@@ -58,7 +58,7 @@ const struct __locale_map *__get_locale(int cat, const char *val)
 
 	if (builtin) {
 		if (cat == LC_CTYPE && val[1]=='.')
-			return (void *)&c_dot_utf8;
+			return (void *)&__c_dot_utf8;
 		return 0;
 	}
 
@@ -117,7 +117,7 @@ const struct __locale_map *__get_locale(int cat, const char *val)
 
 	/* For LC_CTYPE, never return a null pointer unless the
 	 * requested name was "C" or "POSIX". */
-	if (!new && cat == LC_CTYPE) new = (void *)&c_dot_utf8;
+	if (!new && cat == LC_CTYPE) new = (void *)&__c_dot_utf8;
 
 	UNLOCK(lock);
 	return new;
diff --git a/src/locale/newlocale.c b/src/locale/newlocale.c
index 4e0cbd34..89d36b1d 100644
--- a/src/locale/newlocale.c
+++ b/src/locale/newlocale.c
@@ -3,21 +3,52 @@
 #include "locale_impl.h"
 #include "libc.h"
 
+extern const struct __locale_map __c_dot_utf8;
+
+static const struct __locale_struct c_locale = { 0 };
+static const struct __locale_struct c_dot_utf8_locale = {
+	.cat[LC_CTYPE] = &__c_dot_utf8
+};
+
+int __loc_is_allocated(locale_t loc)
+{
+	return loc && loc != &c_locale && loc != &c_dot_utf8_locale;
+}
+
 locale_t __newlocale(int mask, const char *name, locale_t loc)
 {
-	int i;
+	int i, j;
+	struct __locale_struct tmp;
+	const struct __locale_map *lm;
 
-	if (!loc) {
-		loc = malloc(sizeof *loc);
-		if (!loc) return 0;
+	/* For locales with allocated storage, modify in-place. */
+	if (__loc_is_allocated(loc)) {
 		for (i=0; i<LC_ALL; i++)
-			if (!(mask & (1<<i)))
-				loc->cat[i] = __get_locale(i, "");
+			if (mask & (1<<i))
+				loc->cat[i] = __get_locale(i, name);
+		return loc;
+	}
+
+	/* Otherwise, build a temporary locale object, which will only
+	 * be instantiated in allocated storage if it does not match
+	 * one of the built-in static locales. This makes the common
+	 * usage case for newlocale, getting a C locale with predictable
+	 * behavior, very fast, and more importantly, fail-safe. */
+	for (j=i=0; i<LC_ALL; i++) {
+		if (loc && !(mask & (1<<i)))
+			lm = loc->cat[i];
+		else
+			lm = __get_locale(i, mask & (1<<i) ? name : "");
+		if (lm) j++;
+		tmp.cat[i] = lm;
 	}
 
-	for (i=0; i<LC_ALL; i++)
-		if (mask & (1<<i))
-			loc->cat[i] = __get_locale(i, name);
+	if (!j)
+		return (locale_t)&c_locale;
+	if (j==1 && tmp.cat[LC_CTYPE]==c_dot_utf8_locale.cat[LC_CTYPE])
+		return (locale_t)&c_dot_utf8_locale;
+
+	if ((loc = malloc(sizeof *loc))) *loc = tmp;
 
 	return loc;
 }