Fix #74170: locale information change after mime_content_type

Some functions in libmagic (distributed with fileinfo extension) perform this sequence of calls:
func() {
setlocale(LC_TYPE, "C")
.. do some work ..
setlocale(LC_TYPE, "")
}

It effectively resets LC_TYPE if it that was set before the function call.

To avoid manipulations with current locale at all, the problematic functions
were modified to use locale-independent functions.

Author: turchanov

NEWS

@@ -9,6 +9,10 @@ PHP                                                                        NEWS
 - CURL:
   . Implemented FR #77711 (CURLFile should support UNICODE filenames). (cmb)
 
+- Fileinfo:
+  . Fixed bug #74170 (locale information change after mime_content_type).
+    (Sergei Turchanov)
+
 - GD:
   . Fixed bug #78923 (Artifacts when convoluting image with transparency).
     (wilson chen)

ext/fileinfo/libmagic/funcs.c

@@ -477,11 +477,9 @@ file_replace(struct magic_set *ms, const char *pat, const char *rep)
 	zend_string *repl;
 	size_t rep_cnt = 0;
 
-	(void)setlocale(LC_CTYPE, "C");
-
 	opts |= PCRE2_MULTILINE;
 	convert_libmagic_pattern(&patt, (char*)pat, strlen(pat), opts);
-	if ((pce = pcre_get_compiled_regex_cache(Z_STR(patt))) == NULL) {
+	if ((pce = pcre_get_compiled_regex_cache_ex(Z_STR(patt), 0)) == NULL) {
 		zval_ptr_dtor(&patt);
 		rep_cnt = -1;
 		goto out;
@@ -503,7 +501,6 @@ file_replace(struct magic_set *ms, const char *pat, const char *rep)
 	zend_string_release_ex(res, 0);
 
 out:
-	(void)setlocale(LC_CTYPE, "");
 	return rep_cnt;
 }
 

ext/fileinfo/libmagic/readcdf.c

@@ -116,14 +116,24 @@ cdf_app_to_mime(const char *vbuf, const struct nv *nv)
 {
 	size_t i;
 	const char *rv = NULL;
+	char *vbuf_lower;
 
-	(void)setlocale(LC_CTYPE, "C");
-	for (i = 0; nv[i].pattern != NULL; i++)
-		if (strcasestr(vbuf, nv[i].pattern) != NULL) {
+	vbuf_lower = zend_str_tolower_dup(vbuf, strlen(vbuf));
+	for (i = 0; nv[i].pattern != NULL; i++) {
+		char *pattern_lower;
+		int found;
+
+		pattern_lower = zend_str_tolower_dup(nv[i].pattern, strlen(nv[i].pattern));
+		found = (strstr(vbuf_lower, pattern_lower) != NULL);
+		efree(pattern_lower);
+
+		if (found) {
 			rv = nv[i].mime;
 			break;
 		}
-	(void)setlocale(LC_CTYPE, "");
+	}
+
+	efree(vbuf_lower);
 	return rv;
 }
 

ext/fileinfo/libmagic/softmagic.c

@@ -422,27 +422,25 @@ match(struct magic_set *ms, struct magic *magic, uint32_t nmagic,
 private int
 check_fmt(struct magic_set *ms, const char *fmt)
 {
-	pcre2_code *pce;
-	uint32_t re_options, capture_count;
+	pcre_cache_entry *pce;
 	int rv = -1;
 	zend_string *pattern;
 
 	if (strchr(fmt, '%') == NULL)
 		return 0;
 
-	(void)setlocale(LC_CTYPE, "C");
 	pattern = zend_string_init("~%[-0-9\\.]*s~", sizeof("~%[-0-9\\.]*s~") - 1, 0);
-	if ((pce = pcre_get_compiled_regex(pattern, &capture_count, &re_options)) == NULL) {
+	if ((pce = pcre_get_compiled_regex_cache_ex(pattern, 0)) == NULL) {
 		rv = -1;
 	} else {
-		pcre2_match_data *match_data = php_pcre_create_match_data(capture_count, pce);
+		pcre2_code *re = php_pcre_pce_re(pce);
+		pcre2_match_data *match_data = php_pcre_create_match_data(0, re);
 		if (match_data) {
-			rv = pcre2_match(pce, (PCRE2_SPTR)fmt, strlen(fmt), 0, re_options, match_data, php_pcre_mctx()) > 0;
+			rv = pcre2_match(re, (PCRE2_SPTR)fmt, strlen(fmt), 0, 0, match_data, php_pcre_mctx()) > 0;
 			php_pcre_free_match_data(match_data);
 		}
 	}
-	zend_string_release_ex(pattern, 0);
-	(void)setlocale(LC_CTYPE, "");
+	zend_string_release(pattern);
 	return rv;
 }
 

ext/fileinfo/tests/bug74170.phpt

@@ -0,0 +1,22 @@
+--TEST--
+Bug #74170 locale information change after mime_content_type
+--SKIPIF--
+<?php
+if (!class_exists('finfo'))
+	die('skip no fileinfo extension');
+if (!extension_loaded('intl'))
+	die('skip intl extension not enabled');
+if (setlocale(LC_CTYPE, 'ru_RU.koi8r') === false)
+	die('skip ru_RU.koi8r locale is not available');
+?>
+--FILE--
+<?php
+var_dump(setlocale(LC_CTYPE, 'ru_RU.koi8r'));
+var_dump(nl_langinfo(CODESET));
+var_dump(mime_content_type(__DIR__ . '/resources/test.ppt'));
+var_dump(nl_langinfo(CODESET));
+--EXPECT--
+string(11) "ru_RU.koi8r"
+string(6) "KOI8-R"
+string(29) "application/vnd.ms-powerpoint"
+string(6) "KOI8-R"