Imply UTF8 validity in explode function

Author: mvorisek

ext/standard/string.c

@@ -816,12 +816,14 @@ PHP_FUNCTION(wordwrap)
 /* }}} */
 
 /* {{{ php_explode */
-PHPAPI void php_explode(const zend_string *delim, zend_string *str, zval *return_value, zend_long limit)
+PHPAPI void php_explode(const zend_string *delim, const zend_string *str, zval *return_value, zend_long limit)
 {
 	const char *p1 = ZSTR_VAL(str);
 	const char *endp = ZSTR_VAL(str) + ZSTR_LEN(str);
 	const char *p2 = php_memnstr(ZSTR_VAL(str), ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
 	zval  tmp;
+    zend_string *tmp2;
+	uint32_t flags = ZSTR_GET_COPYABLE_CONCAT_PROPERTIES_BOTH(delim, str);
 
 	if (p2 == NULL) {
 		ZVAL_STR_COPY(&tmp, str);
@@ -831,15 +833,19 @@ PHPAPI void php_explode(const zend_string *delim, zend_string *str, zval *return
 		ZEND_HASH_FILL_PACKED(Z_ARRVAL_P(return_value)) {
 			do {
 				ZEND_HASH_FILL_GROW();
-				ZEND_HASH_FILL_SET_STR(zend_string_init_fast(p1, p2 - p1));
+                tmp2 = zend_string_init_fast(p1, p2 - p1);
+                GC_ADD_FLAGS(tmp2, flags);
+				ZEND_HASH_FILL_SET_STR(tmp2);
 				ZEND_HASH_FILL_NEXT();
 				p1 = p2 + ZSTR_LEN(delim);
 				p2 = php_memnstr(p1, ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
 			} while (p2 != NULL && --limit > 1);
 
 			if (p1 <= endp) {
 				ZEND_HASH_FILL_GROW();
-				ZEND_HASH_FILL_SET_STR(zend_string_init_fast(p1, endp - p1));
+                tmp2 = zend_string_init_fast(p1, endp - p1);
+                GC_ADD_FLAGS(tmp2, flags);
+				ZEND_HASH_FILL_SET_STR(tmp2);
 				ZEND_HASH_FILL_NEXT();
 			}
 		} ZEND_HASH_FILL_END();
@@ -848,13 +854,14 @@ PHPAPI void php_explode(const zend_string *delim, zend_string *str, zval *return
 /* }}} */
 
 /* {{{ php_explode_negative_limit */
-PHPAPI void php_explode_negative_limit(const zend_string *delim, zend_string *str, zval *return_value, zend_long limit)
+PHPAPI void php_explode_negative_limit(const zend_string *delim, const zend_string *str, zval *return_value, zend_long limit)
 {
 #define EXPLODE_ALLOC_STEP 64
 	const char *p1 = ZSTR_VAL(str);
 	const char *endp = ZSTR_VAL(str) + ZSTR_LEN(str);
 	const char *p2 = php_memnstr(ZSTR_VAL(str), ZSTR_VAL(delim), ZSTR_LEN(delim), endp);
 	zval  tmp;
+	uint32_t flags = ZSTR_GET_COPYABLE_CONCAT_PROPERTIES_BOTH(delim, str);
 
 	if (p2 == NULL) {
 		/*
@@ -880,6 +887,7 @@ PHPAPI void php_explode_negative_limit(const zend_string *delim, zend_string *st
 		/* limit is at least -1 therefore no need of bounds checking : i will be always less than found */
 		for (i = 0; i < to_return; i++) { /* this checks also for to_return > 0 */
 			ZVAL_STRINGL(&tmp, positions[i], (positions[i+1] - ZSTR_LEN(delim)) - positions[i]);
+            GC_ADD_FLAGS(Z_STR(tmp), flags);
 			zend_hash_next_index_insert_new(Z_ARRVAL_P(return_value), &tmp);
 		}
 		efree((void *)positions);

ext/zend_test/tests/strings_marked_as_utf8.phpt

@@ -142,6 +142,34 @@ var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
 $string_concat = implode('', [1, 1.0, 'a']);
 var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
 
+echo "explode:\n";
+$string = 'můj žlutý kůň';
+$firstByte = substr('ů', 0, 1); // byte present in $string, but itself it is not valid UTF-8
+$dumpUtf8ValidityArrFx = function (array $strings): void {
+    echo (implode(' ', array_map(fn ($v) => zend_test_is_string_marked_as_valid_utf8($v) ? 'true' : 'false', $strings)) ?: 'empty') . "\n";
+};
+$dumpUtf8ValidityArrFx(explode("\xff", ''));
+$dumpUtf8ValidityArrFx(explode('ů', $string));
+$dumpUtf8ValidityArrFx(explode('ů', $string . "\xff"));
+$dumpUtf8ValidityArrFx(explode('ů', $string, 1));
+$dumpUtf8ValidityArrFx(explode('ů', $string . "\xff", 1));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string . "\xff"));
+$dumpUtf8ValidityArrFx(explode("\xff", $string));
+$dumpUtf8ValidityArrFx(explode("\xff", $string . "\xff"));
+$dumpUtf8ValidityArrFx(explode('ů', $string, -1));
+$dumpUtf8ValidityArrFx(explode('ů', $string . "\xff", -1));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string, -1));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string . "\xff", -1));
+$dumpUtf8ValidityArrFx(explode("\xff", $string, -1));
+$dumpUtf8ValidityArrFx(explode("\xff", $string . "\xff", -1));
+$dumpUtf8ValidityArrFx(explode('ů', $string, -2));
+$dumpUtf8ValidityArrFx(explode('ů', $string . "\xff", -2));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string, -2));
+$dumpUtf8ValidityArrFx(explode($firstByte, $string . "\xff", -2));
+$dumpUtf8ValidityArrFx(explode("\xff", $string, -2));
+$dumpUtf8ValidityArrFx(explode("\xff", $string . "\xff", -2));
+
 ?>
 --EXPECT--
 Empty strings:
@@ -199,3 +227,25 @@ bool(true)
 bool(true)
 bool(true)
 bool(true)
+explode:
+false
+true true true
+false false false
+true
+false
+false false false false false
+false false false false false
+false
+false false
+true true
+false false
+false false false false
+false false false false
+empty
+false
+true
+false
+false false false
+false false false
+empty
+empty