Imply UTF8 validity in implode function (#10780)

Sets the UTF-8 valid flag if all parts are valid, or numeric (which are valid UTF-8 by definition).

* remove unuseful comments

* Imply UTF8 validity in implode function

* revert zend_string_dup change

Author: mvorisek

ext/standard/string.c

@@ -928,9 +928,6 @@ PHP_FUNCTION(explode)
 }
 /* }}} */
 
-/* {{{ An alias for implode */
-/* }}} */
-
 /* {{{ php_implode */
 PHPAPI void php_implode(const zend_string *glue, HashTable *pieces, zval *return_value)
 {
@@ -958,11 +955,14 @@ PHPAPI void php_implode(const zend_string *glue, HashTable *pieces, zval *return
 
 	ptr = strings = do_alloca((sizeof(*strings)) * numelems, use_heap);
 
+	uint32_t flags = ZSTR_GET_COPYABLE_CONCAT_PROPERTIES(glue);
+
 	ZEND_HASH_FOREACH_VAL(pieces, tmp) {
 		if (EXPECTED(Z_TYPE_P(tmp) == IS_STRING)) {
 			ptr->str = Z_STR_P(tmp);
 			len += ZSTR_LEN(ptr->str);
 			ptr->lval = 0;
+			flags &= ZSTR_GET_COPYABLE_CONCAT_PROPERTIES(ptr->str);
 			ptr++;
 		} else if (UNEXPECTED(Z_TYPE_P(tmp) == IS_LONG)) {
 			zend_long val = Z_LVAL_P(tmp);
@@ -981,12 +981,14 @@ PHPAPI void php_implode(const zend_string *glue, HashTable *pieces, zval *return
 			ptr->str = zval_get_string_func(tmp);
 			len += ZSTR_LEN(ptr->str);
 			ptr->lval = 1;
+			flags &= ZSTR_GET_COPYABLE_CONCAT_PROPERTIES(ptr->str);
 			ptr++;
 		}
 	} ZEND_HASH_FOREACH_END();
 
 	/* numelems cannot be 0, we checked above */
 	str = zend_string_safe_alloc(numelems - 1, ZSTR_LEN(glue), len, 0);
+	GC_ADD_FLAGS(str, flags);
 	cptr = ZSTR_VAL(str) + ZSTR_LEN(str);
 	*cptr = 0;
 
@@ -1622,9 +1624,6 @@ PHP_FUNCTION(str_ends_with)
 }
 /* }}} */
 
-/* {{{ An alias for strstr */
-/* }}} */
-
 /* {{{ Finds position of first occurrence of a string within another */
 PHP_FUNCTION(strpos)
 {

ext/zend_test/tests/strings_marked_as_utf8.phpt

@@ -115,6 +115,33 @@ $string = "\xff";
 $string_concat = str_repeat($string, 100);
 var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
 
+echo "implode:\n";
+$arr = ['a', 'b'];
+$string_concat = implode('', $arr);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('|', $arr);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', ['c', ...$arr]);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', [...$arr, 'c']);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', ["\xff", ...$arr]);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', [...$arr, "\xff"]);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode("\xff", $arr);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', []);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode("\xff", []);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', ['a']);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode("\xff", ['a']);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+$string_concat = implode('', [1, 1.0, 'a']);
+var_dump(zend_test_is_string_marked_as_valid_utf8($string_concat));
+
 ?>
 --EXPECT--
 Empty strings:
@@ -159,3 +186,16 @@ bool(true)
 str_repeat:
 bool(true)
 bool(false)
+implode:
+bool(true)
+bool(true)
+bool(true)
+bool(true)
+bool(false)
+bool(false)
+bool(false)
+bool(true)
+bool(true)
+bool(true)
+bool(true)
+bool(true)