[PHP-DEV] [PATCH] default session serialization

[Arpad Ray]

Hi,

Attached is a patch (against HEAD) which adapts the default ("php") session
serializer to serialize the whole array using php_var_serialize instead of
calling it once per element.

This simplifies the serialization code, and allows unicode session keys, or
even ascii keys containing the pipe character.

However it would be a significant BC break, as old serialized session
records would be unreadable after upgrading. We could mitigate this by
providing a script to convert old session files.

It also means the serialized data would be a few bytes larger per key, so
there's definitely some weighing up to do.

Any thoughts?

Arpad

Index: ext/session/session.c
===================================================================
--- ext/session/session.c	(revision 291124)
+++ ext/session/session.c	(working copy)
@@ -783,45 +783,10 @@
 {
 	smart_str buf = {0};
 	php_serialize_data_t var_hash;
-	PS_ENCODE_VARS;
 
 	PHP_VAR_SERIALIZE_INIT(var_hash);
+	php_var_serialize(&buf, &PS(http_session_vars), &var_hash TSRMLS_CC);
 
-	PS_UENCODE_LOOP(
-		if (!struc) {
-			smart_str_appendc(&buf, PS_UNDEF_MARKER);
-		}
-
-		if (key_type == HASH_KEY_IS_STRING) {
-			if (memchr(key.s, PS_DELIMITER, key_length)) {
-				PHP_VAR_SERIALIZE_DESTROY(var_hash);
-				smart_str_free(&buf);
-				return FAILURE;
-			}
-			smart_str_appendl(&buf, key.s, key_length);
-		} else {
-			/* HASH_KEY_IS_UNICODE */
-			char *str = NULL;
-			int len;
-			UErrorCode status = U_ZERO_ERROR;
-
-			zend_unicode_to_string_ex(UG(utf8_conv), &str, &len, key.u, key_length, &status);
-			if (U_FAILURE(status) || memchr(str, PS_DELIMITER, key_length)) {
-				PHP_VAR_SERIALIZE_DESTROY(var_hash);
-				smart_str_free(&buf);
-				if (str) { efree(str); }
-				return FAILURE;
-			}
-			smart_str_appendl(&buf, str, len);
-			efree(str);
-		}
-		smart_str_appendc(&buf, PS_DELIMITER);
-
-		if (struc) {
-			php_var_serialize(&buf, struc, &var_hash TSRMLS_CC);
-		}
-	);
-
 	if (newlen) {
 		*newlen = buf.len;
 	}
@@ -835,61 +800,57 @@
 
 PS_SERIALIZER_DECODE_FUNC(php) /* {{{ */
 {
-	const char *p, *q;
-	char *name;
 	const char *endptr = val + vallen;
-	zval *current;
-	int namelen;
-	int has_value;
+	zval **current, *storage;
+	zstr name;
+	uint namelen;
+	zend_uchar utype;
+	ulong num_key;
 	php_unserialize_data_t var_hash;
+	HashPosition pos;
 
+	if (!vallen) {
+		return SUCCESS;
+	}
+
 	PHP_VAR_UNSERIALIZE_INIT(var_hash);
+	ALLOC_INIT_ZVAL(storage);
 
-	p = val;
+	if (!php_var_unserialize(&storage, (const unsigned char **)&val, (const unsigned char *)endptr, &var_hash TSRMLS_CC) || Z_TYPE_P(storage) != IS_ARRAY) {
+		zval_ptr_dtor(&storage);
+		PHP_VAR_UNSERIALIZE_DESTROY(var_hash);
+		return FAILURE;
+	}
 
-	while (p < endptr) {
+	zend_hash_internal_pointer_reset_ex(Z_ARRVAL_P(storage), &pos);
+	while (zend_hash_get_current_data_ex(Z_ARRVAL_P(storage), (void **)&current, &pos) == SUCCESS) {
 		zval **tmp;
-		has_value = 1;
+		zval_add_ref(current);
 
-		q = p;
-		while (*q != PS_DELIMITER) {
-			if (++q >= endptr) goto break_outer_loop;
+		switch (zend_hash_get_current_key_ex(Z_ARRVAL_P(storage), &name, &namelen, &num_key, 0, &pos)) {
+			case HASH_KEY_IS_STRING:
+				utype = IS_STRING;
+				break;
+			case HASH_KEY_IS_UNICODE:
+				utype = IS_UNICODE;
+				break;
+			default:
+				goto skip;
 		}
 
-		if (*p == PS_UNDEF_MARKER) {
-			if (++p >= endptr) goto break_outer_loop;
-
-			has_value = 0;
-		}
-
-		namelen = q - p;
-		name = estrndup(p, namelen);
-		q++;
-
-		if (zend_hash_find(&EG(symbol_table), name, namelen + 1, (void **) &tmp) == SUCCESS) {
+		if (zend_u_hash_find(&EG(symbol_table), utype, name, namelen, (void **)&tmp) == SUCCESS) {
 			if ((Z_TYPE_PP(tmp) == IS_ARRAY && Z_ARRVAL_PP(tmp) == &EG(symbol_table)) || *tmp == PS(http_session_vars)) {
 				goto skip;
 			}
 		}
 
-		if (has_value) {
-			ALLOC_INIT_ZVAL(current);
-			if (php_var_unserialize(&current, (const unsigned char **) &q, (const unsigned char *) endptr, &var_hash TSRMLS_CC)) {
-				zend_utf8_hash_update(Z_ARRVAL_P(PS(http_session_vars)), name, namelen + 1, &current, sizeof(zval *), NULL);
-			} else {
-				zval_ptr_dtor(&current);
-			}
-		}
-		PS_ADD_VARL(name, namelen);
+		zend_u_hash_update(Z_ARRVAL_P(PS(http_session_vars)), utype, name, namelen, current, sizeof(current), NULL);
 skip:
-		efree(name);
-
-		p = q;
+		zend_hash_move_forward_ex(Z_ARRVAL_P(storage), &pos);
 	}
-break_outer_loop:
 
+	zval_ptr_dtor(&storage);
 	PHP_VAR_UNSERIALIZE_DESTROY(var_hash);
-
 	return SUCCESS;
 }
 /* }}} */

-- 
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php