optimized some ignore rules in the phpstan-baseline.neon
#456
Conversation
| $limited = ini_get('pcre.backtrack_limit'); | ||
| ini_set('pcre.backtrack_limit', '4'); | ||
| $things = ['"Google Chrome";v="95", "Chromium";v="95", ";Not A Brand";v="99"']; | ||
| self::assertSame(['"Google Chrome";v="95", "Chromium";v="95", ";Not A Brand";v="99"'], Psr7\Header::normalize($things)); |
There was a problem hiding this comment.
This assertion is bogus. However the backtrack limit is a good reason to not rely on regular expressions here. See: #447
There was a problem hiding this comment.
Agree. Rely on the regular expression may not a good idea. For example, parsing the new Permissions-Policy's value, eg: Permissions-Policy: fullscreen=(self "https://example.com" "https://another.example.com"), geolocation=*, camera=(), there were the parentheses, self, none, empty string and double quotes strings, it shall be so hard to do that.
TheNorthMemory
changed the title
phpstan-baseline.neonphpstan-baseline.neon
|
There are a lot of changes here, and it's not clear if any bugs are actually fixed. Since there are merge conflicts, we're not able to merge at this time. Please feel free to re-submit a PR fixing a specific bug. |
phpstan-baseline.neon;pcre.backtrack_limitsettings may be affected the functionpreg_split's behavior. While there were a customized value (less than 5), it can caused theHeader::parseandHeader::normalizeworking incorrectly. Optimize codes for this case.