Support # comments in regex #3735
Conversation
|
This pull request has been marked as ready for review. |
src/Type/Regex/RegexGroupParser.php
Outdated
| // The regex engine ignores everything after the (?# until the first closing parenthesis | ||
| $regex = preg_replace('/\(\?#[^)]*\)/', '', $regex) ?? ''; |
There was a problem hiding this comment.
general comments which work independent from modifiers
|
Thank you! |
| $modifiers = $this->regexExpressionHelper->getPatternModifiers($regex) ?? ''; | ||
| foreach (self::NOT_SUPPORTED_MODIFIERS as $notSupportedModifier) { | ||
| if (str_contains($modifiers, $notSupportedModifier)) { | ||
| return null; | ||
| } | ||
| } | ||
|
|
||
| if (str_contains($modifiers, 'x')) { | ||
| // in freespacing mode the # character starts a comment and runs until the end of the line | ||
| $regex = preg_replace('/[^?]#.*/', '', $regex) ?? ''; |
There was a problem hiding this comment.
This will also get rid of the character before # due to [^?]. I realize in most sane cases this will be a space and thus it's harmless, but IMO this should rather be a lookbehind like (?<!\?). I assume the goal was to exclude (?#...) but those are already ignored/handled by the lexer https://github.com/phpstan/phpstan-src/blob/2.0.x/resources/RegexGrammar.pp#L80-L82
There was a problem hiding this comment.
will look into it, thanks for the heads up.
(I had to use this additional char before the # as we would otherwise destroy comments in (?# ..) notation). since we replace before lexing we need to make sure we don't turn the comments into something which is no longer a comment
closes phpstan/phpstan#12242
see https://www.regular-expressions.info/freespacing.html#freecomment
will look into whitespace normalization for the
x-modifer in a separate PR