Markus Staab
Markus Staab
Having xdebug loaded while running phpstan is a nogo IMO (if you don't want to debug phpstan itself). In my experience just having xdebug loaded no matter activated or not,...
AFAIR was the carbon phpdoc perf problem fixed a while ago
does the result only influence when the progressbar gets rendered or does it also has a impact on the overall analysis time?
I wonder whether we should print out a warning for the php cli jit - in a similar way we do it for xdebug - as it seems it can...
I am working on it and will provide a PR later today
I think its a bug in the Hoa AST, which we walk here: https://github.com/phpstan/phpstan-src/blob/d842380a1f07014d045dd42d6ccd4204a85e688a/src/Type/Php/RegexArrayShapeMatcher.php#L204-L235 @mvorisek could you have a look at the grammar etc?
thanks. I will have another deeper look then. support would still be welcome :) _edit: fixes in https://github.com/phpstan/phpstan-src/pull/3184_
In my experience breaking the baseline or introduce subtle mistakes by manual editing happens pretty often. Thats the reason why we usually generate the baseline from scratch using the phpstan...
the problem is not related to the used loop. it also happens in regular if statements. it looks like the nested assignment triggers the false-positive: https://phpstan.org/r/2f35dfa2-efc6-42c5-a8bf-5a7328180c69 when we use a...
you need to add phpdocs to make phpstan aware that a function/method is throwing (alternatively create a extension for that). just by looking at the sources, phpstan is not able...