fix: PhpUnitStrictFixer - do not crash on property having the name of method to fix
#7804
Conversation
| @@ -686,17 +686,25 @@ public static function provideFunctionReturnTypeInfoPre80Cases(): iterable | |||
|
|
|||
| /** | |||
| * @dataProvider provideIsTheSameClassCallCases | |||
| * | |||
| * @param list<int> $sameClassCallIndices | |||
| { | ||
| $tokens = Tokens::fromCode($code); | ||
| $analyzer = new FunctionsAnalyzer(); | ||
|
|
||
| self::assertSame($isTheSameClassCall, $analyzer->isTheSameClassCall($tokens, $index)); | ||
| for ($index = $tokens->count(); $index >= 0; --$index) { |
There was a problem hiding this comment.
| for ($index = $tokens->count(); $index >= 0; --$index) { | |
| for ($index = $tokens->count() - 1; $index >= 0; --$index) { |
if we have size==2, then array[2] doesn't exist
There was a problem hiding this comment.
Previous tests had checked for -1: https://github.com/PHP-CS-Fixer/PHP-CS-Fixer/blob/v3.49.0/tests/Tokenizer/Analyzer/FunctionsAnalyzerTest.php#L711-L715
Maybe we want to throw an error for a non-existing index? Definitely, we want to check what will happen if the non-existing index is given.
There was a problem hiding this comment.
to me code looks very counterintuitive to iterate over all existing indexes plus one non-existing , out of the blue it looks almost like a typo error to be fixed (even by sca detecting calling $array[non-exist-index], if sca advanced enough).
i would encourage to have dedicated check for non-existing index.
There was a problem hiding this comment.
Maybe we want to throw an error for a non-existing index?
out of the blue sounds like possible variant, probably how I would drive it if creating this function from scratch [~indexoutofboundsexception].
we could check how code would react, i hope we do not rely on this existing behaviour to return false instead of error
There was a problem hiding this comment.
No description provided.