PathMatcher does not respect pathEqualityUsesCanonicalForm setting about jimfs HOT 5 CLOSED

I'm not very familiar with these settings, but I would have guessed that setNameCanonicalNormalization(CASE_FOLD_ASCII) would make "glob:**.txt" match files with a TXT extension, and it sounds like you're saying it does. Is that the case, and you're just expecting a different behavior (maybe because of setPathEqualityUsesCanonicalForm(true))?

from jimfs.

I'm sorry, i missed out an important negation from my explanation.
I edited my original comment.

from jimfs.

Hmm, that's curious. This should work, and there are even tests that check that the regex Patterns created by PathNormalization.compilePattern are case insensitive when CASE_FOLD_ASCII is one of the normalizations.

Edit: Oh, but it uses the display normalizations for that. Hmm.

I can't test it right now, but I'm assuming that on Windows said PathMatcher would in fact match a file with a TXT extension? It may be that I just need to use the canonical normalizations when path-equality-uses-canonical-form is true.

from jimfs.

Yes, the default PathMatcher on Windows can match files regardless of casing.

from jimfs.

Thanks 👍

from jimfs.