globstar about micromatch HOT 13 CLOSED

console.log(mm.isMatch('a/b/c/d', 'a/**b')) // => false

It does if you fix the pattern: console.log(mm.isMatch('a/b/c/d', 'a/**d')); // => true

from micromatch.

Hmm, okay. I remember implementing behavior based on my (literal) interpretation of the spec. But it seems this is a bit nuanced.

fwiw I'm refactoring this to use a parser like the one implemented in the refactor branch of extglob. this will allow us to easily make changes based on discussions like this

from micromatch.

Is this intentional?

No, it's not. I'll look into it, thx

from micromatch.

@MartinKolarik returns true (maybe a bug?)

I don't think so. It's not a bug for me.

No, it's not.

@jonschlinkert @MartinKolarik, actually it is logical to be true because * means 0 or more. And no matter that it is globstar not single star, it still make sense.

from micromatch.

@tunnckoCore it would make sense for mm.isMatch('a/b', 'a/**b');, but it shouldn't behave like globstar (matching multiple dirs) if it isn't the only thing in a path part.

from micromatch.

I'm not saying it should behave like globstar. I'm not sure for the my last sentence, just saying it make some sense.

And not, it not behaves as globstar, which is good and expected?

console.log(mm.isMatch('a/b/c/d', 'a/**b')) // => false

from micromatch.

The current behavior is correct. From the bash specification

two adjacent ‘’s used as a single pattern will match all files and zero or more directories and subdirectories. If followed by a ‘/’, two adjacent ‘’s will match only directories and subdirectories.

closing

from micromatch.

Well, the linked spec also doesn't say mm.isMatch('a/b/c', 'a/b**'); shouldn't match, but it (I think correctly) doesn't. I wasn't able to find a spec covering this behavior, but it's documented in minimatch's README, and more importantly, it's how real bash works (just tested that).

from micromatch.

fwiw I'm refactoring this to use a parser like the one implemented in the refactor branch of extglob.

Looks good, that should definitely make it easier to handle situations like this.

from micromatch.

indeed, but the best part is that it's also much faster :)

from micromatch.

Really? It already is quite fast. Will definitely take a closer look at the code.

from micromatch.

more specifically it parses and compiles the regex patterns faster. so initialization will be faster, but not necessarily the actual matching. Hmm, I'm going to create an issue to get my thoughts out on this

from micromatch.

more specifically it parses and compiles the regex patterns faster. so initialization will be faster, but not necessarily the actual matching

yeah that makes sense

from micromatch.