Re: .gitattributes override behavior (possible bug, or documentation bug)

From: Jeff King <peff@peff.net>
Date: Mar 20, 2018, 04:04 UTC
Message-ID: <20180320040411.GB12938@sigill.intra.peff.net>
In-Reply-To: <CAHnyXxRcwq40W4tKm=Kscrsnb77yh7=eGDE=r5AZq073MPX9AQ@mail.gmail.com>

On Mon, Mar 19, 2018 at 11:10:26PM -0400, Dakota Hawkins wrote:

Show 6 quoted lines

> > As you noted below, that second line does not match your path, because
> > attributes on a directory aren't applied recursively. And it has nothing
> > to do with overriding. If you remove the png line entirely, you can see
> > that we still do not match it. You need to use "*" to match the paths.
> 
> Ah, yes, I see that. Inconsistent with .gitignore (more below), right?

Yes, it is.

Show 11 quoted lines

> > I could not find anything useful in gitattributes(5). There's some old
> > discussion here:
> >
> >   https://public-inbox.org/git/slrnkldd3g.1l4.jan@majutsushi.net/
> 
> If I follow that correctly: There's some initial speculation that it
> would be OK to apply the attributes recursively, which is then shot
> down because it wasn't designed to be recursive (though I don't see a
> different, technical reason for that), followed by finding a (this
> same?) solution/workaround for the original problem. Is that about
> right?

Right. The technical reason is mostly "that is not how it was designed, and it would possibly break some corner cases if we switched it now".

Show 7 quoted lines

> > I think just "/.readme-docs/**" should be sufficient for your case. You
> > could also probably write "*" inside ".readme-docs/.gitattributes",
> > which may be simpler (you don't need "**" there because patterns without
> > a slash are just matched directly against the basename).
> 
> Wouldn't that make the "*" inside ".readme-docs/.gitattributes",
> technically recursive when "*" matches a directory?

Sort of. The pattern is applied recursively to each basename, but the pattern itself does not match recursively. That's probably splitting hairs, though. :)

Show 5 quoted lines

> It's always seemed to me that both were necessary to explicitly match
> things in a directory and its subdirectories (example, IIRC: "git
> ls-files -- .gitattributes" vs "git ls-files -- .gitattributes
> **/.gitattributes"). Maybe that example is peculiar in that its a
> dotfile and can't have a wildcard before the dot?

Those are pathspecs, which are not quite the same as gitattributes. They don't do the magic basename matching.

For pathspecs a "*" matches across slashes, which is what allows "git log -- '*.h'" to work (but not a suffix wildcard like 'foo*').

> I guess my takeaway is that it would be _good_ if the gitattributes
> documentation contained the caveat about not matching directories
> recursively, but _great_ if gitattributes and gitignore (and whatever
> else there is) were consistent.

I agree it would be nice if they were consistent (and pathspecs, too). But unfortunately at this point there's a maze of backwards compatibility to deal with.

-Peff