[PATCH 1/2] fnmatch: allow character class names with 'z'

[Adhemerval Zanella Netto]

On 23/05/23 18:55, Carlo Arenas wrote:
> On Tue, May 23, 2023 at 11:09 AM Adhemerval Zanella Netto
> <adhemerval.zanella@linaro.org> wrote:
>> . Also, current practice is to both add
>> a bug report if this is a user-visible issue (which seems so) along with
>> a testcase to avoid any potential regression.  Cold you provide both?
> 
> Sure; but I would like to clarify that the bug I was really targeting
> has a bugzilla[1] entry already and the fix[2] for it includes "part
> 2" of a fix for this.

Thanks, we are now enforcing regression tests on every bug report.  Since
it already have a reproducer, just follow other fnmatch tests (for instance
posix/tst-fnmatch7.c).

> 
> My assumption was that this bug is too old and has no user effect
> (unless someone adds a custom class name with 'z' in their name), and
> in the 20 years that had gone by, there are only a handful of those.
> 
> Either way, I will be adding tests for both bugs in a v2, but wanted
> to be sure you would have them split (which I would normally agree
> with), or maybe I should have squashed both commits instead.

I would say to just squash them on same patch.

> 
> Carlo
> 
> [1] https://sourceware.org/bugzilla/show_bug.cgi?id=30483
> [2] https://patchwork.sourceware.org/project/glibc/patch/20230523073732.6956-3-carenas@gmail.com/