This is the mail archive of the libc-locales@sourceware.org mailing list for the GNU libc locales project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH][BZ 18934] hu_HU: Fix multiple sorting bugs.


Hi,

Please use the patch I attach now to this mail, not to the previous
one. Sorry for the confusion!

I checked the previous patch many times, yet I missed something that
I've just discovered after sending the previous mail. I forgot one of
the compound letters from the unittest.

The only change from the previous patch is the addition of these few
more lines in the unittest, so it has an even better coverage. The
patch to the locale definiton is unchanged.

I've re-run the test and of course it still passes :)

Thanks,
egmont

On Tue, Oct 13, 2015 at 11:56 PM, Egmont Koblinger <egmont@gmail.com> wrote:
> Hi,
>
> Could you please review and apply the attached patch?
>
> Recommended commit message body (feel free to edit as you please):
> -----
> Fix sorting of long consonants, a regression introduced by #13547. Fix
> inconsistencies in uppercase vs. lowercase sorting. Fix diacritic
> ordering. Fix ordering of foreign accents.
>
> Add an extensive test file.
>
>     [BZ #18934]
>     * locales/hu_HU: Fix multiple bugs.
>     * hu_HU.in: New file.
>     * Makefile (test-input): Add hu_HU.UTF-8.
> -----
>
> I know that generally one patch per issue is a cleaner approach, but
> this time apologize for an all-in-one: the patches would heavily
> conflict, and it would be really cumbersome to unittest an incremental
> series. Instead, think about it as TDD (test driven development): I
> attach a decent unittest with explanations and pointers to the rules,
> and a locale definition that implements them.
>
> The addressed bugs are:
>
> - The fix to bug 13547 was incorrect and introduced a regression. It
> fixed a corner case, whereas I didn't realize it broke a more typical
> once. See details over there.
>
> - Two minor bugs/inconsistencies wrt. sorting upper/lowercase values,
> as described in bug 18587.
>
> - Someone enabled backwards ordering of diacrits by default (bug
> 17750), breaking tons of locales including Hungarian. So disable
> backwards ordering in this locale definition.
>
> - Foreign accents should be sorted after the native Hungarian ones, it
> wasn't the case so far.
>
> Plus, a unittest is added which is far more extensive than any other
> locale has. It includes all the examples from the official rules of
> Hungarian orthography's corresponding sections, as well as thorough
> testing of all corner cases I could think of, created by me; and
> comments all around.
>
> In addition to fixing a(n unfortunately relatively unsignificant)
> locale, I hope that this unittest file will encourage other locale
> maintainers to create similarly extensive tests, increasing the
> quality of other locales in the long run and preventing regressions
> (such as the backward diacritics ordering) from sneaking in.
>
> Thanks a lot,
> egmont

Attachment: glibc-18934-hu-collate-v5.patch
Description: application/mbox


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]