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

[Carlos O'Donell <carlos at redhat dot com>]

On 01/31/2017 06:16 PM, Egmont Koblinger wrote:
> Hi guys,
> 
> Could we please return to this old topic and get it addressed once and for all?
> 
> It's approximately the 8th (!!!) time I'm sending this out and/or
> pinging you. The patch has been there for more than a year. Not a
> single complaint, concern, question etc. was raised.
> 
> The patch not only addresses multiple bugs, but also backs it up with
> by far the most exhaustive, clean, modular collation unittest any
> locale has.
> 
> Mike, the last time (in last April) you said "i'm inclined to merge
> this"... but then nothing happend, I wonder why.
> 
> Please guys, either commit this patch, or give me a clean and
> actionable feedback about your concerns. Let me know if there's
> anything else you need, but pretty please, let's not delay this issue
> any further.

Egmont,

Thank you for your patience. Perhaps the best way to restart this conversation
is to cover what, if any, review, the changes have received and reference old
discussions about them.

My questions are:

* How does this compare to CLDR?

We are looking to try and harmonize between glibc and CLDR so we have common
collation across all languages that use the two collation APIs. This may be
an impossible goal to get perfect, but harmonization is just that, getting as
close as we can. I say impossible because glibc needs to be updated to actually
use the Unicode collation algorithm instead of the current code before any true
harmonization could happen.

If we had a test harness that could compare glibc to CLDR that would be great,
but we don't currently. So I'm not asking for this, but perhaps a spot test of
certain problematic values to see how CLDR compares would be good enough.

* Does the regression test pass?

* What kind of consequences might this have on existing programs?

* Can you find a Hungarian speaker to review and validate your changes?

I try to get secondary review from a native speaker for any Spanish work that
I do, that way at least it has had some peer review.

-- 
Cheers,
Carlos.