From: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Date: Wed, 25 May 2011 09:36:13 GMT
Subject: Re: What's cooking in git.git (May 2011, #11; Tue, 24)
Message-ID: <BANLkTi=5koa_f_xqaGvKPrEnKMh7ziOT0g@mail.gmail.com>
In-Reply-To: <4DDCA0D2.2070604@viscovery.net>

On Wed, May 25, 2011 at 08:25, Johannes Sixt <j.sixt@viscovery.net> wrote:
> Am 5/24/2011 23:45, schrieb Junio C Hamano:
>> * ab/i18n-scripts (2011-05-21) 48 commits
>>   (merged to 'next' on 2011-05-23 at 69164a3)
>>  ...
>>
>> Rerolled.
>
> I see you promoted ab/i18n-scripts-basic to master. What are your plans
> with this topic? In its current form, git-submodule breaks in an obvious
> way on Windows, and other pitfalls may hide elsewhere.

Aren't you confusing ab/i18n-scripts-basic with ab/i18n-scripts? The
former just adds the infrastructure, the latter adds some eval_gettext
invocations that break on Windows while I haven't submitted some fixup
patches for what's now in *next*, not master.

If anything's breaking on Windows in master it would be the
t/t0201-gettext-fallbacks.sh test which uses some eval_gettext()
calls, but nothing else uses it currently.

> I don't have enough energy to help out with this issue, particularly
> because i18n is not exactly my own itch. Personally, I wouldn't mind the
> fate of the topic if there were a NO_GETTEXT switch that simply avoids all
> the problems on Windows. Is there such a switch?

There isn't and can't be without major re-organizations because
shellscripts don't have something like the C preprocessor.

But I'll just fix the Windows issue soon and submit a new series for
inclusion in next.