Re: [PATCH] i18n: use test_i18ncmp in t2020 (checkout --detach)

From: Jeff King <peff@peff.net>
Date: Apr 14, 2012, 08:22 UTC
Message-ID: <20120414082213.GC11124@sigill.intra.peff.net>
In-Reply-To: <20120414050234.GE1791@burratino>

On Sat, Apr 14, 2012 at 12:02:34AM -0500, Jonathan Nieder wrote:

Show 9 quoted lines

> Jeff King wrote:
> 
> > Yes, but this wasn't gettext-ed at all until 2012-02-02, and then it
> > didn't get merged into master until the week after. How often do you
> > run GETTEXT_POISON tests? I know I don't, and obviously Junio does not
> > include them as part of his process before pushing out master.
> 
> The theory is that it should be convenient to run them when we are
> considering a "gettextize" patch.

I was thinking when I wrote the above that it might be something worth running on every test run. But it's really not. It really only matters if you are gettextizing a string (or introducing a new string). And we should generally catch that in code review, I would think.

So it's still a good thing to run once in a while, and when there is a big gettext patch, but it's probably overkill to run it all the time.

> Maybe something like the following would make it easier for some
> people to always build with GETTEXT_POISON and run tests with
> GIT_GETTEXT_POISON only occasionally.

I found "make GETTEXT_POISON=1 test" to be sufficiently easy (and then my next "make" will un-poison.

> I'd rather have a real poison locale since this would not require any
> runtime support in the git binary, though.  Does the value of
> LC_MESSAGES have to be a valid locale?  Would something like
> en_US@poison work?

That would nice. Poisoning seems like it should be a property of the test run rather than the build. It seems like putting junk into LC_MESSAGES will just cause the default messages to be shown (seems like a sane thing to do), so I suspect you'd have to actually make a poison locale (though you could probably generate one via script without too much effort).

-Peff