From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Lehmann Subject: Re: What's cooking in git.git (Apr 2011, #02; Wed, 6) Date: Mon, 11 Apr 2011 19:55:38 +0200 Message-ID: <4DA3409A.9090206@web.de> References: <7v8vvni1t3.fsf@alter.siamese.dyndns.org> <4DA23551.4040304@web.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Junio C Hamano , git@vger.kernel.org To: =?UTF-8?B?w4Z2YXIgQXJuZmrDtnLDsCBCamFybWFzb24=?= X-From: git-owner@vger.kernel.org Mon Apr 11 19:55:48 2011 Return-path: Envelope-to: gcvg-git-2@lo.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by lo.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1Q9LL1-0008Aa-RB for gcvg-git-2@lo.gmane.org; Mon, 11 Apr 2011 19:55:48 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754845Ab1DKRzn convert rfc822-to-quoted-printable (ORCPT ); Mon, 11 Apr 2011 13:55:43 -0400 Received: from fmmailgate01.web.de ([217.72.192.221]:55683 "EHLO fmmailgate01.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754775Ab1DKRzm (ORCPT ); Mon, 11 Apr 2011 13:55:42 -0400 Received: from smtp01.web.de ( [172.20.0.243]) by fmmailgate01.web.de (Postfix) with ESMTP id 2874E18C0F126; Mon, 11 Apr 2011 19:55:41 +0200 (CEST) Received: from [93.240.123.81] (helo=[192.168.178.43]) by smtp01.web.de with asmtp (WEB.DE 4.110 #2) id 1Q9LKv-0005Zw-00; Mon, 11 Apr 2011 19:55:41 +0200 User-Agent: Mozilla/5.0 (X11; U; Linux i686; de; rv:1.9.2.15) Gecko/20110303 Lightning/1.0b2 Thunderbird/3.1.9 In-Reply-To: X-Sender: Jens.Lehmann@web.de X-Provags-ID: V01U2FsdGVkX1/6pbikpOU8YpgqKT1aDoDkz90k3nhC8cIkKg7J mwlB3xTAS35iLieziFQD85I5iBqO+lKfznNCG9FAxWY9HcqDLa /9VIW15CBCQDBWQ/epZw== Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: Am 11.04.2011 11:00, schrieb =C3=86var Arnfj=C3=B6r=C3=B0 Bjarmason: > On Mon, Apr 11, 2011 at 00:55, Jens Lehmann wro= te: >> Am 10.04.2011 20:52, schrieb =C3=86var Arnfj=C3=B6r=C3=B0 Bjarmason: >>> On Wed, Apr 6, 2011 at 22:04, Junio C Hamano wr= ote: >>> >>>> * jl/submodule-fetch-on-demand (2011-03-06) 7 commits >>>> (merged to 'next' on 2011-03-20 at a5e452d) >>>> + fetch/pull: Describe --recurse-submodule restrictions in the BU= GS section >>>> + submodule update: Don't fetch when the submodule commit is alre= ady present >>>> + fetch/pull: Don't recurse into a submodule when commits are alr= eady present >>>> + Submodules: Add 'on-demand' value for the 'fetchRecurseSubmodul= e' option >>>> + config: teach the fetch.recurseSubmodules option the 'on-demand= ' value >>>> + fetch/pull: Add the 'on-demand' value to the --recurse-submodul= es option >>>> + fetch/pull: recurse into submodules when necessary >>> >>> This series needs to be fixed to not fail tests with GETTEXT_POISON= =3DYesPlease. >> >> Maybe something like this? Junio, do you want me to squash the chang= es in >> the patch series and resend it or should I send a new commit? Ok, as that series already hit master I'll hack up a new commit. > Sorry about the short mail yesterday, I had to take care of something= =2E Same here :-) > Anyway, that looks like it'll work, but the preferred way of doing it > is like it's done at the bottom of t3700-add.sh. I.e.: >=20 > test_expect_success 'git add --dry-run --ignore-missing of > non-existing file' ' > test_must_fail git add --dry-run --ignore-missing track-this > ignored-file >actual.out 2>actual.err > ' >=20 > test_expect_success C_LOCALE_OUTPUT 'git add --dry-run > --ignore-missing of non-existing file output' ' > test_cmp expect.out actual.out && > test_cmp expect.err actual.err > ' >=20 > That is, split up each test into a test that tests the functionality > (i.e. exit codes) and one that tests the output. >=20 > That way we make sure that only things dependent on C locale output > are skipped under non-C locales. Thanks for explaining, that makes sense. Jonathan mentioned some time a= go that t5526 would benefit from testing the functionality rather than the output like it does now, so rewriting that test is on my maintenance li= st anyway. I'll have to find a locale independent way to check if the submodule paths are printed correctly, but I'll see if I can chop off t= he language specific parts of the output before doing the compare. > Sometimes that's hard, in which case it's fine to just skip the entir= e thing. That shouldn't be necessary for t5526, I'll tackle that when I find som= e time. The question is: Do we need a commit now adding the prerequisite = as a band aid fix to restore the ability to successfully run the test suit= e with GETTEXT_POISON=3DYesPlease, or do we have some time so I can rewri= te t5526 properly to not depend on the language anymore?