From: Johannes Sixt <j.sixt@viscovery.net>
Date: Tue, 10 Aug 2010 14:12:46 GMT
Subject: Re: Black smoke from git rebase -i exec
Message-ID: <4C615E5E.3090301@viscovery.net>
In-Reply-To: <AANLkTimgRu=nRFpO+QW758SWbQ+Vs+8gtpAc4N-cNWr6@mail.gmail.com>

Am 8/10/2010 15:57, schrieb Ævar Arnfjörð Bjarmason:
> On Tue, Aug 10, 2010 at 13:37, Matthieu Moy
> <Matthieu.Moy@grenoble-inp.fr> wrote:
>> Ęvar Arnfjörš Bjarmason <avarab@gmail.com> writes:
>>
>>> There's some black smoke in pu after the git rebase -i series was
>>> applied: http://smoke.git.nix.is/app/projects/report_details/14
>>
>> Strange, I can't reproduce this on my box (tried on RHEL x86_64 and
>> Debian i686).
> 
> Hi. The issue appears to be that there's some non-POSIX code in your
> patch (but I didn't check what). The test works for me with bash, but
> fails with dash (which is the Debian testing /bin/sh).
> 
> Can you try with dash or some other non-bash POSIX shell and see if it
> fails?

The culprit is commands like these:

	FAKE_LINES="exec_echo_foo_>file1 1" \
		test_must_fail git rebase -i HEAD^ &&

You cannot apply single-command-export if the command is a shell function.
You must rewrite this as:

	(
		export FAKE_LINES="..." &&
		test_must_fail git rebase ....
	) &&

-- Hannes