Re: git 2.10.1 test regression in t4014-format-patch.sh

[Jeremy Huddleston Sequoia <jeremyhu@freedesktop.org>]

[-- Attachment #1: Type: text/plain, Size: 1833 bytes --]


> On Oct 9, 2016, at 17:18, Josh Triplett <josh@joshtriplett.org> wrote:
> 
> On October 9, 2016 5:15:22 PM PDT, Jeremy Huddleston Sequoia <jeremyhu@freedesktop.org> wrote:
>> Hey Josh,
>> 
>> Hope you're doing well.
>> 
>> I wanted to let you know that this patch of yours, which landed in git
>> 2.10.1, introduced some test failures, seen on macOS.
>> 
>> Let me know if you need any additional information to track these down.
>> 
>> Thanks,
>> Jeremy
>> 
>> 
>> not ok 65 - format-patch default signature
>> #	
>> #		git format-patch --stdout -1 | tail -n 3 >output &&
>> #		signature >expect &&
>> #		test_cmp expect output
>> #	
>> 
>> not ok 132 - format-patch --base
>> #	
>> #		git checkout side &&
>> #		git format-patch --stdout --base=HEAD~3 -1 | tail -n 7 >actual &&
>> #		echo >expected &&
>> #		echo "base-commit: $(git rev-parse HEAD~3)" >>expected &&
>> #		echo "prerequisite-patch-id: $(git show --patch HEAD~2 | git
>> patch-id --stable | awk "{print \$1}")" >>expected &&
>> #		echo "prerequisite-patch-id: $(git show --patch HEAD~1 | git
>> patch-id --stable | awk "{print \$1}")" >>expected &&
>> #		signature >> expected &&
>> #		test_cmp expected actual
>> #	
> 
> Can you run the test with the option to show the expected and actual strings?
> Did the testsuite run with the wrong git somehow?

Nope, it's the right version being tested.  The failure seems due to your git_version change not liking our formatting

$ git --version
git version 2.10.1 (Apple Git-99)

(the 'Apple Git-XX' being added because this was from a build that had Apple's patch series, including https://github.com/jeremyhu/git/commit/f99905d0752d923e5ec61e14c675a300c6d04284)

We modify DEF_VER, which confused your regex.  Simple patch inc in a separate email.

Thanks,
Jeremy


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 4465 bytes --]