Disable tests that break because of git 2.4

Disable tests that break because of git 2.4

BUG=487172
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=295375

[Adrian Kuegel, 2015-05-20 14:28:35]

akuegel@chromium.org changed reviewers:
+ iannucci@chromium.org

[Adrian Kuegel, 2015-05-20 14:28:36]

Robbie, can you please review this CL?

https://codereview.chromium.org/1150663002/diff/1/tests/git_number_test.py
File tests/git_number_test.py (right):

https://codereview.chromium.org/1150663002/diff/1/tests/git_number_test.py#ne...
tests/git_number_test.py:86: ))
Looks like there was no newline at the end of this file before.

This change was actually not intended; I only edited this file because I
temporarily removed the '3.7' parameter to be able to run presubmits locally
without needing to install the python-coverage package. I can revert that change
if you prefer that.

[Adrian Kuegel, 2015-05-22 09:19:25]

Robbie, friendly ping?

[iannucci, 2015-05-22 09:59:56]

lgtm, though it seems pretty clear to me that the branches() function just isn't
WAI under 2.4. Did you get a chance to see what the failure is?

https://codereview.chromium.org/1150663002/diff/20001/tests/git_common_test.py
File tests/git_common_test.py (right):

https://codereview.chromium.org/1150663002/diff/20001/tests/git_common_test.p...
tests/git_common_test.py:236: self.assertEqual(self.repo.run(set,
self.gc.branches()),
looks like the branches() function is broken then? Did you see what's wrong with
it by any chance?

[Adrian Kuegel, 2015-05-22 10:08:14]

https://codereview.chromium.org/1150663002/diff/20001/tests/git_common_test.py
File tests/git_common_test.py (right):

https://codereview.chromium.org/1150663002/diff/20001/tests/git_common_test.p...
tests/git_common_test.py:236: self.assertEqual(self.repo.run(set,
self.gc.branches()),
On 2015/05/22 09:59:56, iannucci wrote:
> looks like the branches() function is broken then? Did you see what's wrong
with
> it by any chance?

Here is the error message:

FAIL: testBranches (__main__.GitReadOnlyFunctionsTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "tests/git_common_test.py", line 235, in testBranches
    {'master', 'branch_D', 'root_A'})
AssertionError: Items in the first set but not the second:
'696a838)'

Not sure what is going on here, and I don't have git 2.4 locally installed.

[Adrian Kuegel, 2015-05-22 10:08:35]

The CQ bit was checked by akuegel@chromium.org

[commit-bot: I haz the power, 2015-05-22 10:08:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1150663002/20001

[commit-bot: I haz the power, 2015-05-22 10:12:07]

Committed patchset #2 (id:20001) as
http://src.chromium.org/viewvc/chrome?view=rev&revision=295375

[iannucci, 2015-05-22 10:40:37]

On 2015/05/22 10:12:07, I haz the power (commit-bot) wrote:
> Committed patchset #2 (id:20001) as
> http://src.chromium.org/viewvc/chrome?view=rev&revision=295375

I suspect that the output of the git-branch command changed. This is the
implementation of the branches() function:

https://chromium.googlesource.com/chromium/tools/depot_tools/+/a1bf72dd1e6128...

It looks like 2.4 changed the wording of the git-branch output:

https://github.com/git/git/blob/v2.4.0/Documentation/RelNotes/2.4.0.txt

I suspect that you can fix the bug by adding the string "* (detached at " to the
NO_BRANCH tuple, but I don't have a machine with 2.4 handy to verify.

[iannucci, 2015-05-22 10:41:58]

On 2015/05/22 10:40:37, iannucci wrote:
> On 2015/05/22 10:12:07, I haz the power (commit-bot) wrote:
> > Committed patchset #2 (id:20001) as
> > http://src.chromium.org/viewvc/chrome?view=rev&revision=295375
> 
> I suspect that the output of the git-branch command changed. This is the
> implementation of the branches() function:
> 
>
https://chromium.googlesource.com/chromium/tools/depot_tools/+/a1bf72dd1e6128...
> 
> It looks like 2.4 changed the wording of the git-branch output:
> 
> https://github.com/git/git/blob/v2.4.0/Documentation/RelNotes/2.4.0.txt
> 
> I suspect that you can fix the bug by adding the string "* (detached at " to
the
> NO_BRANCH tuple, but I don't have a machine with 2.4 handy to verify.

or, just remove the " from" part of the string that's already there

[Adrian Kuegel, 2015-05-22 11:07:58]

On 2015/05/22 10:41:58, iannucci wrote:
> On 2015/05/22 10:40:37, iannucci wrote:
> > On 2015/05/22 10:12:07, I haz the power (commit-bot) wrote:
> > > Committed patchset #2 (id:20001) as
> > > http://src.chromium.org/viewvc/chrome?view=rev&revision=295375
> > 
> > I suspect that the output of the git-branch command changed. This is the
> > implementation of the branches() function:
> > 
> >
>
https://chromium.googlesource.com/chromium/tools/depot_tools/+/a1bf72dd1e6128...
> > 
> > It looks like 2.4 changed the wording of the git-branch output:
> > 
> > https://github.com/git/git/blob/v2.4.0/Documentation/RelNotes/2.4.0.txt
> > 
> > I suspect that you can fix the bug by adding the string "* (detached at " to
> the
> > NO_BRANCH tuple, but I don't have a machine with 2.4 handy to verify.
> 
> or, just remove the " from" part of the string that's already there

Ok, I will try that on the CQ host.