Fix test failures on Ubuntu/Trusty.

Fix test failures on Ubuntu/Trusty.

I ran into several failures running the presubmit checks
on my Ubuntu/Trusty laptop and it looks like a some of
the tests were broken by recent versions of svn and git.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=293341

[Sam Clegg, 2014-12-10 21:25:37]

sbc@chromium.org changed reviewers:
+ maruel@chromium.org

[Sam Clegg, 2014-12-10 21:25:39]

Am I missing something here?  Should these tests not already run on Trusty?

https://codereview.chromium.org/795723003/diff/1/tests/checkout_test.py
File tests/checkout_test.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/checkout_test.py#newco...
tests/checkout_test.py:614: # checksum seems to vary with svn version so we
can't check it
New versions of svn also seem to include a lot more info here.

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py
File tests/gclient_smoketest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py#n...
tests/gclient_smoketest.py:1578: self.assertEquals(3, len(out[0]))
I'm really not sure why the number of elements changed from 2 to 3 here.

The 3 elements array I have looks like this:


[['running', 'svn update --revision BASE --ignore-externals',
'/tmp/trialy3bcnW/__main__.GClientSmokeFromCheckout.testRevertAndStatus/foo/bar'],
"Updating '.':", 'At revision 1.']

Not sure what it is supposed to look like.

The use of checkString() here looks like its a bug right?

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py
File tests/scm_unittest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:512: #self._capture(['add', join('new_dir', 'subdir',
'newfile')])
I was getting svn errors here, presumably because the first 'add' operation is
recursive (at least in my version of svn).

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:530: #self.assertEquals('', self._capture(['status']))
I couldn't figure this one out.

[M-A Ruel, 2014-12-11 01:22:47]

lgtm, thanks!

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py
File tests/gclient_smoketest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py#n...
tests/gclient_smoketest.py:1578: self.assertEquals(3, len(out[0]))
On 2014/12/10 21:25:38, Sam Clegg wrote:
> I'm really not sure why the number of elements changed from 2 to 3 here.
> 
> The 3 elements array I have looks like this:
> 
> 
> [['running', 'svn update --revision BASE --ignore-externals',
>
'/tmp/trialy3bcnW/__main__.GClientSmokeFromCheckout.testRevertAndStatus/foo/bar'],
> "Updating '.':", 'At revision 1.']
> 
> Not sure what it is supposed to look like.
> 
> The use of checkString() here looks like its a bug right?

Bah, this could be just removed, it's not going to be changed significantly in
the future. And yes it looks like a bug.

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py
File tests/scm_unittest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:512: #self._capture(['add', join('new_dir', 'subdir',
'newfile')])
On 2014/12/10 21:25:38, Sam Clegg wrote:
> I was getting svn errors here, presumably because the first 'add' operation is
> recursive (at least in my version of svn).

Ok, it wasn't previously.

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:530: #self.assertEquals('', self._capture(['status']))
On 2014/12/10 21:25:38, Sam Clegg wrote:
> I couldn't figure this one out.

np

[Sam Clegg, 2014-12-11 01:28:45]

So I guess others are not running the presubmits then? Or ignoring the errors?

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py
File tests/gclient_smoketest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py#n...
tests/gclient_smoketest.py:1578: self.assertEquals(3, len(out[0]))
On 2014/12/11 01:22:46, M-A Ruel wrote:
> On 2014/12/10 21:25:38, Sam Clegg wrote:
> > I'm really not sure why the number of elements changed from 2 to 3 here.
> > 
> > The 3 elements array I have looks like this:
> > 
> > 
> > [['running', 'svn update --revision BASE --ignore-externals',
> >
>
'/tmp/trialy3bcnW/__main__.GClientSmokeFromCheckout.testRevertAndStatus/foo/bar'],
> > "Updating '.':", 'At revision 1.']
> > 
> > Not sure what it is supposed to look like.
> > 
> > The use of checkString() here looks like its a bug right?
> 
> Bah, this could be just removed, it's not going to be changed significantly in
> the future. And yes it looks like a bug.

Should I remove this test then?

I could find a precise machine to test this change on if you think the test is
useful.

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py
File tests/scm_unittest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:530: #self.assertEquals('', self._capture(['status']))
On 2014/12/11 01:22:46, M-A Ruel wrote:
> On 2014/12/10 21:25:38, Sam Clegg wrote:
> > I couldn't figure this one out.
> 
> np

Should I disable this test then?  Or delete it?  Maybe I'll see it passes on
precise and disable it with a comment if it does.

[M-A Ruel, 2014-12-11 01:57:00]

On 2014/12/11 01:28:45, Sam Clegg wrote:
> So I guess others are not running the presubmits then? Or ignoring the errors?

I'm on precise and the CQ is on precise, that's probably why.

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py
File tests/gclient_smoketest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py#n...
tests/gclient_smoketest.py:1578: self.assertEquals(3, len(out[0]))
On 2014/12/11 01:28:45, Sam Clegg wrote:
> On 2014/12/11 01:22:46, M-A Ruel wrote:
> > On 2014/12/10 21:25:38, Sam Clegg wrote:
> > > I'm really not sure why the number of elements changed from 2 to 3 here.
> > > 
> > > The 3 elements array I have looks like this:
> > > 
> > > 
> > > [['running', 'svn update --revision BASE --ignore-externals',
> > >
> >
>
'/tmp/trialy3bcnW/__main__.GClientSmokeFromCheckout.testRevertAndStatus/foo/bar'],
> > > "Updating '.':", 'At revision 1.']
> > > 
> > > Not sure what it is supposed to look like.
> > > 
> > > The use of checkString() here looks like its a bug right?
> > 
> > Bah, this could be just removed, it's not going to be changed significantly
in
> > the future. And yes it looks like a bug.
> 
> Should I remove this test then?
> 
> I could find a precise machine to test this change on if you think the test is
> useful.

No need.

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py
File tests/scm_unittest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:530: #self.assertEquals('', self._capture(['status']))
On 2014/12/11 01:28:45, Sam Clegg wrote:
> On 2014/12/11 01:22:46, M-A Ruel wrote:
> > On 2014/12/10 21:25:38, Sam Clegg wrote:
> > > I couldn't figure this one out.
> > 
> > np
> 
> Should I disable this test then?  Or delete it?  Maybe I'll see it passes on
> precise and disable it with a comment if it does.

Delete.

[Sam Clegg, 2014-12-11 02:24:25]

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py
File tests/gclient_smoketest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/gclient_smoketest.py#n...
tests/gclient_smoketest.py:1578: self.assertEquals(3, len(out[0]))
On 2014/12/11 01:57:00, M-A Ruel wrote:
> On 2014/12/11 01:28:45, Sam Clegg wrote:
> > On 2014/12/11 01:22:46, M-A Ruel wrote:
> > > On 2014/12/10 21:25:38, Sam Clegg wrote:
> > > > I'm really not sure why the number of elements changed from 2 to 3 here.
> > > > 
> > > > The 3 elements array I have looks like this:
> > > > 
> > > > 
> > > > [['running', 'svn update --revision BASE --ignore-externals',
> > > >
> > >
> >
>
'/tmp/trialy3bcnW/__main__.GClientSmokeFromCheckout.testRevertAndStatus/foo/bar'],
> > > > "Updating '.':", 'At revision 1.']
> > > > 
> > > > Not sure what it is supposed to look like.
> > > > 
> > > > The use of checkString() here looks like its a bug right?
> > > 
> > > Bah, this could be just removed, it's not going to be changed
significantly
> in
> > > the future. And yes it looks like a bug.
> > 
> > Should I remove this test then?
> > 
> > I could find a precise machine to test this change on if you think the test
is
> > useful.
> 
> No need.

I looked into it and added comment.  Can remove this test if you'd rather.

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py
File tests/scm_unittest.py (right):

https://codereview.chromium.org/795723003/diff/1/tests/scm_unittest.py#newcod...
tests/scm_unittest.py:530: #self.assertEquals('', self._capture(['status']))
On 2014/12/11 01:57:00, M-A Ruel wrote:
> On 2014/12/11 01:28:45, Sam Clegg wrote:
> > On 2014/12/11 01:22:46, M-A Ruel wrote:
> > > On 2014/12/10 21:25:38, Sam Clegg wrote:
> > > > I couldn't figure this one out.
> > > 
> > > np
> > 
> > Should I disable this test then?  Or delete it?  Maybe I'll see it passes on
> > precise and disable it with a comment if it does.
> 
> Delete.

Done.

[Sam Clegg, 2014-12-11 03:03:57]

The CQ bit was checked by sbc@chromium.org

[commit-bot: I haz the power, 2014-12-11 03:04:40]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/795723003/20001

[commit-bot: I haz the power, 2014-12-11 03:06:50]

Committed patchset #2 (id:20001) as
http://src.chromium.org/viewvc/chrome?view=rev&revision=293341