Create a PRESUBMIT which verifies that 9999 and stable are in sync

Create a PRESUBMIT which verifies that 9999 and stable are in sync

BUG=4993
TEST=Verified that there is no error when in sync.
Verified that an error is reported when 9999 does not exist.
Verified that an error is reported when not in sync.

Change-Id: I734ba8bd4e83cc3eebc514bdf9ec7c73d9cbb65b

[Chris Masone, 2010-07-26 22:23:33]

http://codereview.chromium.org/3015033/diff/1/2
File PRESUBMIT.py (right):

http://codereview.chromium.org/3015033/diff/1/2#newcode34
PRESUBMIT.py:34: devebuild_path =  os.path.join(dir, ebuild + '-9999.ebuild')
will this work if I have chromeos-login-0.5.0.ebuild and
chromeos-login-9999.ebuild?

[Mandeep Singh Baines, 2010-07-26 22:35:21]

cmasone@chromium.org (cmasone@chromium.org) wrote:
>
> http://codereview.chromium.org/3015033/diff/1/2
> File PRESUBMIT.py (right):
>
> http://codereview.chromium.org/3015033/diff/1/2#newcode34
> PRESUBMIT.py:34: devebuild_path =  os.path.join(dir, ebuild +
> '-9999.ebuild')
> will this work if I have chromeos-login-0.5.0.ebuild and
> chromeos-login-9999.ebuild?
>

Yes. It will. I tested with the kernel ebuild.

> http://codereview.chromium.org/3015033/show

[Chris Masone, 2010-07-26 22:37:47]

How?  I don't understand what strips off the other version number?

On Mon, Jul 26, 2010 at 3:35 PM, Mandeep Singh Baines <msb@chromium.org> wrote:
> cmasone@chromium.org (cmasone@chromium.org) wrote:
>>
>> http://codereview.chromium.org/3015033/diff/1/2
>> File PRESUBMIT.py (right):
>>
>> http://codereview.chromium.org/3015033/diff/1/2#newcode34
>> PRESUBMIT.py:34: devebuild_path =  os.path.join(dir, ebuild +
>> '-9999.ebuild')
>> will this work if I have chromeos-login-0.5.0.ebuild and
>> chromeos-login-9999.ebuild?
>>
>
> Yes. It will. I tested with the kernel ebuild.
>
>> http://codereview.chromium.org/3015033/show
>

[Mandeep Singh Baines, 2010-07-26 23:01:10]

Chris Masone (cmasone@chromium.org) wrote:
> How?  I don't understand what strips off the other version number?
> 

f.LocalPath() = chromeos-base/chromeos-login/chromeos-login-0.5.0.ebuild
dir = chromeos-base/chromeos-login
ebuild = chromeos-login
devebuild_path = chromeos-base/chromeos-login/ + 'chromeos-login' +
'-9999.ebuild'

> On Mon, Jul 26, 2010 at 3:35 PM, Mandeep Singh Baines <msb@chromium.org>
wrote:
> > cmasone@chromium.org (cmasone@chromium.org) wrote:
> >>
> >> http://codereview.chromium.org/3015033/diff/1/2
> >> File PRESUBMIT.py (right):
> >>
> >> http://codereview.chromium.org/3015033/diff/1/2#newcode34
> >> PRESUBMIT.py:34: devebuild_path = ?os.path.join(dir, ebuild +
> >> '-9999.ebuild')
> >> will this work if I have chromeos-login-0.5.0.ebuild and
> >> chromeos-login-9999.ebuild?
> >>
> >
> > Yes. It will. I tested with the kernel ebuild.
> >
> >> http://codereview.chromium.org/3015033/show
> >

[Chris Masone, 2010-07-26 23:06:30]

cool

lgtm

On Mon, Jul 26, 2010 at 3:59 PM, Mandeep Singh Baines <msb@chromium.org> wrote:
> Chris Masone (cmasone@chromium.org) wrote:
>> How?  I don't understand what strips off the other version number?
>>
>
> f.LocalPath() = chromeos-base/chromeos-login/chromeos-login-0.5.0.ebuild
> dir = chromeos-base/chromeos-login
> ebuild = chromeos-login
> devebuild_path = chromeos-base/chromeos-login/ + 'chromeos-login' +
'-9999.ebuild'
>
>> On Mon, Jul 26, 2010 at 3:35 PM, Mandeep Singh Baines <msb@chromium.org>
wrote:
>> > cmasone@chromium.org (cmasone@chromium.org) wrote:
>> >>
>> >> http://codereview.chromium.org/3015033/diff/1/2
>> >> File PRESUBMIT.py (right):
>> >>
>> >> http://codereview.chromium.org/3015033/diff/1/2#newcode34
>> >> PRESUBMIT.py:34: devebuild_path = ?os.path.join(dir, ebuild +
>> >> '-9999.ebuild')
>> >> will this work if I have chromeos-login-0.5.0.ebuild and
>> >> chromeos-login-9999.ebuild?
>> >>
>> >
>> > Yes. It will. I tested with the kernel ebuild.
>> >
>> >> http://codereview.chromium.org/3015033/show
>> >
>

[sosa, 2010-07-27 00:30:38]

I know this is late but can you answer this?

http://codereview.chromium.org/3015033/diff/1/2
File PRESUBMIT.py (right):

http://codereview.chromium.org/3015033/diff/1/2#newcode19
PRESUBMIT.py:19: return ebuild_contents.count('inherit cros-workon') > 0
What happens if an ebuild contains "inherit something_else cros-workon"?

[Mandeep Singh Baines, 2010-07-27 01:13:46]

sosa@chromium.org (sosa@chromium.org) wrote:
> I know this is late but can you answer this?
>

Sorry was P0 so wanted to get in ASAP.

>
> http://codereview.chromium.org/3015033/diff/1/2
> File PRESUBMIT.py (right):
>
> http://codereview.chromium.org/3015033/diff/1/2#newcode19
> PRESUBMIT.py:19: return ebuild_contents.count('inherit cros-workon') > 0
> What happens if an ebuild contains "inherit something_else cros-workon"?
>

Good point. I mistakenly confused our python coding convention with our
ebuild one and thought that wasn't allowed. But that's not correct.

I'll send a follow-up CL to make the search more permissive.

> http://codereview.chromium.org/3015033/show