Make the users empty when there is no passowrd file.

Make the users empty when there is no passowrd file.

Without this a default checkout that doesn't touch the trymaster
fails presubmit checks.


BUG=None
TEST=local

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=147431

[Peter Mayo, 2012-07-19 02:15:51]

This got in the way of a sync & retest another change of mine.

I wonder if M-A put a password file on the commit-queue?

[M-A Ruel, 2012-07-19 11:57:27]

https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
File scripts/master/try_job_rietveld.py (right):

https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
scripts/master/try_job_rietveld.py:68: try:
I'd prefer a 
if os.path.isfile()

and be sure to log when the file is not loaded.

[(NOT FOR CODE REVIEWS), 2012-07-19 12:34:40]

Lgtm

MA: your comment about logging made me think that it would be a good idea to
also log at line 219 when a job is rejected due to user not being allowed. It
would also be good to send an email or maybe post a message to the issue letting
the user know. What do you think? Peter: I'm not suggesting any of this as part
of this cl.

[M-A Ruel, 2012-07-19 12:48:24]

On 2012/07/19 12:34:40, rogerta1 wrote:
> Lgtm
> 
> MA: your comment about logging made me think that it would be a good idea to
> also log at line 219 when a job is rejected due to user not being allowed. It
> would also be good to send an email or maybe post a message to the issue
letting
> the user know. What do you think? Peter: I'm not suggesting any of this as
part
> of this cl.

Since this is affecting the CQ, I'll have it be committed right away. Some can
then update the style afterward.

[commit-bot: I haz the power, 2012-07-19 12:48:31]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/petermayo@chromium.org/10796027/1

[commit-bot: I haz the power, 2012-07-19 12:49:48]

Change committed as 147431

[Peter Mayo, 2012-07-19 19:12:43]

On 2012/07/19 12:48:24, Marc-Antoine Ruel wrote:
> On 2012/07/19 12:34:40, rogerta1 wrote:
> > Lgtm
> > 
> > MA: your comment about logging made me think that it would be a good idea to
> > also log at line 219 when a job is rejected due to user not being allowed.
It
> > would also be good to send an email or maybe post a message to the issue
> letting
> > the user know. What do you think? Peter: I'm not suggesting any of this as
> part
> > of this cl.
> 
> Since this is affecting the CQ, I'll have it be committed right away. Some can
> then update the style afterward.

Thx.

[Peter Mayo, 2012-07-19 19:12:51]

https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
File scripts/master/try_job_rietveld.py (right):

https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
scripts/master/try_job_rietveld.py:68: try:
On 2012/07/19 11:57:28, Marc-Antoine Ruel wrote:
> I'd prefer a 
> if os.path.isfile()
> 
> and be sure to log when the file is not loaded.

SGTM ... once running I'm not sure if the preferred behavior should be to
continue with the previous set, or not allow any until the permissions/contents
of the file are fixed.  Similarly for the web service being unavailable.

Did you want a third cook in this kitchen, or are you and Roger happy to
continue improvements from here?

[Peter Mayo, 2012-07-19 21:37:57]

On 2012/07/19 19:12:51, Peter Mayo wrote:
>
https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
> File scripts/master/try_job_rietveld.py (right):
> 
>
https://chromiumcodereview.appspot.com/10796027/diff/1/scripts/master/try_job...
> scripts/master/try_job_rietveld.py:68: try:
> On 2012/07/19 11:57:28, Marc-Antoine Ruel wrote:
> > I'd prefer a 
> > if os.path.isfile()
> > 
> > and be sure to log when the file is not loaded.
> 
> SGTM ... once running I'm not sure if the preferred behavior should be to
> continue with the previous set, or not allow any until the
permissions/contents
> of the file are fixed.  Similarly for the web service being unavailable.
> 
> Did you want a third cook in this kitchen, or are you and Roger happy to
> continue improvements from here?

2012-07-19 17:09:48-0400 [-] error in ValidUserPoller
        Traceback (most recent call last):
          File
"/usr/local/google/home/petermayo/satfix/build/third_party/twisted_10_2/twisted/internet/defer.py",
line 133, in maybeDeferred
            result = f(*args, **kw)
          File
"/usr/local/google/home/petermayo/satfix/build/scripts/master/try_job_rietveld.py",
line 57, in _poll
            d.addCallback(self._MakeSet)
          File
"/usr/local/google/home/petermayo/satfix/build/third_party/twisted_10_2/twisted/internet/defer.py",
line 297, in addCallback
            callbackKeywords=kw)
          File
"/usr/local/google/home/petermayo/satfix/build/third_party/twisted_10_2/twisted/internet/defer.py",
line 286, in addCallbacks
            self._runCallbacks()
        --- <exception caught here> ---
          File
"/usr/local/google/home/petermayo/satfix/build/third_party/twisted_10_2/twisted/internet/defer.py",
line 542, in _runCallbacks
            current.result = callback(current.result, *args, **kw)
          File
"/usr/local/google/home/petermayo/satfix/build/scripts/master/try_job_rietveld.py",
line 86, in _MakeSet
            emails = (email.strip() for email in data.splitlines())
        exceptions.AttributeError: 'frozenset' object has no attribute
'splitlines'


D'oh - wrong docstring - followup coming.

[Roger Tawa OOO till Jul 10th, 2012-07-19 22:23:11]

Bad docstring my fault, sorry about that.