> On March 3, 2015, 3:21 p.m., Zameer Manji wrote:
> > src/test/python/apache/aurora/client/cli/test_supdate.py, line 211
> > <https://reviews.apache.org/r/31710/diff/1/?file=883801#file883801line211>
> >
> >     For the calls like this you can do 
> > `mock_api.pause_job_update.assert_called_once_with(self.TEST_JOBKEY)` 
> > instead of comparing against an array of call objects.
> 
> Bill Farner wrote:
>     In case the diff didn't make it clear, i think we should settle on 
> `mock_calls` for uniformity.  Additionally, i think the error messages from 
> asserting on `mock_calls` is better:
>     
>     ```
>     _mock_self = <MagicMock name='ctime' id='140532173693328'>, args = (1,)
>     kwargs = {}, self = <MagicMock name='ctime' id='140532173693328'>
>     msg = 'Expected to be called once. Called 9 times.'
>     
>     >   ???
>     E   AssertionError: Expected to be called once. Called 9 times.
>     ```
>     
>     vs
>     
>     ```
>     >       assert self.mock_ctime.mock_calls == [call(1)]
>     E       AssertionError: assert [call(1),\n ca...(8),\n call(9)] == 
> [call(1)]
>     E         Left contains more items, first extra item: call(2)
>     E         Full diff:
>     E         - [call(1),
>     E         ?         ^
>     E         + [call(1)]
>     E         ?         ^
>     E         -  call(2),
>     E         -  call(3),
>     E         -  call(4),
>     E         -  call(5),
>     E         -  call(6),
>     E         -  call(7),
>     E         -  call(8),
>     E         -  call(9)]
>     ```
>     
>     Moreover, `assert x.mock_calls == [...]` is not vulnerable to typos.  In 
> this test, the first typoed line passes, but the second fails.
>     
>     ```
>           self.mock_ctime.assret_called_once_with(1)
>           assert self.mock_ctime.mock_clls == [call(n) for n in range(1, 10)]
>     ```

Your reasoning is sound. I would like to point out the nice error messages come 
from using `assert` statements under a py.test runner.


- Zameer


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/31710/#review75083
-----------------------------------------------------------


On March 3, 2015, 3:16 p.m., Bill Farner wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/31710/
> -----------------------------------------------------------
> 
> (Updated March 3, 2015, 3:16 p.m.)
> 
> 
> Review request for Aurora and Maxim Khutornenko.
> 
> 
> Bugs: AURORA-1165 and AURORA-1166
>     https://issues.apache.org/jira/browse/AURORA-1165
>     https://issues.apache.org/jira/browse/AURORA-1166
> 
> 
> Repository: aurora
> 
> 
> Description
> -------
> 
> Discovered these while working on AURORA-1157
> 
> 
> Diffs
> -----
> 
>   src/main/python/apache/aurora/client/cli/update.py 
> ee8146be0cb6e3f3139a5491c883864950f245bd 
>   src/test/python/apache/aurora/client/cli/test_supdate.py 
> 1173650998d397a68a7fabf38201fb9feebf2977 
> 
> Diff: https://reviews.apache.org/r/31710/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Bill Farner
> 
>

Reply via email to