Re: [Duplicity-talk] Failure testing duplicity 3.0.0

[Kenneth Loafman]

Here's the patch.

...Ken

On Mon, Sep 2, 2024 at 12:09 PM Kenneth Loafman <kenneth@loafman.com> wrote:

OK, it's not a timing issue, but a test case issue.  I'll fix it and get back to you with a patch for 3.02.

Seems Python or pytest sometimes remembers stuff from previous tests.  dup_time.curtimestr was set in test_basic and since that ran before test_suffix all was well.  

Interesting that this has not cropped up earlier since that file has not changed in a long time.

...Ken

On Mon, Sep 2, 2024 at 11:28 AM Kenneth Loafman <kenneth@loafman.com> wrote:

Scott,

Exactly how do you run your tests?  What command line?

I do know that this is not a timing issue.  test_file_naming does not actually create files, just filenames.  The tests that do backup, etc.. have a 1 second delay in between, so no timing issue.

Are you perhaps running this under Docker?  We've had timing issues there, but going to the 1 second delay solved that problem years ago.

...Ken

On Mon, Sep 2, 2024 at 7:23 AM Scott Hannahs via Duplicity-talk <duplicity-talk@nongnu.org> wrote:

Ede,
It might be.  But it is intermittent!  The last 3 tests built without errors.

What is that test that does an assertion "assert dup_time.curtimestr”
It is trying to do a get of a gzipped incremental file name from the manifest, but with no volume number?

Is the machine too fast during testing that it is creating files with duplicate date/time stamps?

-Scott


> On Sep 2, 2024, at 06:36, edgar.soldin--- via Duplicity-talk <duplicity-talk@nongnu.org> wrote:
>
> hey Scott,
>
> looks like the brew build runs fine https://formulae.brew.sh/formula/duplicity . something with your specific setup?
>
> can you give more details os version etc. to the box you are building on? unfortunately i can't run latest macOS versions virtually so far, so it' difficult to try to reproduce.
>
> sunny regards ..ede
>
> On 02.09.2024 03:46, Scott Hannahs via Duplicity-talk wrote:
>> First, I get a lot of errors trying to build 3.0.2, so I will get to them next.
>>
>> However I get this failure with 3.0.0.  It seems that there is an issue with the time of the files?  Are they being built too quickly and the timer does not have time to increment?  This doesn’t happen on every build.
>>
>>
>> -Scott
>>
>> testing/unit/test_file_naming.py .F.......                               [ 92%]
>> ….
>>
>> =================================== FAILURES ===================================
>> ____________________________ FileNaming.test_suffix ____________________________
>>
>> self = <testing.unit.test_file_naming.FileNaming testMethod=test_suffix>
>>
>>     def test_suffix(self):
>>         """Test suffix (encrypt/compressed) encoding and generation"""
>>         file_naming.prepare_regex(force=True)
>>>       filename = file_naming.get("inc", manifest=1, gzipped=1)
>>
>> self       = <testing.unit.test_file_naming.FileNaming testMethod=test_suffix>
>>
>> /opt/sw/src/fink.build/duplicity-3.0.0-1/duplicity-rel.3.0.0/testing/unit/test_file_naming.py:102:
>> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>>
>> type = 'inc', volume_number = None, manifest = 1, encrypted = False, gzipped = 1
>> partial = False
>>
>>     def get(
>>         type,
>>         volume_number=None,
>>         manifest=False,  # pylint: disable=redefined-builtin
>>         encrypted=False,
>>         gzipped=False,
>>         partial=False,
>>     ):
>>         """
>>         Return duplicity filename of specified type
>>
>>         type can be "full", "inc", "full-sig", "new-sig", "full-stat", "inc-stat". volume_number
>>         can be given with the full and inc types.  If manifest is true the
>>         filename is of a full or inc manifest file.
>>         """
>>>       assert dup_time.curtimestr
>> E       AssertionError
>>
>> encrypted  = False
>> gzipped    = 1
>> manifest   = 1
>> partial    = False
>> type       = 'inc'
>> volume_number = None
>>
>> /opt/sw/src/fink.build/duplicity-3.0.0-1/duplicity-rel.3.0.0/duplicity/file_naming.py:253: AssertionError
>>
>>
>> _______________________________________________
>> Duplicity-talk mailing list
>> Duplicity-talk@nongnu.org
>> https://lists.nongnu.org/mailman/listinfo/duplicity-talk
>
> _______________________________________________
> Duplicity-talk mailing list
> Duplicity-talk@nongnu.org
> https://lists.nongnu.org/mailman/listinfo/duplicity-talk


_______________________________________________
Duplicity-talk mailing list
Duplicity-talk@nongnu.org
https://lists.nongnu.org/mailman/listinfo/duplicity-talk

chg_test__Fix_unit_test_file_naming_py_.patch
Description: Binary data