>>>>> "martin" == Martin Pool <email address hidden> writes:
>> Trivial but worth the exposure.
>>
>> There should be lighter solutions to avoid that kind of easy traps...
martin> Would it be better for branchbuilder to default to
martin> utc rather than to the local timezone?
May be, but that will make it different from the rest of the code
base...
What I had in mind was more about having easier ways to create
commits for test purposes, having to specify even the timestamp
shouldn't be necessary or specified only once or something.
This is rarely needed except for log (and annotate here) and I'd
prefer to address it locally there. I started something along
these lines in blackbox.test_log.py: TestLog.commit_options(). It
makes the test clearer where used, but this still needs more
love...
>>>>> "martin" == Martin Pool <email address hidden> writes:
>> Trivial but worth the exposure.
>>
>> There should be lighter solutions to avoid that kind of easy traps...
martin> Would it be better for branchbuilder to default to
martin> utc rather than to the local timezone?
May be, but that will make it different from the rest of the code
base...
What I had in mind was more about having easier ways to create
commits for test purposes, having to specify even the timestamp
shouldn't be necessary or specified only once or something.
This is rarely needed except for log (and annotate here) and I'd test_log. py: TestLog. commit_ options( ). It
prefer to address it locally there. I started something along
these lines in blackbox.
makes the test clearer where used, but this still needs more
love...