On Jun 16, 2005, at 10:21 PM, Vivek wrote:
> Thanks for the review Fitz. I had sent this patch as a follow up of
> issue #2301. But since there is going to be a new file created, some
> further date related tests were also added.
>
> On Thu, 2005-06-16 at 23:32, Brian W. Fitzpatrick wrote:
>
>> Comments inline
>>
>>
>>> +
>>> + for i in range(0,10):
>>>
>>
>> Don't use 10, use len(time_array)
>>
> OK.
>
>>
>>
>>> + out, err = svntest.main.run_svn(None,'log', mu_path, '-r',
>>> '{' +
>>> + str(time_array[i]) + '}'
>>> + )
>>> + if err != '' and out[3] != 'Log message for revision 1.\n' :
>>>
>>
>> instead of '!=' use 'if not' ...
>>
> OK.
>
>>
>>
>>> + raise svntest.Failure
>>> +
>>>
>>>
>> +#-------------------------------------------------------------------
>> ---
>>
>>> +# The output of GNU date was not being accepted by svn commands
>>> +# Test the fix.
>>>
>>>
>>
>> Why are you using popen for this instead of the python time
>> functions?
>>
>
> This is a specific test for the GNU date/time bug. The relavent
> post is
> http://svn.haxx.se/dev/archive-2005-04/1361.shtml. This test is very
> platform specific.
Why can't just mock up the format that GNU date outputs using strftime?
-Fitz
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Jun 17 17:31:57 2005