[TIP] unittestgui (was: Green and Red bars)

[Michael Foord]

On 29/10/2010 23:01, Mark Roddy wrote:
> On Fri, Oct 29, 2010 at 10:59 PM, Michael Foord
> <fuzzyman at voidspace.org.uk>  wrote:
>> On 29/10/2010 22:51, Mark Roddy wrote:
>>> On Fri, Oct 29, 2010 at 9:48 PM, Michael Foord
>>> <fuzzyman at voidspace.org.uk>    wrote:
>>>> On 27/10/2010 21:33, Mark Roddy wrote:
>>>>> [snip..]
>>>>> I've added the last two of Michael's feature requests (fix window
>>>>> resizing issue and ability to set advanced discovery options).
>>>>> They're in my bitbucket repo now:
>>>>> https://bitbucket.org/markroddy/unittestgui
>>>>>
>>>>> Questions/comments/feedback are welcome and appreciated.
>>>> Hey Mark,
>>>>
>>>> I'm currently away but due back home soon. I couldn't get the new version
>>>> of
>>>> unittestgui to run the mock tests on this laptop. The error I'm getting
>>>> is:
>>>>
>>>>     raise ImportError('Start directory is not importable: %r' % start_dir)
>>>> ImportError: Start directory is not importable:
>>>> '/Users/michael/Dev/Packages/mock'
>>>>
>>>> Is the top level directory defaulting to the current working directory?
>>>> It
>>>> should default to the same directory as the start directory *unless* a
>>>> start
>>>> directory has been selected. I can look at this more when I return.
>>>>
>>>> It would be nice to be able to pick a top level directory with a
>>>> directory
>>>> picker too (but just put it in the text box - being able to edit the
>>>> directory is useful).
>>>>
>>>> All the best,
>>>>
>>>> Michael
>>>>
>>>>> -Mark
>>>> --
>>>> http://www.voidspace.org.uk/
>>>>
>>>>
>>> Michael,
>>> I was passing in an empty string for the top_level_directory when one
>>> has not been specified instead of a None value which the discover()
>>> function explicitly checks for.  I've pushed an update to make sure
>>> None is passed in in this case and also added a directory dialog for
>>> selecting the top level directory.  Please let me know if you are
>>> still experiencing the bug.  I was able to reproduce it on my end, and
>>> it does not appear to still be an issue.
>>>
>> Now it works great the first time, but if I then choose a *different*
>> directory and click start it fails with the same error as before. Some state
>> is obviously hanging around between test runs.
>>
>> Thanks for the quick response!
>>
>> All the best,
>>
>> Michael
>>
>>> -Mark
>>
> When you select a different directory to discover from or a different
> top level directory?
Select a different directory to discover from.

Michael

> -Mark


-- 
http://www.voidspace.org.uk/