[PATCH] test_timestamp.py changes for win32

[John Arbash Meinel]

John Arbash Meinel has voted +1.
Status is now: Approved
Comment:
The only thing I would comment about this, is that we still have a 
problem for "nearly null" timestamps, and for "before epoch".

So if you have "timestamp = 1*3600" and you have "tz = -2*3600" you will 
still have a problem with gmtime on windows.

However, we are unlikely to get many conversions that go all the way 
back to 1970, so I don't think it is strictly a problem.

Probably where we hit EPOCH the most is because "bzr diff" uses it when 
the revision is null (added or deleted files have one side as null:).

And this fixes those cases just fine.


For details, see: 
http://bundlebuggy.aaronbentley.com/request/%3Ce01316480704190106u44b92ec9n24e0f7a9ba27e1ad%40mail.gmail.com%3E