[Python-Dev] bug-fixes in cnri-16-start branch

Guido van Rossum guido@beopen.com
Mon, 07 Aug 2000 19:21:43 -0500


> I have a question on the right procedure for fixing a simple
> bug in the 1.6 release branch.
> 
> Bug #111162 appeared because the tests for math.rint() are
> already contained in the cnri-16-start revision of test_math.py
> while the "try: ... except AttributeError: ..." construct which
> was checked in shortly after was not.
> 
> Now the correct bugfix is already known (and has been
> applied to the main branch). I have updated the test_math.py
> file in my working version with "-r cnri-16-start" and
> made the changes.
> 
> Now I probably should just commit, close the patch
> (with an appropriate follow-up) and be happy.

That would work, except that I prefer to remove math.rint altogether,
as explained by Tim Peters.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)