[Python-Dev] tests expecting but not finding errors due to bug fixes

Michael Hudson mwh at python.net
Mon Oct 27 14:26:20 EST 2003


Alex Martelli <aleaxit at yahoo.com> writes:

> Switching to the 2.3 maintenance branch (where test_bsdddb runs just fine),
> I got "make test" failures on test_re.py.  Turns out that the 2.3-branch 
> test_re.py was apparently not updated when the RE recursion bug was
> fixed -- it still expects a couple of exceptions to be raised and they don't
> get raised any more because the bugfix itself WAS backported.
>
> On general principles, in cases of this ilk, IS it all right to just backport 
> the corrected unit-test (from the 2.4 to the 2.3 branch) and commit the
> fix, or should one be more circumspect about it...?

I'd say go for it.  It sounds like just a partially missed backport
(and someone checking things in without running make test, tsk).

Cheers,
mwh

-- 
  Roll on a game of competetive offence-taking.
                                            -- Dan Sheppard, ucam.chat



More information about the Python-Dev mailing list