[issue2402] get rid of warnings in regrtest with -3

Benjamin Peterson <[EMAIL PROTECTED]> added the comment: Brett has been doing this. -- resolution: -> duplicate status: open -> closed ___ Python tracker <[EMAIL PROTECTED]> ___ _

[issue2402] get rid of warnings in regrtest with -3

Changes by Quentin Gallet-Gilles <[EMAIL PROTECTED]>: -- nosy: +quentin.gallet-gilles __ Tracker <[EMAIL PROTECTED]> __ ___ Python-bugs-list mai

[issue2402] get rid of warnings in regrtest with -3

Changes by Benjamin Peterson <[EMAIL PROTECTED]>: -- components: +Tests __ Tracker <[EMAIL PROTECTED]> __ ___ Python-bugs-list mailing list Uns

[issue2402] get rid of warnings in regrtest with -3

Changes by Benjamin Peterson <[EMAIL PROTECTED]>: -- type: -> behavior __ Tracker <[EMAIL PROTECTED]> __ ___ Python-bugs-list mailing list Uns

[issue2402] get rid of warnings in regrtest with -3

Benjamin Peterson <[EMAIL PROTECTED]> added the comment: I made a little decorator which silences py3k warnings. It could be useful for test.test_support. def silence_py3k(func): def decorator(*args, **kwargs): warnings.simplefilter("ignore", warnings.DeprecationWarning) func

[issue2402] get rid of warnings in regrtest with -3

Steven Bethard <[EMAIL PROTECTED]> added the comment: Fair enough. I agree that the deprecated APIs should still be tested. But there are a lot of other warnings, e.g. where the standard lib uses ``has_key`` instead of ``in``. These should be removed. The only warnings should be those that are ac

[issue2402] get rid of warnings in regrtest with -3

Benjamin Peterson <[EMAIL PROTECTED]> added the comment: No, *correct usage* of the stdlib should not result in Py3k warnings. Old APIs should still be tested, which I suppose is part of the reason that there are so many warnings from testing. -- nosy: +benjamin.peterson ___