Hi Jakub, On Tue, 14 Jun 2016 16:34:19 Jakub Wilk wrote: > * Stuart Prescott <stu...@debian.org>, 2016-06-15, 00:02: > >+ @property > >+ def date(self): > >+ """ a datetime object for the removal action """ > >+ try: > >+ from dateutil.parser import parse > >+ except ImportError: > >+ warnings.warn("Install the python-dateutil or python3-dateutil > >" + "package to process dates in package removals.") + > > raise > >+ return parse(self['date']) > > There's a RFC 822 date parser in Python stdlib > (email.utils.parsedate_tz), so if we jumped through a hops or two, we > could avoid this extra dependency.
good thought -- that might indeed be better. (and importing repeatedly inside a function isn't very pretty at all) > Hmm, I wonder how other parts of python-debian parse dates... I can't spot any other places, actually. The changelog parser doesn't even try, leaving them as strings instead; I'm wary about trying to do too much in this code and wondering if I should not parse the date at all and leave it as a string for the consumer to parse if that is actually wanted. > >+ self.assertEqual(r.bug, ['753912']) > >+ self.assertEqual(r.also_wnpp, ['123456']) > > Shouldn't bug numbers be integers? I wondered about that too, and then wondered about (a) trying to do too much and (b) zero-padding in a few years. The changelog parser does convert to integers, so perhaps this should too for consistency. thanks for the feedback! cheers Stuart -- Stuart Prescott http://www.nanonanonano.net/ stu...@nanonanonano.net Debian Developer http://www.debian.org/ stu...@debian.org GPG fingerprint 90E2 D2C1 AD14 6A1B 7EBB 891D BBC1 7EBB 1396 F2F7