Re: Possible Mass Bug Filing: String Exceptions Removed in Python 2.6
* Scott Kitterman , 2010-06-05, 22:43: As was recently discussed on debian-python: http://lists.debian.org/debian-python/2010/05/msg00111.html String exceptions are no longer supported at all in Python 2.6. Many string exceptions has been fixed since then, but - interestingly - a few new has appeared, too. -- Jakub Wilk episoder_0.6.4-1 /usr/share/episoder/pyepisoder/plugins.py:137 ibus-xkbc_1.3.3.20100804-1 /usr/share/ibus-xkbc/setup/xkbc/xkb_config.py:325 ibus-xkbc_1.3.3.20100804-1 /usr/share/ibus-xkbc/vkb/vkb_key.py:137 salome_5.1.3-9 /usr/share/pyshared/salome/geompyDC.py:194 slicer_3.6.0~svn13936-1 /usr/lib/Slicer3/SlicerBaseGUI/Python/Slicer.py:596 sugar-physics-activity_5+dfsg-1 /usr/share/sugar/activities/Physics.activity/olpcgames/mesh.py:268 sugar-physics-activity_5+dfsg-1 /usr/share/sugar/activities/Physics.activity/olpcgames/mesh.py:304 sugar-physics-activity_5+dfsg-1 /usr/share/sugar/activities/Physics.activity/olpcgames/mesh.py:307 sugar-pippy-activity_36~dfsg-1 /usr/share/sugar/activities/Pippy.activity/groupthink/stringtree.py:134 Adam C. Powell, IV salome (U) Dominique Belhachemi slicer (U) LI Daobing ibus-xkbc (U) Debian OLPC sugar-physics-activity sugar-pippy-activity Debian Science Maintainers salome Debian Science Team slicer Luke Faraone sugar-physics-activity (U) Asias He ibus-xkbc (U) IME Packaging Team ibus-xkbc Stefan Ott episoder Python Applications Packaging Team episoder (U) Jonas Smedegaard sugar-pippy-activity (U) signature.asc Description: Digital signature
Quick Update Review
http://paste.debian.net/83034/ captures what I think are the essential policy changes for Squeeze from our discussion about the future of XS-Python-Version and Python 3. Please give it a quick review and let me know if I misrepresented the consensus. I intentionally did not include the elimination of lists of individual versions as being valid since that would break existing packages (and the release team would not let that in). Scott K signature.asc Description: This is a digitally signed message part.
Getting rid of dh_python
We have currently 4 different Python helpers in the archive[0], which is about 3 too much. dh_python has been deprecated for almost 4 years, yet there are still ~80 packages that use it. [0] There are also some packages that doesn't bytecompile their Python modules at all. Or that just ship *.pyc/*.pyo in their binary packages. Or that use existing helper in "interesting" ways. Yay for diversity. -- Jakub Wilk Loic Dachary (OuoU) poker-engine poker-network pypoker-eval Cosimo Alfarano python-xmpp (U) Ernesto Nadir Crespo Avila pyching python-uncertainities Sebastien Bacher gdeskcal mboxcheck-applet Michael Banck gamin opensync Dima Barsky python-pam Jérémy Bobbio python-clientform (U) Ivan Borzenkov ntfs-config Ben Burton python-pgsql snappea Marco Cabizza galago-gtk-python galago-python Pierre Chifflier inguma nufw prewikka Robert Collins bicyclerepair Kevin Coyner htmlgen Joost Yervante Damad wireshark (U) Julien Danjou mbot Debian GNOME Maintainers gamin (U) Debian HA Maintainers heartbeat Debian Java Maintainers eclipse-pydev Debian OpenOffice Team openoffice.org Debian Python Modules Team pexpect (U) python-biggles (U) python-pysnmp4 (U) quixote1 Debian/Ubuntu Zope Team python-clientform Sebastian Dröge gamin (U) pitivi (U) service-discovery-applet Dirk Eddelbuettel quantlib-swig Rene Engelhard openoffice.org (U) ExpPsy Maintainers pyode Alexandre Fayolle psyco pyqonsole python-unit xmldiff Bartosz Fenski pystatgrab Arnaud Fontaine python-pysnmp4 Decklin Foster py-libmpdclient pygmy Gustavo Franco directoryassistant Sylvain Le Gall libiptcdata (U) Radovan Garabík ludevit John Goerzen dictclient dictdlib forg pygopherd Debian QA Group pythoncard smart-notifier Yaroslav Halchenko pyepl (U) pyode (U) Chris Halls openoffice.org (U) Michael Hanke pyepl (U) pyode (U) Simon Horman heartbeat (U) Michael Janssen bittorrent LaMont Jones nmap Philipp Kaluza pida Matthias Klose eclipse-pydev (U) python-clientform (U) Matthias Klose python-scientific Steve Kowalik python-gendoc Chris Lawrence lincredits python-bibtex routeplanner Ana Beatriz Guerrero Lopez empy Martin Loschwitz heartbeat (U) Jan Luebbe pida (U) python-pysnmp4 (U) Maintainers of GStreamer packages pigment-python Daniel Martin pyrite-publisher Ricardo Javier Cardenes Medina fnorb Loic Minier gamin (U) pigment-python (U) pitivi service-discovery-applet (U) Sam Morris pymsnt Josselin Mouette gamin (U) tessa Alexey Nezhdanov python-xmpp Javier Fernandez-Sanguino Pen~a spkproxy Frederic Peters gnome-blog lasso Martin Pitt postgresql-8.4 Franz Pletz pympd pype Arnaud Quette pigment-python (U) Ganesan Rajagopal pexpect Balint Reczey wireshark Santiago Ruano Rincón quixote1 (U) Anibal Monsalve Salazar heartbeat (U) Frederik Schüler heartbeat (U) Riccardo Setti galago-gtk-python (U) galago-python (U) Sjoerd Simons gamin (U) Brian Sutherland python-clientform (U) python-tcpwrap NeuroDebian Team pyepl Paul van Tilburg pigment-python (U) Fabio Tranchitella psycopg2 python-clientform (U) Deepak Tripathi python-biggles Luis Uribe xtalk Michael Vogt unattended-upgrades David Watson geximon ntlmaps Ian Wienand libiptcdata Jamie Wilkinson pymad signature.asc Description: Digital signature
Re: Quick Update Review
On Friday, August 13, 2010 11:10:04 am Scott Kitterman wrote: > http://paste.debian.net/83034/ captures what I think are the essential > policy changes for Squeeze from our discussion about the future of > XS-Python-Version and Python 3. Please give it a quick review and let me > know if I > misrepresented the consensus. I intentionally did not include the > elimination of lists of individual versions as being valid since that > would break existing packages (and the release team would not let that > in). > > Scott K Thank you to those that took a look. We're going ahead now. Scott K signature.asc Description: This is a digitally signed message part.
Re: Quick Update Review
Hello, On Fri, Aug 13, 2010 at 22:47, Scott Kitterman wrote: > On Friday, August 13, 2010 11:10:04 am Scott Kitterman wrote: >> http://paste.debian.net/83034/ captures what I think are the essential >> policy changes for Squeeze from our discussion about the future of >> XS-Python-Version and Python 3. I don't think there was all that consensus over X-P-V: could you point to msgids? Also in relation to XB-P-V that's still mentioned (at least in the diff on that paste.d.n it's still mentioned), that will generate a naming difference (and yes, XB-P-V should be dropped, now). > Thank you to those that took a look. We're going ahead now. waiting only 5 hours in august/summer/holiday time is quite a bit short (it's even less than deborphan screw up!). but oh well... Regards, -- Sandro Tosi (aka morph, morpheus, matrixhasu) My website: http://matrixhasu.altervista.org/ Me at Debian: http://wiki.debian.org/SandroTosi -- To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktim37kkq8+xapzg_nbab2j5jbbdkwpk1jm0ef...@mail.gmail.com
Re: Quick Update Review
On Friday, August 13, 2010 04:58:37 pm Sandro Tosi wrote: > Hello, > > On Fri, Aug 13, 2010 at 22:47, Scott Kitterman wrote: > > On Friday, August 13, 2010 11:10:04 am Scott Kitterman wrote: > >> http://paste.debian.net/83034/ captures what I think are the essential > >> policy changes for Squeeze from our discussion about the future of > >> XS-Python-Version and Python 3. > > I don't think there was all that consensus over X-P-V: could you point > to msgids? Also in relation to XB-P-V that's still mentioned (at least > in the diff on that paste.d.n it's still mentioned), that will > generate a naming difference (and yes, XB-P-V should be dropped, now). My understanding of the rough consensus was that XB-P-V should be allowed to naturally fade away and that we would not push a transition. Even if we wanted to push a transition, it won't happen for Squeeze, so the policy change can wait. I agree there was rough consensus to drop XB-P-V and I will gladly do that in the first post-Squeeze Python policy upload. IMO the essential thing for Squeeze was getting X-P-V and X3-P-V supported. Since we are trying to get approval to go into Squeeze during the freeze, I didn't want to take on more. The last time we discussed this started in Resent-Message-ID: <3ighbuystun.a.sxc.zw6...@liszt>. It didn't generate a lot of discussion and I think the only realy push back was from you and you seemed (as I read it) to think this was good, but want more too. More will have to wait for Squeeze +1. I had hoped to have more time to work on this in July, but it didn't work out. > > Thank you to those that took a look. We're going ahead now. > > waiting only 5 hours in august/summer/holiday time is quite a bit > short (it's even less than deborphan screw up!). but oh well... I didn't do this as a formal last call and wait a few days because I don't want to give the freeze any more chance to solidify than I really have to and I don't expect this to be controversial. If there are real problems in this, I'm open to addressing them. Scott K signature.asc Description: This is a digitally signed message part.
Re: Quick Update Review
* Scott Kitterman , 2010-08-13, 17:32: IMO the essential thing for Squeeze was getting X-P-V and X3-P-V supported. Well, but python-support, the most widely used helper, doesn't support X-P-V. So what's the point of adding it to policy now? -- Jakub Wilk signature.asc Description: Digital signature
Re: Getting rid of dh_python
On 13/08/10 22:22, Jakub Wilk wrote: > We have currently 4 different Python helpers in the archive[0], which is > about 3 too much. dh_python has been deprecated for almost 4 years, yet > there are still ~80 packages that use it. Debian GNOME Maintainers gamin (U) This is fixed in our SVN, though the changes are too intrusive packaging-wise, so I won't upload them until Squeeze is released. Regards, Emilio -- To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4c65bf97.2050...@debian.org
Re: Quick Update Review
On Friday, August 13, 2010 05:42:35 pm Jakub Wilk wrote: > * Scott Kitterman , 2010-08-13, 17:32: > >IMO the essential thing for Squeeze was getting X-P-V and X3-P-V > >supported. > > Well, but python-support, the most widely used helper, doesn't support > X-P-V. So what's the point of adding it to policy now? Because the current (before today) implementation of pyversions and py3versions both used XS-P-V and silently ignored versions over/under 3 so were doing something that the community pretty clearly didn't want. Switching py3versions to use X3-P-V instead of XS-P-V was the most important thing and it would have seemed wrong to me to implement X3-P-V and not also X-P-V along with it. Bottom line is I didn't want to release with an implementation that the community had rejected (and I thought I'd get to it last month, but didn't). Scott K signature.asc Description: This is a digitally signed message part.
