4 ms·

Interesting. I know Scipy are one of the biggest communities of Python users. Once Django (and all it's dependencies, specifically the database backends) are p

Interesting. I know Scipy are one of the biggest communities of Python users. Once Django (and all it's dependencies, specifically the database backends) are ported over to Python 3, I think a critical mass will have been established in the move to Python 3.

mattyb 16y ago
From the Django 1.2 release notes: A roadmap for Django’s overall 2.x Python support, and eventual transition to Python 3.x, is currently being developed, and will be announced prior to the release of Django 1.3. Anybody know what this roadmap currently looks like?
- jacobolus 16y ago
  There's some info here: http://reinout.vanrees.org/weblog/2010/05/25/django-technical-design.html http://reinout.vanrees.org/weblog/2010/05/25/django-technica... Question: What are the plans for python 3? We haven’t made any plans for 2017 yet. We’re slowly dropping older releases. Probably python 2.4 sticks around in django 1.3, though. There has been work done on a python 3 port. Apparently the tests already run, though only 60% succeed. He considered 50% already a huge success, so they (=students at a canadian university, if I heard correcly) did a good job. Python 3 support will depend on an influx of new developers. (I’d guess this doesn’t seriously mean that a Py3k port won’t be done until 2017.)
illumen 16y ago
It's been heart breaking after putting so much effort myself into working on python3 stuff to see the Django people, and the other big slow python framework... zope/plone ignore it. Boo to Django.
- ebtalley 16y ago
  I can sympathize but I also see those with a need for the stability that comes with long term support. Django has reached a level of acceptance like Debian, or CentOS where intensive refactors are unacceptable. Maybe it's up to someone else to fork Django.
- jacobolus 16y ago
  One problem for Django and similar projects is that many hosted systems, etc., still only run 2.3 or 2.4, and they don’t want to cut those people off for as long as they can. As far as I know it’s rather difficult to write code that works on both 2.3/2.4 and 3.x, since the 2to3 tool assumes a recent 2.x version.... so my understanding is that they’re phasing out one python version or so each release, with the hopes of eventually getting to a 2.6+/3.x release. I don’t think they’re really “ignoring” Py3k. Then again, I’m not really an expert.
  - illumen 16y ago
    It's not that hard to support 2.4-3.1 with the same code base. It's been done, and documented by a bunch of projects now.
AndyKelley 16y ago
I don't understand why we can't have python 3 code which imports python 2 modules. If that was supported, everything would be moved over to python 3 in no time.
- superchink 16y ago
  Here's a quick explanation, but I'm sure some googling will find you more details explaining why they chose to break backward compatibility. http://arstechnica.com/open-source/news/2008/12/python-3-0-sheds-backwards-compatibility.ars http://arstechnica.com/open-source/news/2008/12/python-3-0-s...

mattyb 16y ago

From the Django 1.2 release notes: A roadmap for Django’s overall 2.x Python support, and eventual transition to Python 3.x, is currently being developed, and will be announced prior to the release of Django 1.3. Anybody know what this roadmap currently looks like?

jacobolus 16y ago
There's some info here: http://reinout.vanrees.org/weblog/2010/05/25/django-technical-design.html http://reinout.vanrees.org/weblog/2010/05/25/django-technica... Question: What are the plans for python 3? We haven’t made any plans for 2017 yet. We’re slowly dropping older releases. Probably python 2.4 sticks around in django 1.3, though. There has been work done on a python 3 port. Apparently the tests already run, though only 60% succeed. He considered 50% already a huge success, so they (=students at a canadian university, if I heard correcly) did a good job. Python 3 support will depend on an influx of new developers. (I’d guess this doesn’t seriously mean that a Py3k port won’t be done until 2017.)

illumen 16y ago

It's been heart breaking after putting so much effort myself into working on python3 stuff to see the Django people, and the other big slow python framework... zope/plone ignore it. Boo to Django.

ebtalley 16y ago
I can sympathize but I also see those with a need for the stability that comes with long term support. Django has reached a level of acceptance like Debian, or CentOS where intensive refactors are unacceptable. Maybe it's up to someone else to fork Django.
jacobolus 16y ago
One problem for Django and similar projects is that many hosted systems, etc., still only run 2.3 or 2.4, and they don’t want to cut those people off for as long as they can. As far as I know it’s rather difficult to write code that works on both 2.3/2.4 and 3.x, since the 2to3 tool assumes a recent 2.x version.... so my understanding is that they’re phasing out one python version or so each release, with the hopes of eventually getting to a 2.6+/3.x release. I don’t think they’re really “ignoring” Py3k. Then again, I’m not really an expert.
- illumen 16y ago
  It's not that hard to support 2.4-3.1 with the same code base. It's been done, and documented by a bunch of projects now.

AndyKelley 16y ago

I don't understand why we can't have python 3 code which imports python 2 modules. If that was supported, everything would be moved over to python 3 in no time.

superchink 16y ago
Here's a quick explanation, but I'm sure some googling will find you more details explaining why they chose to break backward compatibility. http://arstechnica.com/open-source/news/2008/12/python-3-0-sheds-backwards-compatibility.ars http://arstechnica.com/open-source/news/2008/12/python-3-0-s...