| « RESTful JSON | Writing High-Efficiency Large Python Systems--Lesson #3: Banish lazy imports » |
I'm categorically rejecting the 2to3 approach--for myself anyway. If you think it would help, feel free to:
Me, I'd rather just drop cherrypy/ into 3k and skip steps 1-5.
Changes I had to make so far (http://www.cherrypy.org/changeset/2029):
At the moment, I'm a bit blocked importing wsgiserver--we had a nonblocking version of makefile that subclassed the old socket._fileobject class. Looks like the whole socket implementation has changed (and much of it pushed down into C). Not looking forward to reimplementing that.
I would recommend you at least reinclude step 5. 80-90% of the items you listed can be fixed by 2to3.
I enjoyed this post. Thanks!
Ha, John, I was looking for a link to a spam site when I read your comment. Good thing I know you. :) You're always welcome.
Thanks for this interesting post! I'm really happy to know you are updating CherryPy to Python 3.0.
Where can I get the Python 3000 compatible version of Cherrypy? Thanks in advance.
Hmmm,
According to me, subclassing concrete classes which aren't meant to be subclassed is an anti-pattern IMHO (especially implementation details starting with an underscore). Apart from that, where are you at with the implementation? Python 3.0 is definitely one of the cleanest languages, and I hope most current libraries and support software (such as easy-install) will update.
Any updates on this? WSGI now has py3k support, it'd be nice to be able to use it within a framework.
Py3k support in CP is one of my goals before PyCon 09. Feel free to join the effort.