This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author rhettinger
Recipients belopolsky, benjamin.peterson, georg.brandl, loewis, rhettinger, vdupras
Date 2008-02-24.01:35:00
SpamBayes Score 0.0026288023
Marked as misclassified No
Message-id <1203816902.01.0.345714874623.issue2144@psf.upfronthosting.co.za>
In-reply-to
Content
Py3.0 updte:  UserDict is going to survive into Py3.0 and will be moved 
into the collections module.  DictMixin is replaced by the abstract 
base classes for mappings.

I think the discussion here has grown far beyond the original bug 
report.  I recommend this one be closed and that you start a thread on 
python-3000 or python-dev if you want to propose altering basic APIs or 
moving the ABC code into C.

I can't say that I support your "UserDict is a kludge ..." comment.  
The class has its uses and the standard library will continue to use it 
where appropriate.  

FWIW, at one time I also thought all uses of UserDict would ultimately 
be supplanted by dict subclassing, but I found that it was somewhat 
difficult to remove in some circumstances and that its API had some 
advantages (i.e. it's easier to write subclass code like self.data
[computedkey]=computedvalue than to dispatch to dict.__setitem__(self, 
computedkey, computedvalue) -- the latter form is somewhat errorprone 
when the subclass methods become more complex).

Will look at 1367711 as you suggest.
History
Date User Action Args
2008-02-24 01:35:02rhettingersetspambayes_score: 0.0026288 -> 0.0026288023
recipients: + rhettinger, loewis, georg.brandl, belopolsky, benjamin.peterson, vdupras
2008-02-24 01:35:02rhettingersetspambayes_score: 0.0026288 -> 0.0026288
messageid: <1203816902.01.0.345714874623.issue2144@psf.upfronthosting.co.za>
2008-02-24 01:35:01rhettingerlinkissue2144 messages
2008-02-24 01:35:00rhettingercreate