[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Is there any doc for python SWIG bindings?

From: Barry Scott <barry_at_barrys-emacs.org>
Date: 2005-02-21 00:19:02 CET

>> I may be somewhat biased here, but I can't really make out any sense
>> in
>> 'alternative' binding implementations. If the 'originals' are buggy,
>> try to
>> fix it. If they don't expose all functionality, add it. If they are
>> clumsy to
>> use, add a wrapper.
>
> I agree with the sentiment, though I can take a pretty good guess at
> why pysvn exists as it is: SWIG is nastily complex to learn.
>
> PySVN wraps much less of the API than the SWIG-Python bindings do, but
> does it in a higher level (and documented) way - it's all about
> tradeoffs, really.
>
> Max.

Spot on Max.

 From the pysvn FAQ

3. What's wrong with the subversion python bindings?
The subversion python binds have a number of problems:

     * they are not object oriented or python like
     * they are low level, about 650 functions and 70 structs
     * users end up read the Subverison C header files to learn usage

4. Why don't you use SWIG to wrap an svn API?
Using PyCXX a python like interface is easy to implement. To achieve
the same interface with SWIG is a lot harder.

I will be adding the svnlook functionality next. I've not been asked
for, or personally required, anything other
then client API and the svnlook API for use in hook procedures.

Barry

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Feb 21 00:20:23 2005

This is an archived mail posted to the Subversion Dev mailing list.