Python bug?

[Bernd Gehrmann]

On Wed, 2 May 2001, Richard Dale wrote:
> On Wed, 02 May 2001, Marcel Strittmattery wrote:
> It's still at the 'proof of concept' stage - there are only about 4 methods
> done I believe. But it uses the standard embedded python api, so all it needs
> are more wrapped methods - it's going in the right direction...

I have implemented a new set of python bindings for DCOP last week.
If I get this finished, this makes all DCOP interfaces immediately
available from python, without any interface generator or so.
In contrast to kdebindings/dcoppython, it also allows for user-
defined marshallers implemented in python, so e.g. it is possible 
to return even non-trivial classes like ParsedClass via DCOP.

Hopefully I can finish a usable version this week...
although my thesis, the good weather and kdevelop are at
least one conflicting thing too much ;-(

> I've recently been looking at the Qt/KDE python bindings code, and it has
> really good Qt support, but only KDE 1.x with 100 classess or so. I think a
> proper KDE binding for 2.x needs around 300+ classes (I've done about 450 for
> java). The SIP (Qt/KDE python bindings generator) stuff seems quite labour
> intensive, and so a lot of work to do 400 classes. But perhaps we could use it
> to do a useful subset.

I think having KApplication and KMainWindow would already be 
nice. They provide most of the KDE look & feel. Qt/PyQt is
already quite complete for itself.

Bernd.


-
to unsubscribe from this list send an email to kdevelop-request at kdevelop.org with the following body:
unsubscribe »your-email-address«