[gst-devel] python bindings

[Leif Johnson]

Hi Thomas -

On Sat, 04 Oct 2003, Thomas Vander Stichele wrote:

> I dug in and added an Element () constructor so that
> a) source = Element ("fakesrc", "mysource") works
> b) source = Element ("nogoodname", "badelement") triggers a RuntimeError

Nice ! I think this is much smoother than the long function name before.

> I also put in a testsuite dir to test bindings functionality as well, as
> a regression check.

Also nice. For some reason it's so much easier to code---even if that
means writing tests---in Python. Must be a mental block. : )

> Python heads (Dave, Leif, others ?), could you check if I did the right
> thing, and if I should get started on fixing other things as I go ?

The code looks ok to me, but I'm still kinda new to this wrapper writing
thing.

> I've also been told by the gnome python bindings people that it's not
> good to autogenerate gstreamer.defs by hand.  What do you think ?
> Maybe we should have a gstreamer-auto.defs that we regularly remake to
> check for new functions to wrap, but keep gstreamer.defs in CVS.

Yeah, it would be extra nice if we could get gst-python to compile with
either gst-0.6 or gst-0.7 (atm I can only get it to work with 0.7 if I
delete a function from the generated gstreamer.c file).

I'll try to see if I can do this. It's about time I helped out some
around here. : )

leif

--
Leif Morgan Johnson : http://ambient.2y.net/leif/