Hi All,
I'm trying to write some python bindings for Anjuta.
I have hit an issue with pygtk which is reported as bug #503668 (
http://bugzilla.gnome.org/show_bug.cgi?id=503668). Anjuta library uses
quite a lots of GInterface and an object (typically a plugin) can
implement several interfaces. But two interfaces could have functions
with the same name.
In C, it's not a problem because all interface functions are prefixed by
the interface name. In Python, it is an issue because it's not the case.
I have written a patch to fix this. It adds an additional attribute
vmethod allowing to specify the name of the overridden C function which
can be different from the python function name. If this attribute is not
present the same name is used for both, so the backward compatibility is
kept (it's mandatory). By example I can have
(define-virtual clear_indicable
(of-object "IAnjutaIndicable")
(c-name "ianjuta_indicable_clear")
(vmethod "clear")
(return-type "none")
(parameters
'("GError**" "err")
)
)
Here, the python function name is clear_indicable, while the C method is
clear.
I have get some comments about this patch and I have taken them into
account but it's now stuck for several months. I haven't pushed it that
much because I have get other priorities (Anjuta 2.4.x). But now we
would like to have working python bindings for the next release of GNOME
and a few people have started to work on this, there is a small web site
too http://libanjutapython.sourceforge.net/
Could I get this patch upstream, or could you tell me what should be
improved or why it cannot be committed ?
Regards,
Sébastien
_______________________________________________
pygtk mailing list [email protected]
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/
