Re: gstreamermm / gmmproc Problems
- From: "Milosz Derezynski" <internalerror gmail com>
- To: "Murray Cumming" <murrayc murrayc com>
- Cc: Gtkmm Mailing List <gtkmm-list gnome org>
- Subject: Re: gstreamermm / gmmproc Problems
- Date: Thu, 21 Jun 2007 17:45:44 +0200
This works. Thanks! I'll submit a patch for the gmmproc docs.
Now there is another problem. I end up with literal _CONVERT macros inside the code which to my understanding should be post-processed again (it seems like it's a 2-stage thing where it picks up code, creates _CONVERT macros, which are then processed again).
But, the code still contains stuff like (i've prepended the namespace here, since it's Gst::Object, not Glib::Object, just making sure..):
<code>
bool Gst::Object::set_name(const Glib::ustring& name)
{
return _CONVERT(gboolean,bool,gst_object_set_name`'(ifelse(`',1,const_cast<__CNAME__*>(gobj()),gobj())`'ifelse(`_CONVERT(const Glib::ustring&,gchar*,name,0)',,,`, ')_CONVERT(const Glib::ustring&,gchar*,name,0)));
}
</code>
The wrapper inside gstobject.hg looks like this:
<code>
_WRAP_METHOD(bool set_name (const Glib::ustring& name), gst_object_set_name)
</code>
I have this here in gst_method.defs:
<code>
(define-method set_name
(of-object "GstObject")
(c-name "gst_object_set_name")
(return-type "gboolean")
(parameters
'("gchar*" "name")
)
)
</code>
And also basic conversion definitions for the basic pointer conversions inside convert_gst.m4 (which now works thanks to the previous hint, thanks again!)
_CONVERSION(`GstObject*',`Glib::RefPtr<Gst::Object>',`Glib::wrap($3)')
_CONVERSION(`Glib::RefPtr<Gst::Object>',`GstObject*',__CONVERT_REFPTR_TO_P)
_CONVERSION(`Glib::RefPtr<Gst::Object>&',`GstObject*',__CONVERT_REFPTR_TO_P)
What could be wrong?
--Milosz
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]