[gstreamer-bugs] [Bug 413418] Make GstAppSrc better

GStreamer (bugzilla.gnome.org) bugzilla-daemon at bugzilla.gnome.org
Sat Mar 3 01:01:17 PST 2007


Do not reply to this via email (we are currently unable to handle email
responses and they get discarded).  You can add comments to this bug at
http://bugzilla.gnome.org/show_bug.cgi?id=413418

  GStreamer | gst-plugins-bad | Ver: HEAD CVS


David Schleef changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ds at schleef.org
OtherBugsDependingO|339897                      |
              nThis|                            |




------- Comment #7 from David Schleef  2007-03-03 08:59 UTC -------
> (2) are the inclusion of config.h, unistd.h, fcntl.h and string.h needed?

config.h, always

> (4) are the enum, gst_app_src_get_properties and gst_app_src_set_properties
> needed if there are no properties ?

Doesn't matter.  I always leave them.

> (6) in gst_app_src_unlock, ->unlock is accessed and the condition variable
> signalled without holding the mutex : shouldn't it hold the mutex ?

no.

> (7) in gst_app_src_unlock, wouldn't it be safer to broadcast the condition
> variable (although the current state of the code makes me think signal is
> enough)

no.

> (8) in gst_app_src_create, shouldn't ->caps be tested for NULL before use ?

no.

> (9) if that class is to be used by a program not based on GStreamer, shouldn't
> it have a method
> void gst_app_src_push_data (GstAppSrc *appsrc,
>                             gpointer data,
>                             guint length);
> which builds a GstBuffer then calls gst_app_src_push_buffer with it ?
> 

This would need a callback and a closure.  I don't think it's a good idea at
this time.


-- 
Configure bugmail: http://bugzilla.gnome.org/userprefs.cgi?tab=email




More information about the Gstreamer-bugs mailing list