Previous blitting code could potentially write past the
end of the buffer if the x or y position was odd, and for
the same underlying reason, didn't get the chroma registration
correct in the odd position case.
https://bugzilla.gnome.org/show_bug.cgi?id=647830
Remove the android/ top dir
Fixe the Makefile.am to be androgenized
To build gstreamer for android we are now using androgenizer which generates the needed Android.mk files.
Androgenizer can be found here: http://git.collabora.co.uk/?p=user/derek/androgenizer.git
The rationale is that it can't be properly used right now when using
it to encode mpeg2video because of the needs-to-be-rewritten properties
and format negotiation. Other encoders will negotiate in a much saner
fashion.
One such example is that when you pick mpeg2enc for mpeg2video, the
default value for the 'format' property is "Generic MPEG-1", which is
completely wrong if downstream caps are mpeg2. The whole negotiation
code needs some serious loving before this plugin can be bumped back
up to a higher rank.
It seems xvidcore overreads its input buffer, so a nasty workaround
is to allocate some more memory (16 bytes seem to be enough).
There is no apparent image corruption with these extra bytes set to 0,
valgrind is much happier, and the crashes go away.
It is ugly, and slower though. But then, xviddec is currently
not autoplugged for playback anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=334107
When passing an unexistent file to rsvgoverlay it would
crash because the svg loading would fail without setting
an error.
This patch makes it check if the handle was actually created
and logs an error in case it didn't. Maybe it should post an
error to the bus, but the previous error handling didn't, so
I just followed the same logic.
Sync both pads by waiting in the kate chain function.
Do not reset our internal segment from segment updates, in order
to be able to map video running time to kate running time, to
give libtiger the timestamp it expects. This allows us to use
running time to sync to video, which is The Right Way.
https://bugzilla.gnome.org/show_bug.cgi?id=600929
Don't leak source caps. Use GST_PTR_FORMAT to log caps, so
we don't need to leak strings from gst_caps_to_string().
No need to use GST_DEBUG_FUNCPTR for vfuncs where the base
class will never look them up (like property getters/setters).
Don't use g_return_*_if_fail() for things that aren't directly
programming errors (by the application developer).
Fixes kate unit test under valgrind.
When Totem switches streams, tiger will be reset, and start receiving
buffers from the middle of the stream, without being sent headers.
If this happens, try to get headers from the caps.
https://bugzilla.gnome.org/show_bug.cgi?id=638004
