Commit graph

10244 commits

Author SHA1 Message Date
Wim Taymans
be50d18a57 warptv: port to 0.11 2011-07-06 18:50:26 +02:00
Wim Taymans
a462c14b9d dice: keep track of info 2011-07-06 18:50:15 +02:00
Wim Taymans
a1b4b089ec effectv: port dice 2011-07-06 18:32:45 +02:00
Wim Taymans
50b0751799 effectv: port agingtv 2011-07-06 18:09:49 +02:00
Wim Taymans
da28ebfbe3 aasink: port to new video API 2011-07-06 17:50:54 +02:00
Wim Taymans
1a0a6f54bb cacasink: port to 0.11 2011-07-06 17:40:20 +02:00
Wim Taymans
f70da0a542 jpeg: beginnings of porting to 0.11 2011-07-06 16:51:36 +02:00
Mark Nauwelaerts
a6d9995705 wavparse: use ALWAYS source pad rather than SOMETIMES 2011-07-06 16:32:13 +02:00
Mark Nauwelaerts
e9146e7aec wavparse: port to 0.11 2011-07-06 16:32:13 +02:00
Mark Nauwelaerts
9cf287ebcc wavenc: port to 0.11 2011-07-06 16:32:11 +02:00
Mark Nauwelaerts
88f107b30b qtdemux: adjust to unsigned segment fields 2011-07-06 16:32:11 +02:00
Wim Taymans
fdf5a49422 speex: port speex elements 2011-07-06 15:57:23 +02:00
Wim Taymans
3fd1106b7e Merge branch 'master' into 0.11 2011-07-06 12:05:12 +02:00
Mark Nauwelaerts
ef02634dc6 rtpmanager: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer and corresponding rtp and rtcp buffer interfaces
* misc caps and segment handling changes

FIXME: also relies on being able to pass caps along with a buffer,
which has no evident equivalent yet, so that either needs one,
or still needs quite some code path modification to drag along caps.
2011-07-06 10:16:12 +02:00
René Stadler
ae87731de5 pulsesink: prevent race condition causing ref leak
Since commit 8bfd80, gst_pulseringbuffer_stop doesn't wait for the
deferred call to be run before returning. This causes a race when
READY->NULL is executed shortly after, which stops the mainloop. This
leaks the element reference which is passed as userdata for the callback
(introduced in commit 7cf996, bug #614765).

The correct fix is to wait in READY->NULL for all outstanding calls to
be fired (since libpulse doesn't provide a DestroyNotify for the
userdata). We get rid of the reference passing from 7cf996 altogether,
since finalization from the callback would anyways lead to a deadlock.

Re-fixes bug #614765.
2011-07-05 16:36:17 +02:00
René Stadler
f8456e2a1a pulsesink: small cleanup of copy-paste code 2011-07-05 16:36:17 +02:00
René Stadler
3589cee762 pulsesink: remove unused member variable and misleading log message
Wim changed it in commit 8bfd80 so that pa_defer_ran is not read
anywhere.

The log message used to annotate a mainloop_wait call which is gone.
2011-07-05 16:36:17 +02:00
Wim Taymans
794320a202 videoflip: fix caps 2011-07-05 15:37:52 +02:00
Wim Taymans
301e3e32ff effectv: port edgetv 2011-07-05 11:46:16 +02:00
Tim-Philipp Müller
4c70f79e35 Add -DGST_USE_UNSTABLE_API to the compiler flags to avoid warnings 2011-07-05 10:12:25 +01:00
David Schleef
f2c8761a50 goom: Don't answer lantency queries before negotiation 2011-07-04 12:59:50 -07:00
Wim Taymans
f1df4c4b32 udp: port to new API 2011-07-04 18:15:42 +02:00
Wim Taymans
8b040cfae2 pulse: remove implementsinterface 2011-07-04 18:12:56 +02:00
Wim Taymans
99c102578f alpha: fix caps 2011-07-04 18:10:55 +02:00
Wim Taymans
b9b007e1be alpha: port to new video API 2011-07-04 18:06:48 +02:00
Wim Taymans
3c27f2435a alpha: more porting 2011-07-04 17:00:34 +02:00
Wim Taymans
476e2184f4 port to new video api 2011-07-04 16:49:55 +02:00
Wim Taymans
ead60adb33 video: port to new video apis 2011-07-04 16:49:47 +02:00
Mark Nauwelaerts
0c25863253 jpegdec: avoid crashing on invalid input without components 2011-07-04 14:32:27 +02:00
Mark Nauwelaerts
54c9b13ea7 flv: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer
* misc segment and caps changes
2011-07-04 11:50:02 +02:00
Mark Nauwelaerts
d59a00aa1c Merge branch 'master' into 0.11
Conflicts:
	ext/pulse/pulsesink.c
2011-07-04 11:48:13 +02:00
Mark Nauwelaerts
b307c74e9e flvmux: pass along segment info to collectpads
... so it can track this and be subsequently used to determine running time etc.
2011-07-04 11:36:48 +02:00
Mark Nauwelaerts
350b9b9d2d flvdemux: indicate raw format in aac caps 2011-07-04 11:36:46 +02:00
Mark Nauwelaerts
611c950868 qtmux: mind requested name for request pad 2011-07-04 11:07:13 +02:00
Mark Nauwelaerts
1ea692bc63 avidemux: free scheduling query 2011-07-04 11:06:54 +02:00
David Schleef
f69dcaab58 pulse: Increase ranks to PRIMARY + 10
So that pulsesrc/pulsesink get chosen over other possible PRIMARY
src/sinks by autoaudiosink.  Presumably, if pulse is available, it
is always preferred over another src/sink.

Fixes: #647540.
2011-07-03 19:53:42 -07:00
David Schleef
32aa80cda8 multipartmux: Add \r\n to tail of pushed buffers
Clients such as Firefox require the \r\n after the payload.
2011-07-03 18:27:17 -07:00
Branko Subasic
e385108f84 matroskademux: avoid looping when searching for clusters
Fixes some bugs that results in the demuxer looping when seaching
for clusters in non-finalized files.

https://bugzilla.gnome.org/show_bug.cgi?id=652195
2011-07-01 17:19:40 +01:00
Mark Nauwelaerts
70b561bc13 multifile: port to 0.10
* use G_DEFINE_TYPE
* adjust to new GstBuffer
* misc caps handling
2011-06-30 12:35:35 +02:00
Mark Nauwelaerts
01fcf3f2ed cutter: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer
* minor misc
2011-06-30 12:35:34 +02:00
Mark Nauwelaerts
19145ce5f5 replaygain: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer
2011-06-30 12:35:33 +02:00
Mark Nauwelaerts
8d73748509 spectrum: remove deprecated property 2011-06-30 12:35:32 +02:00
Mark Nauwelaerts
2d9bcbdf03 spectrum: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer
2011-06-30 12:35:31 +02:00
Mark Nauwelaerts
696ffc85d7 level: port to 0.11
* use G_DEFINE_TYPE
* adjust to new GstBuffer
2011-06-30 11:19:12 +02:00
Mark Nauwelaerts
d9ba4b7616 equalizer: port to 0.11 2011-06-30 10:30:16 +02:00
Debarshi Ray
c8619aa0fd matroskaparse: fix reference counting of parse->streamheader
https://bugzilla.gnome.org/show_bug.cgi?id=652286

Signed-off-by: David Schleef <ds@schleef.org>
2011-06-29 23:56:53 -07:00
David Schleef
2f94df8032 jpegenc: Don't round up size of encoded buffers
For some reason, in code dating to 2001, encoded jpeg buffers were
rounded up to multiples of 4 bytes.  With the added bonus that the
extra bytes are unwritten, causing valgrind issues.  Oops.  I can't
think of any reason why JPEG buffers need to be multiples of 4 bytes,
so I removed the padding.  There might be some code somewhere that
depends on this behavior, so if this needs to be reverted, please fix
the valgrind issues.
2011-06-29 23:55:33 -07:00
Mark Nauwelaerts
f886681367 isomp4: port to 0.11 2011-06-29 12:46:20 +02:00
Mark Nauwelaerts
65bb271a95 avidemux: tweak some ported segment handling
... to avoid losing duration during push mode seeking, and to properly
accumulate running time when segment seeking.
2011-06-29 12:16:38 +02:00
Mark Nauwelaerts
cc526d2e52 qtmux: free date tag 2011-06-29 12:05:23 +02:00