Commit graph

10477 commits

Author SHA1 Message Date
Mark Nauwelaerts
436a85ac8a avidemux: tweak DEFAULT format duration query response 2012-01-20 17:10:25 +01:00
Mark Nauwelaerts
14fe651b15 alphacolor: remove redundant statement 2012-01-20 17:10:21 +01:00
Mark Nauwelaerts
1911812572 flacdec: improve upstream peer duration querying
... to avoid accepting unhandled duration query result.
2012-01-20 17:10:19 +01:00
Mark Nauwelaerts
e44d930289 pulsesrc: additional error condition checking 2012-01-20 17:10:17 +01:00
Mark Nauwelaerts
3168b77e04 pulsesink: additional error condition checking 2012-01-20 17:10:14 +01:00
Mark Nauwelaerts
ad11ec4121 jpegenc: check _alloc_buffer result and perform fallback alloc if needed
... rather than carrying on with NULL buffer.
2012-01-20 17:10:11 +01:00
Tim-Philipp Müller
9ae5ddd8f9 Add --disable-fatal-warnings configure option 2012-01-19 14:24:04 +00:00
Vincent Penquerc'h
cf81df8058 cutter: fix leak of unused GValue 2012-01-16 17:51:18 +00:00
Vincent Penquerc'h
e9e0bebde3 tests: fix autodetect test not testing correctly for state change success
State change to PAUSED can be done async, so if this happens, we need
to wait for the change to be done (or failed).
2012-01-16 16:10:08 +00:00
Vincent Penquerc'h
2a7a38ca07 rtph263ppay: fix caps leak 2012-01-16 15:42:46 +00:00
Vincent Penquerc'h
4d51c68fc0 deinterlace: make interlacedness test deterministic
If the interlaced flag is not present in the caps, we assume the
data is not interlaced, instead of leaving the boolean uninitialized.
2012-01-16 12:13:50 +00:00
Vincent Penquerc'h
f0ac29113c pulsesrc: fix wrong error check
pa_stream_* functions return negative on error, despite the defines
for error codes being positive.

I only got to repro the error twice, so I'm not sure 100% sure this
fixes the issue (the negative var being uninitialized after returning
from pa_stream_get_latency).
2012-01-13 18:11:36 +00:00
Vincent Penquerc'h
f799fb2c90 oss4: fix caps leaks 2012-01-13 17:43:49 +00:00
Vincent Penquerc'h
b0608128b1 v4l2src: fix caps leak 2012-01-13 17:25:59 +00:00
Vincent Penquerc'h
4c2a2063cf tests: fix caps leak in videotestsrc test 2012-01-13 15:57:20 +00:00
Tim-Philipp Müller
f10e8192fa rtpptdemux: plug pad leak in error code path
Based on patch by: Stig Sandnes <stig.sandnes@cisco.com>

Don't leak srcpad if there are no caps.

https://bugzilla.gnome.org/show_bug.cgi?id=667820
2012-01-13 11:02:24 +00:00
Stig Sandnes
ffe010ee4d osxvideo: Fix leak of NSOpenGLPixelFormat object
https://bugzilla.gnome.org/show_bug.cgi?id=667818
2012-01-13 10:14:34 +00:00
Havard Graff
150403ab13 v4l2src: Don't assert when the interface is not implemented.
Simply return FALSE instead.

https://bugzilla.gnome.org/show_bug.cgi?id=667817
2012-01-13 10:07:20 +00:00
Raimo Järvi
e886ef8f5b waveformsink: Fix mingw warnings
https://bugzilla.gnome.org/show_bug.cgi?id=667719
2012-01-13 09:43:47 +00:00
Vincent Penquerc'h
654a04f90c gstrtpssrcdemux: fix element leak 2012-01-12 18:23:42 +00:00
Vincent Penquerc'h
f04d812719 matroska: do not leak attachment buffers 2012-01-12 14:19:22 +00:00
Vincent Penquerc'h
483514528a flacenc: do not drop the first data buffer on the floor (and leak it either) 2012-01-12 10:30:56 +00:00
Reynaldo H. Verdejo Pinochet
86053d9772 Temporarily disabling multifile for the Android build
There is a hard dependency on inotify comming from gio. We
are not currently bundling inotify with the Android dist so
I'm disabling multifile for now until someone gets around
to sort this out.

This change fixes building on Android
2012-01-11 18:57:43 -03:00
Stefan Sauer
bc1fa747a7 jack: add a transport mode enum
Clients can configure the desired behaviour via "transport" property. The
default behaviour is ignoring the transport state. Other modes are master and
slave.
2012-01-11 14:52:14 +01:00
Tim-Philipp Müller
834e58be9e tests: fix wavenc test on big endian
wavenc only accepts little-endian PCM, but most of our
elements such as audiotestsrc only produce or process
audio in native endianness, so we need to plug a
converter before wavenc on big endian systems.
2012-01-11 01:47:38 +00:00
Stefan Sauer
747e63f4e7 jack: deactivate the request_state code
When qjackctl is started, transport is stopped by default. This would be a
regression for gstreamer apps that before just started to play right away.
2012-01-10 23:02:45 +01:00
Stefan Sauer
7d4044aa46 jack: add transport control handling
This feature allows to start and stop playback from other jack applications (e.g. qjackctl).
2012-01-10 22:35:02 +01:00
Nicola Murino
d1bb060d71 matroskamux: fix codec_priv leaks
https://bugzilla.gnome.org/show_bug.cgi?id=667419
2012-01-10 18:29:06 +00:00
Stefan Sauer
0280ab04ed jack: use jack type for the callback
Jack headers have a typedef for the shutdown callback as well.
2012-01-10 15:08:16 +01:00
Vincent Penquerc'h
2b2c0940f1 matroskamux: fix codec string leaks 2012-01-09 17:28:17 +00:00
Vincent Penquerc'h
a6d9f6a3ce isomp4: fix caps leak 2012-01-05 19:25:33 +00:00
Vincent Penquerc'h
3489b691c5 isomp4: remove dead assignment 2012-01-05 19:08:03 +00:00
Tim-Philipp Müller
12ff707854 tests: fix some leaks and remove files when done in qtmux test 2012-01-04 18:05:41 +00:00
Peter Seiderer
b82d9ca240 multifilesink: post better error message when we run out of disk space
Map write errno ENOSPC to GST_RESOURCE_ERROR_NO_SPACE_LEFT.
2012-01-04 15:38:29 +00:00
Tim-Philipp Müller
2be912cb45 waveformsink: fix compiler warnings with MingW
https://bugzilla.gnome.org/show_bug.cgi?id=666485
2011-12-27 22:59:03 +00:00
Tim-Philipp Müller
56c1003f7a tests: make git ignore new unit test binary 2011-12-27 12:06:25 +00:00
Tim-Philipp Müller
86cd5bd7f2 udpsrc: fix valgrind warning
https://bugzilla.gnome.org/show_bug.cgi?id=666644
2011-12-27 11:50:03 +00:00
Tim-Philipp Müller
b744ad03cc udpsrc: add unit test that sends 0-size packet
https://bugzilla.gnome.org/show_bug.cgi?id=666644
2011-12-27 11:49:10 +00:00
John Ogness
0c4b60f010 udpsrc: drop dataless UDP packets
It is allowed to send/receive UDP packets with no data. When such
a packet is available, select() will return with success but
ioctl(FIONREAD) will return 0. But a read() must still occur in
order to clear off the UDP packet from the queue.

This patch will read the dataless packet from the socket. If
select() was woken for other reasons (and FIONREAD returns 0),
this may result in a UDP packet getting accidentally dropped.
But since UDP is not reliable, this is acceptable.

NOTE: This patch fixes a nasty bug where sending a dataless
      UDP packet to a udpsrc instance will cause an infinite
      loop.

https://bugzilla.gnome.org/show_bug.cgi?id=666644

Signed-off-by: John Ogness <john.ogness@linutronix.de>
2011-12-27 01:40:34 +00:00
Yaakov Selkowitz
e72b55b6ac waveform: add autotools bits for waveform plugin
https://bugzilla.gnome.org/show_bug.cgi?id=666485
2011-12-26 22:22:59 +00:00
Nicola Murino
7202d37c9d jpegdec: fix peer_caps leak
https://bugzilla.gnome.org/show_bug.cgi?id=666688
2011-12-26 15:34:47 -03:00
Tim-Philipp Müller
89521166b2 flvmux: don't try to push already-freed buffers
Fixes unit test.
2011-12-25 14:23:29 +00:00
Sebastian Dröge
348afe1669 wavparse: Use scale_ceil() functions from core instead of custom ones 2011-12-24 10:57:42 +01:00
Branko Subasic
a7d6690f92 matroskademux: do not consider duration of non-finalized file
... to avoid it clamping requested seek position.

Non-finalized file case, determined by whether
_parse_blockgroup_or_simpleblock ever updates the segment duration.

Fixes #652195.
2011-12-21 18:14:26 +01:00
Mark Nauwelaerts
628ae5c1a1 matroskademux: improve decision to fall back to scanning when seeking
... which is basically iff not streaming and no entry found in index
2011-12-21 15:45:36 +01:00
Oleksij Rempel (Alexey Fisher)
4f98b4ec85 ac3parse: remove unused variable
remove unused variable to fix compile error:
make -C audioparsers
make[3]: Betrete Verzeichnis '/home/lex/tmp/gst-plugins-good/gst/audioparsers'
  CC     libgstaudioparsers_la-gstaacparse.lo
gstaacparse.c: In function 'gst_aac_parse_read_loas_audio_specific_config':
gstaacparse.c:446:12: error: variable 'sbr' set but not used [-Werror=unused-but-set-variable]
cc1: all warnings being treated as errors

Signed-off-by: Oleksij Rempel (Alexey Fisher) <bug-track@fisher-privat.net>
2011-12-21 13:16:44 +00:00
Vincent Penquerc'h
93ac6fb47e ac3parse: let bsid 9 and 10 through
Files with 9 and 10 happen, and seem to comply with the <= 8
format, so let them through.
The spec says nothing about 9 and 10.

https://bugzilla.gnome.org/show_bug.cgi?id=658546
2011-12-20 19:29:42 +00:00
Mark Nauwelaerts
b175d1cbc4 flvmux: properly determine final duration
... which can be authoratively obtained from our own written timestamps.
2011-12-19 17:45:07 +01:00
Mark Nauwelaerts
fa5c02b211 flvmux: only write full metadata at start
... rather than having (potentially) unnecessary duplicates written all over,
or even contradictory varying filesize info, or duration info that will not
be rewritten upon header rewrite.
2011-12-19 17:45:04 +01:00
Mark Nauwelaerts
8c32195d51 flvmux: use GstCollectPads2 buffer callback and running time clipper
... since the default collection heuristics suffice.
2011-12-19 17:45:01 +01:00