Tim-Philipp Müller [Sat, 17 Mar 2012 16:09:31 +0000 (16:09 +0000)]
gst-element-maker: fix -template-video caps string
Don't forget separator.
Tim-Philipp Müller [Sat, 17 Mar 2012 16:05:20 +0000 (16:05 +0000)]
gst-element-maker: add video pad template and use it in videofilter class
Would be nicer if one could just supplement the generic template
from the element template though.
Also, I would really have liked to just add those sections from the
pads template into the element templet directly (so I can cater for
src template caps == sink template caps), but that didn't seem to
work.
Tim-Philipp Müller [Sat, 17 Mar 2012 15:49:45 +0000 (15:49 +0000)]
gst-element-maker: add {src,sink}pad-template, since many base classes create pads for us
And use them in the videofilter template.
Tim-Philipp Müller [Sat, 17 Mar 2012 15:16:29 +0000 (15:16 +0000)]
tools: add videofilter template for gst-element-maker
Tim-Philipp Müller [Sat, 17 Mar 2012 12:47:52 +0000 (12:47 +0000)]
opencv: fix configure check and build with opencv 2.1
AC_CHECK_HEADERS() calls action-if-not-found also if just one of
the headers checked for is missing, which is not what we wanted.
Also, check for highgui_c.h instead of highgui.hpp.
https://bugzilla.gnome.org/show_bug.cgi?id=672226
Stefan Sauer [Fri, 16 Mar 2012 16:07:46 +0000 (17:07 +0100)]
facedetect: initialize some arbitrary variable that gcc-4.6 can track properly
The access to them is clearly guarded by the booleans though.
Stefan Sauer [Fri, 16 Mar 2012 12:14:48 +0000 (13:14 +0100)]
opencv: cosmetic code changes
Check for the availability of the detectors in the processing function to avoid
setting up the ROIs. Use the booleans for have_xxx more.
Stefan Sauer [Fri, 16 Mar 2012 11:57:05 +0000 (12:57 +0100)]
opencv: use opencv prefix from pkgconfig to lookup datafiles
Don't hardconde /usr to lookup datafiles (such as haar cascades).
Vincent Penquerc'h [Thu, 15 Mar 2012 17:10:18 +0000 (17:10 +0000)]
rsndvdsrc: post a message with title number and durations
https://bugzilla.gnome.org/show_bug.cgi?id=672165
Stefan Sauer [Fri, 16 Mar 2012 11:10:10 +0000 (12:10 +0100)]
faceoverlay: remove blank line in gtk-doc comment
Stefan Sauer [Fri, 16 Mar 2012 11:04:11 +0000 (12:04 +0100)]
facedetect: skip detection for unavailable detectors
Stefan Sauer [Thu, 15 Mar 2012 22:17:55 +0000 (23:17 +0100)]
facedetect: move the detector to an own method to hide the ugly ifdefs
This improves the readability in the actual processing function.
Stefan Sauer [Thu, 15 Mar 2012 22:08:02 +0000 (23:08 +0100)]
facedetect: always send facedetect message
The application might like to know, when we don't see the face anymore.
Stefan Sauer [Thu, 15 Mar 2012 21:58:11 +0000 (22:58 +0100)]
faceoverlay: code cleanups
Use glib types and simply expressions in the message handler.
Stefan Sauer [Thu, 15 Mar 2012 21:49:51 +0000 (22:49 +0100)]
faceoverlay: don't access message fields directly
Stefan Sauer [Thu, 15 Mar 2012 21:22:12 +0000 (22:22 +0100)]
facedetect: don't skip faces
Skipping faces at this point makes us lie about the num-ber of faces and also
causes leaks.
Tim-Philipp Müller [Thu, 15 Mar 2012 18:28:01 +0000 (18:28 +0000)]
rsvgoverlay: cosmetic change
Rename PROP_FILENAME to PROP_LOCATION to match the name of
the property. Add some debug logging.
Tim-Philipp Müller [Thu, 15 Mar 2012 18:26:24 +0000 (18:26 +0000)]
faceoverlay: clear overlay if the face disappears
Tim-Philipp Müller [Thu, 15 Mar 2012 18:21:37 +0000 (18:21 +0000)]
faceoverlay: add some locking for properties, free previous location when it changes
Tim-Philipp Müller [Thu, 15 Mar 2012 18:08:09 +0000 (18:08 +0000)]
faceoverlay: don't make rsvgoverlay reload the SVG for every single frame
Only set location on rsvgoverlay if it has changed.
Tim-Philipp Müller [Thu, 15 Mar 2012 17:47:58 +0000 (17:47 +0000)]
faceoverlay: move face handling into own function and handle 0 face count
When a face disappears, we seem to get a message from facedetect with
a face count of 0, which we want to just ignore instead of trying to
access face #-1, which causes nasty warnings.
Tim-Philipp Müller [Thu, 15 Mar 2012 17:42:56 +0000 (17:42 +0000)]
faceoverlay: fix pad templates
Use generic and unspecififed rgb/caps for now. The exact caps
supported depend on the facedetect element and rsvgoverlay. It's
not clear how this worked before, since facedetect only accepts
24-bit RGB, but the caps advertised 32-bit ARGB/BGRA. In any case,
we don't want to force anything really, so that if any of those
elements acquires support for additional formats we pick those up
automatically.
Tim-Philipp Müller [Thu, 15 Mar 2012 17:26:14 +0000 (17:26 +0000)]
rsvgoverlay: fix crash due to double adapter unref
Tim-Philipp Müller [Thu, 15 Mar 2012 16:52:44 +0000 (16:52 +0000)]
faceoverlay: fix weird pad creation code
The element would create normal pads in its instance_init function,
and then later in NULL->READY create the elements it needs, remove
the pads created in the instance_init function, and add new ghost
pads instead. Not without saving the external peer pads of the old
pads of course, which it would promptly re-link to the new ghost
pads. Do all of that a bit differently.
Fixes the generic/states.check unit test.
https://bugzilla.gnome.org/show_bug.cgi?id=670588
Tim-Philipp Müller [Thu, 15 Mar 2012 15:47:24 +0000 (15:47 +0000)]
faceoverlay: minor cosmetic changes
Remove template maker comments, use our defines for
package name and origin.
Tim-Philipp Müller [Thu, 15 Mar 2012 00:13:40 +0000 (00:13 +0000)]
.gitignore: ignore more test binaries and stamp files
Tim-Philipp Müller [Thu, 15 Mar 2012 00:05:41 +0000 (00:05 +0000)]
opencv: make build with opencv 2.3.1 as in debian sid
Where highgui.h doesn't exist any more, but only opencv2/highgui/highgui.hpp.
Also, not quite sure why we're checking for cvaux.h, it's not used anywhere.
Dmitry Ketov [Wed, 14 Mar 2012 09:24:05 +0000 (10:24 +0100)]
winscreencap: Integrate into autotools build system
Fixes bug #672031.
Tim-Philipp Müller [Tue, 13 Mar 2012 20:50:04 +0000 (20:50 +0000)]
inter: fix warnings when gst-inspecting interaudio{src,sink}
The channel property isn't implemented, so don't register it.
Sreerenj Balachandran [Wed, 7 Mar 2012 23:48:56 +0000 (01:48 +0200)]
waylandsink: Fix warnings, proper structuring, dead code removal, adding doc section.
Oleksij Rempel (Alexey Fisher) [Mon, 12 Mar 2012 14:42:04 +0000 (15:42 +0100)]
geometrictransform: make sure gt->map not freed twice
current cheese can create situation where gt->map is freed twice.
This patch set map to null to avoid it.
https://bugzilla.gnome.org/show_bug.cgi?id=671910
Gil Pedersen [Tue, 28 Feb 2012 15:40:31 +0000 (16:40 +0100)]
hlsdemux: add floating point segment duration support
Thibault Saunier [Sat, 10 Mar 2012 23:53:54 +0000 (20:53 -0300)]
Fix an unwanted double negation from last commit
Thibault Saunier [Sat, 10 Mar 2012 22:10:52 +0000 (19:10 -0300)]
Fix 'ignoring return value of function declared with const attribute'
This always happens with GstByteReader/Writer and friends when
not taking into account returned boolean of the _read/_write functions
(which is actually wrong).
Make use of the *_unchecked variant as much as possible, or take the
returned value into account.
Mark Nauwelaerts [Sat, 10 Mar 2012 10:36:25 +0000 (11:36 +0100)]
dtsdec: avoid runaway loop when resyncing during parse
... by checking for the correct (decremented) size variable.
Fixes #671756.
Tim-Philipp Müller [Fri, 9 Mar 2012 17:16:24 +0000 (17:16 +0000)]
configure: fix AC_LANG_SOURCE-related warnings when doing autogen.sh
Danilo Cesar Lemes de Paula [Thu, 1 Mar 2012 17:59:55 +0000 (14:59 -0300)]
Fixing rtpvp8 compatibility with the third draft
https://bugzilla.gnome.org/show_bug.cgi?id=671073
Mark Nauwelaerts [Wed, 7 Mar 2012 11:59:28 +0000 (12:59 +0100)]
opusenc: only request and process 1 frame at a time
... since it is specified in _finish_frame that input buffer may be invalidated
after calling it, and is as such not reliably available for further encoding.
Also, requesting or allowing several frames is only useful if subclass intends
to process these "in 1 run" (as in, 1 output buffer), not for having another
(inner) loop in subclass where the baseclass one will do just fine.
Mark Nauwelaerts [Wed, 7 Mar 2012 11:55:43 +0000 (12:55 +0100)]
opusenc: configure baseclass requested samples really in samples
... as opposed to bytes.
Edward Hervey [Tue, 6 Mar 2012 17:49:11 +0000 (18:49 +0100)]
video: Update/add docs
Edward Hervey [Tue, 6 Mar 2012 17:11:35 +0000 (18:11 +0100)]
Add missing GLIB_DISABLE_DEPRECATION_WARNINGS
Suppress warnings about deprecated threading and GValueArray
API, so git compiles with -Werror.
Mark Nauwelaerts [Tue, 6 Mar 2012 17:33:25 +0000 (18:33 +0100)]
sirendec: port to audiodecoder
Mark Nauwelaerts [Tue, 6 Mar 2012 17:33:17 +0000 (18:33 +0100)]
sirenenc: port to audioencoder
Mark Nauwelaerts [Tue, 6 Mar 2012 15:11:52 +0000 (16:11 +0100)]
celtdec: use base class tag handling helper
... so as to ensure these to be handled and sent at proper time.
Mark Nauwelaerts [Tue, 6 Mar 2012 15:11:30 +0000 (16:11 +0100)]
dtsdec: use base class tag handling helper
... so as to ensure these to be handled and sent at proper time.
Mark Nauwelaerts [Tue, 6 Mar 2012 17:32:09 +0000 (18:32 +0100)]
dccp: avoid indefinite looping upon error
... and fix 'uninitialized' compilation warning as well.
Sebastian Dröge [Tue, 6 Mar 2012 14:21:17 +0000 (15:21 +0100)]
vp8enc: Fix 'argument to 'sizeof' in 'memset' call is the same expression as the destination' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:57:45 +0000 (14:57 +0100)]
gsettings: Fix 'if statement has empty body' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:56:20 +0000 (14:56 +0100)]
dfbvideosink: Fix 'implicit conversion from enumeration type 'DirectResult' to different enumeration type 'DFBResult'' compiler warning
We're not checking the return value of that function anyway.
Sebastian Dröge [Tue, 6 Mar 2012 13:51:57 +0000 (14:51 +0100)]
curl: Fix 'equality comparison with extraneous parentheses' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:51:26 +0000 (14:51 +0100)]
cogcolorspace: Use correct enum type for return value
Sebastian Dröge [Tue, 6 Mar 2012 13:50:06 +0000 (14:50 +0100)]
vdpau: Fix loop to not read before the valid memory area
Sebastian Dröge [Tue, 6 Mar 2012 13:47:17 +0000 (14:47 +0100)]
removesilence: Fix 'argument to 'sizeof' in 'memset' call is the same expression as the destination' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:46:03 +0000 (14:46 +0100)]
mxfdemux: Fix 'comparison of unsigned expression >= 0 is always true' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:44:25 +0000 (14:44 +0100)]
mpegtsparse: Fix 'variable 'pid' is uninitialized when used here' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:43:29 +0000 (14:43 +0100)]
mpegdemux: Fix 'expression result unused' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:41:00 +0000 (14:41 +0100)]
dccp: Fix 'comparison of unsigned expression < 0 is always false' compiler warning
Sebastian Dröge [Tue, 6 Mar 2012 13:38:47 +0000 (14:38 +0100)]
baseaudiovisualizer: Fix 'comparison of unsigned expression >= 0 is always true' compiler warning
Edward Hervey [Mon, 5 Mar 2012 11:46:13 +0000 (12:46 +0100)]
mpegtspacketizer: Handle rollover in offset calculations
Edward Hervey [Mon, 5 Mar 2012 10:27:51 +0000 (11:27 +0100)]
tsdemux: printf fix
Edward Hervey [Mon, 5 Mar 2012 09:14:46 +0000 (10:14 +0100)]
tsdemux: Update TODO
Edward Hervey [Mon, 5 Mar 2012 08:46:17 +0000 (09:46 +0100)]
mpegts: Switch rank of mpegtsdemux and tsdemux
More effort has been put lately in tsdemux and works in more use cases
Edward Hervey [Mon, 5 Mar 2012 08:38:57 +0000 (09:38 +0100)]
tsdemux: Push packets as early as possible
When the PES header tells us how big the outgoing packet is, push the
packet downstream as soon as we have the specified size instead of waiting
for the beginning of the next packet.
Reduces latency and removes issues with very sparse streams (like subtitles
and subpictures).
Stefan Sauer [Sun, 4 Mar 2012 20:54:08 +0000 (21:54 +0100)]
shm: add minimal doc blobs
Sreerenj Balachandran [Tue, 28 Feb 2012 19:00:24 +0000 (21:00 +0200)]
waylandsink: Fix in shell_surface_set_fullscreen
Sreerenj Balachandran [Thu, 16 Feb 2012 20:27:23 +0000 (22:27 +0200)]
waylandsink: More fixes -- update to Wayland 0.85 protocol -- Added shell surface -- Added format query for shm format -- Create the window based on width and height of video extracted from upstream caps.
Sreerenj Balachandran [Wed, 2 Nov 2011 14:23:02 +0000 (16:23 +0200)]
Update with new APIs in wayland-client.
Sreerenj Balachandran [Wed, 2 Nov 2011 11:51:13 +0000 (13:51 +0200)]
More Fixes: * Adding buffer_alloc * perform buffer_damage before surface_attach * Fix typo, Remove Dead code etc.
Sreerenj Balachandran [Wed, 2 Nov 2011 09:02:11 +0000 (11:02 +0200)]
Initial Commit: Adding Wayland Video Sink
Stefan Sauer [Tue, 13 Dec 2011 22:43:59 +0000 (23:43 +0100)]
resindvd: send duration message on duration updates
When we know a new duration, report it. Add more logging.
Stefan Sauer [Tue, 13 Dec 2011 09:26:50 +0000 (10:26 +0100)]
audiovisualizers: planning + example update
Руслан Ижбулатов [Fri, 2 Mar 2012 17:41:39 +0000 (21:41 +0400)]
d3dvideosink: fix compiler warnings and build failure with mingw
shared.d3ddev was removed a while back, not sure how this still
compiles for anyone (tpm).
https://bugzilla.gnome.org/show_bug.cgi?id=653718
https://bugzilla.gnome.org/show_bug.cgi?id=670143
https://bugzilla.gnome.org/show_bug.cgi?id=656905
Tim-Philipp Müller [Sat, 3 Mar 2012 15:57:11 +0000 (15:57 +0000)]
configure: use AS_LIBTOOL and bump version for ABI change in h.264 codec parser
Backported from the 0.10 release branch, where v.23 was used.
https://bugzilla.gnome.org/show_bug.cgi?id=671203
Gwenole Beauchesne [Fri, 2 Mar 2012 10:45:41 +0000 (11:45 +0100)]
codecparsers: h264: record number of emulation prevention bytes in slice_header().
Some hardware video decode acceleration API (VA-API, DXVA) require
a bit count to the first macroblock, minus the number of emulation
prevention bytes. So, instead of having the consumer of the library
scan the slice_header() again, just record that number while parsing.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
https://bugzilla.gnome.org/show_bug.cgi?id=671203
Tim-Philipp Müller [Sat, 3 Mar 2012 15:47:01 +0000 (15:47 +0000)]
tsdemux: fix confusing variable name
Edward Hervey [Fri, 2 Mar 2012 16:11:33 +0000 (17:11 +0100)]
tsdemux: Fix segment start position
If we *really* can't figure out the first start position, that most
likely means the data to push out doesn't have any timestamp.
Use a default value of 0 then
Edward Hervey [Fri, 2 Mar 2012 16:10:28 +0000 (17:10 +0100)]
mpegtsbase: Flush out the packetizer on flushing seeks
... else we end up with bogus data/offsets
Thibault Saunier [Wed, 29 Feb 2012 14:14:15 +0000 (11:14 -0300)]
tsdemux: Activate pads only when receiving data for the stream
https://bugzilla.gnome.org/show_bug.cgi?id=670921
Andoni Morales Alastruey [Fri, 2 Mar 2012 14:37:57 +0000 (15:37 +0100)]
d3dvideosink: add missing directx headers
Edward Hervey [Fri, 2 Mar 2012 10:07:47 +0000 (11:07 +0100)]
mpegvideoparse: Fix level code
Valid ranges are from 4 to 10
Valid ranges >> 1 are from 2 to 5
Valid ranges >> 1 - 2 are from 0 to 3
(and not from 1 to 4)
Edward Hervey [Fri, 2 Mar 2012 09:54:48 +0000 (10:54 +0100)]
mpegtsbase: Check harder for program changes
And don't just rely on the section crc/version_number changing to
indicate that the program actually changed.
Tim-Philipp Müller [Thu, 1 Mar 2012 18:46:51 +0000 (18:46 +0000)]
vdpau: relicense four source files from GPL3+ to LGPL2+
These source files were mistakenly licensed as GPL3, the
author (Carl-Anton Ingmarsson) has given permission to
relicense them to LGPL2+.
https://bugzilla.gnome.org/show_bug.cgi?id=671142
Edward Hervey [Thu, 1 Mar 2012 17:05:17 +0000 (18:05 +0100)]
mpegtsbase/tsdemux: Refactor seek and segment handling
All calculations go through the mpegtspacketizer
Remove unused variables/code
Edward Hervey [Thu, 1 Mar 2012 16:59:57 +0000 (17:59 +0100)]
mpegtsbase: Error out on EOS without any pad
Avoids ending up with hanging pipelines
Edward Hervey [Thu, 1 Mar 2012 16:56:34 +0000 (17:56 +0100)]
mpegtspacketizer: Offset calculation
Allows PCR<=>PTS<=>offset estimation/calculation
Right now the calculation is very naive, but can be extended later on
without disrupting the code in tsdemux/mpegtsbase
Edward Hervey [Thu, 1 Mar 2012 16:53:41 +0000 (17:53 +0100)]
mpegtspacketizer: clock fix and minor addition
* Don't take into account packets that arrived at the same time as
previous ones for clock skew estimation
* Add convenience method for processing the next ts packet
Edward Hervey [Wed, 29 Feb 2012 17:19:23 +0000 (18:19 +0100)]
Revert "mpegtsdemux: Not apply various time the same PMT to a program when repetead"
This reverts commit
8cb0e87f5623836ddc361eb91dcf9b50b48048e7.
mpegtspacketizer already checks if it's a new PMT or not
Edward Hervey [Wed, 29 Feb 2012 09:33:00 +0000 (10:33 +0100)]
tsdemux: Remove all seeking code
It was more than bogus
Xavier Queralt [Tue, 14 Feb 2012 10:57:00 +0000 (11:57 +0100)]
hlsdemux: don't set the fetcher's location twice
Calling gst_element_make_from_uri already does it and filesrc
fails when using file protocol in a location
Guillaume Desmottes [Tue, 28 Feb 2012 15:27:55 +0000 (16:27 +0100)]
Add missing GLIB_DISABLE_DEPRECATION_WARNINGS
Suppress warnings about deprecated threading and GValueArray
API, so git compiles with -Werror.
https://bugzilla.gnome.org/show_bug.cgi?id=670982
Xavier Queralt [Fri, 24 Feb 2012 08:54:35 +0000 (09:54 +0100)]
hlsdemux: don't fail for playlists smaller than fragments-cache
https://bugzilla.gnome.org/show_bug.cgi?id=670963
Thibault Saunier [Mon, 27 Feb 2012 12:45:29 +0000 (09:45 -0300)]
mpegtspacketizer: catch section lengths extending past the buffer length
This is probably the cause for an occasional crash while streaming
MPEG. Blind fix after staring at the code and following logic, so
may or may not fix the issue, I cannot test.
(Port of
4275a70cb55d375afa702917f7359ec117ed49d4 from mpegdemux)
Thibault Saunier [Mon, 27 Feb 2012 12:42:27 +0000 (09:42 -0300)]
mpegtspacketizer: support more character set encodings
Support UTF-16BE, EUC-KR (KSX1001), GB2312 and ISO-10646/UTF8 text
encoding and fixed new line for multibyte encoding
https://bugzilla.gnome.org/show_bug.cgi?id=664257
(Port of
9759d66407f2be8ec29975b0eff3230bb1dae0ef from the mpegtsdemux
element)
Thibault Saunier [Fri, 24 Feb 2012 22:08:40 +0000 (19:08 -0300)]
tsbase: Fix parsing of PSI table IDs
(Ported from mpegtsdemux
d8fd874f5290e4911437120057ee885cdb68b4af)
Thibault Saunier [Fri, 24 Feb 2012 18:26:21 +0000 (15:26 -0300)]
tsdemux: Avoid unlinkely leaks and segfault
Edward Hervey [Fri, 24 Feb 2012 16:53:52 +0000 (17:53 +0100)]
tsdemux: Use new clock skew estimation for outgoing timestamps
Only used in live mode
Edward Hervey [Fri, 24 Feb 2012 16:52:33 +0000 (17:52 +0100)]
tsdemux: Various code cleanup and dead code removal
Edward Hervey [Fri, 24 Feb 2012 16:47:38 +0000 (17:47 +0100)]
mpegtspacketizer: Fix issues with skew code
* Always try to get a timestamp for the algorithm
* Remove dead variable
* Return proper default value
Edward Hervey [Fri, 24 Feb 2012 16:07:43 +0000 (17:07 +0100)]
dvbsrc: Timestamp output by default
Allows downstream elements (like demuxers or parsers) to do remote
clock rate/skew estimation.
Thibault Saunier [Thu, 23 Feb 2012 17:46:09 +0000 (14:46 -0300)]
mpegtspacketizer: Estimate clock skew and retimestamps buffers more precisly
Apply the EPTLA algotithm to estimate clock skew.
Reusing code from -good/gst/rtpmanager/rtpjitterbuffer.c