platform/upstream/gstreamer.git
13 years agoceltenc: allow actually overriding frame size
Vincent Penquerc'h [Wed, 14 Sep 2011 16:07:50 +0000 (17:07 +0100)]
celtenc: allow actually overriding frame size

https://bugzilla.gnome.org/show_bug.cgi?id=659087

13 years agokate: flush on seek, and ensure we cancel any waiting buffer
Vincent Penquerc'h [Tue, 13 Sep 2011 15:24:32 +0000 (16:24 +0100)]
kate: flush on seek, and ensure we cancel any waiting buffer

This is what pango textoverlay does, and I've had that condition
variable get stuck once.

https://bugzilla.gnome.org/show_bug.cgi?id=658940

13 years agoh264parse: do not try to read from a non existent VUI
Vincent Penquerc'h [Mon, 12 Sep 2011 10:17:20 +0000 (11:17 +0100)]
h264parse: do not try to read from a non existent VUI

https://bugzilla.gnome.org/show_bug.cgi?id=658800

13 years agocamerabin2: Adding 2 more log messages
Thiago Santos [Fri, 16 Sep 2011 03:32:02 +0000 (00:32 -0300)]
camerabin2: Adding 2 more log messages

13 years agocamerabin2: replace deprecated g_atomic_int_exchange_and_add
Vincent Penquerc'h [Wed, 14 Sep 2011 14:20:09 +0000 (15:20 +0100)]
camerabin2: replace deprecated g_atomic_int_exchange_and_add

It was replaced by g_atomic_int_add, which now returns the
old value.

https://bugzilla.gnome.org/show_bug.cgi?id=659061

13 years agocamerabin2: Reset last state change result on state-locked elements
Thiago Santos [Thu, 15 Sep 2011 20:35:44 +0000 (17:35 -0300)]
camerabin2: Reset last state change result on state-locked elements

An element stores the result for the last state change it did and
GstBin's state change handler will use this last result for state
locked elements to decide if its state change was successfull or not.

In camerabin2, the filesinks have their state locked and when they
fail switching states, this last failure will be used if the application
tries to change camerabin2's state, causing any state change to fail.

This patch makes camerabin2 reset this last change failure, avoiding
that camerabin2 fails on its next state changes.

13 years agospanplc: Fix typos
Olivier Crête [Thu, 15 Sep 2011 22:17:24 +0000 (18:17 -0400)]
spanplc: Fix typos

13 years agompegvideoparse: Handle properly frame detection
Thibault Saunier [Tue, 13 Sep 2011 23:40:35 +0000 (20:40 -0300)]
mpegvideoparse: Handle properly frame detection

Handle the case where we have buffers containing no startcode
in the middle of a frame properly.

13 years agocodecparsers: mpegvideo: Debugging symbols enhancements
Thibault Saunier [Tue, 13 Sep 2011 23:38:27 +0000 (20:38 -0300)]
codecparsers: mpegvideo: Debugging symbols enhancements

13 years agocodecparsers: mpegvideo: Add the ChromaType enum
Thibault Saunier [Mon, 12 Sep 2011 16:54:37 +0000 (13:54 -0300)]
codecparsers: mpegvideo: Add the ChromaType enum

13 years agocodecparsers: mpegvideo: Change broken_gop to broken_link
Thibault Saunier [Mon, 12 Sep 2011 16:41:41 +0000 (13:41 -0300)]
codecparsers: mpegvideo: Change broken_gop to broken_link

This way we respect the standard naming

13 years agompegvideoparse: send a video-codec tag
Mark Nauwelaerts [Wed, 14 Sep 2011 14:14:24 +0000 (16:14 +0200)]
mpegvideoparse: send a video-codec tag

https://bugzilla.gnome.org/show_bug.cgi?id=569305

13 years agocamerabin2: Check zoom notify callback from camera-source
Thiago Santos [Mon, 12 Sep 2011 19:50:10 +0000 (16:50 -0300)]
camerabin2: Check zoom notify callback from camera-source

Camerabin2 has a zoom property that is simply proxied to its
internal camera-source element. This patch makes camerabin2 listen
to 'notify' signals from it so it can update its zoom property value
when camera-source changes its zoom as a side-effect of another operation
or because the user set the zoom directly to it, instead of doing
it from camerabin2.

13 years agortmpsink: don't block the main thread with RTMP_Connect
Alessandro Decina [Mon, 29 Aug 2011 06:00:22 +0000 (08:00 +0200)]
rtmpsink: don't block the main thread with RTMP_Connect

Move the RTMP_Connect call from the main thread (::start) to the streaming
thread (::render).

13 years agocodecparsers: fix compiler warnings
Alessandro Decina [Mon, 12 Sep 2011 08:33:06 +0000 (10:33 +0200)]
codecparsers: fix compiler warnings

13 years agompegvideoparse: fix compiler warnings
Alessandro Decina [Mon, 12 Sep 2011 08:23:19 +0000 (10:23 +0200)]
mpegvideoparse: fix compiler warnings

13 years agoh264parse: Make use of the h264_nal_parser_free function
Thibault Saunier [Fri, 9 Sep 2011 17:22:39 +0000 (14:22 -0300)]
h264parse: Make use of the h264_nal_parser_free function

13 years agocodecparsers: h264: Implement the nal_parser_free function
Thibault Saunier [Fri, 9 Sep 2011 17:15:21 +0000 (14:15 -0300)]
codecparsers: h264: Implement the nal_parser_free function

13 years agocodecparsers: h264: Use g_slice instead of g_malloc in nal_parser_new
Thibault Saunier [Wed, 7 Sep 2011 00:29:28 +0000 (21:29 -0300)]
codecparsers: h264: Use g_slice instead of g_malloc in nal_parser_new

13 years agocodecparsers: h264: Update the documentation
Thibault Saunier [Wed, 7 Sep 2011 00:28:45 +0000 (21:28 -0300)]
codecparsers: h264: Update the documentation

Fix a few error.
Document missing parts.
Add links to the specifications

13 years agohlsdemux: Fix merge mistake from last commit
Sebastian Dröge [Fri, 9 Sep 2011 10:27:57 +0000 (12:27 +0200)]
hlsdemux: Fix merge mistake from last commit

13 years agohlsdemux: Report the correct timestamp for live streams
Youness Alaoui [Thu, 8 Sep 2011 23:56:33 +0000 (23:56 +0000)]
hlsdemux: Report the correct timestamp for live streams

Buffers would always start with timestamp 0 and we'd start streaming
from the first buffer, but live streams always start streaming from
the last fragment - 3 fragments in the playlist, which makes its
timestamp, as returned by get_next_fragment, be whatever position
they had in the playlist. This makes sure the position correctly
reports the position of the buffer in the playlist, and added a shifting
variable to allow seeking in the middle of fragments.

13 years agohlsdemux/tsdemux: Add HP copyright on work done on HLS
Youness Alaoui [Thu, 8 Sep 2011 22:13:19 +0000 (22:13 +0000)]
hlsdemux/tsdemux: Add HP copyright on work done on HLS

13 years agohlsdemux: set playlist to NULL after unreffing it to avoid a double-unref in the...
Youness Alaoui [Thu, 8 Sep 2011 20:29:42 +0000 (20:29 +0000)]
hlsdemux: set playlist to NULL after unreffing it to avoid a double-unref in the dispose

13 years agohlsdemux: Do not lock the mutex of the fetcher during EOS
Youness Alaoui [Thu, 8 Sep 2011 18:25:00 +0000 (18:25 +0000)]
hlsdemux: Do not lock the mutex of the fetcher during EOS

The fetch_location might call stop_fetcher which could generate an EOS
and cause a deadlock

13 years agospandsp: Adding spandsp plugin with spanplc element for packet loss concealment
Youness Alaoui [Wed, 7 Sep 2011 02:11:06 +0000 (22:11 -0400)]
spandsp: Adding spandsp plugin with spanplc element for packet loss concealment

13 years agohlsdemux: Do not join the task from the task's thread itself. join it only in the...
Youness Alaoui [Wed, 7 Sep 2011 20:09:12 +0000 (20:09 +0000)]
hlsdemux: Do not join the task from the task's thread itself. join it only in the state change

13 years agodocs: cleanup makefiles
Stefan Sauer [Wed, 7 Sep 2011 13:48:33 +0000 (15:48 +0200)]
docs: cleanup makefiles

Remove commented out parts that we don't need. Remove "the wingo addition" - no
so useful after all. Narrow down file-globs for plugin docs.

13 years agodocs: fix camerabin2 names in sections.txt
Stefan Sauer [Wed, 7 Sep 2011 13:42:37 +0000 (15:42 +0200)]
docs: fix camerabin2 names in sections.txt

13 years agokateparse: fix crap in object macros
Stefan Sauer [Wed, 7 Sep 2011 13:37:53 +0000 (15:37 +0200)]
kateparse: fix crap in object macros

13 years agompegvideoparse: make this build with older glib
Vincent Penquerc'h [Wed, 7 Sep 2011 11:32:48 +0000 (12:32 +0100)]
mpegvideoparse: make this build with older glib

g_list_free_full is a new convenience function.
Replace it with g_list_foreach and g_list_free.

https://bugzilla.gnome.org/show_bug.cgi?id=658450

13 years agovideoparsers: Make the h263 and (new) mpeg video parser autopluggable
Sebastian Dröge [Wed, 7 Sep 2011 11:46:49 +0000 (13:46 +0200)]
videoparsers: Make the h263 and (new) mpeg video parser autopluggable

13 years agoh264parse: Fix IDR position marking when the IDR is split into several NALs
George Kiagiadakis [Mon, 5 Sep 2011 17:02:57 +0000 (20:02 +0300)]
h264parse: Fix IDR position marking when the IDR is split into several NALs

13 years agoh263parse: Add all always present fields to the src template caps
Sebastian Dröge [Wed, 7 Sep 2011 09:27:56 +0000 (11:27 +0200)]
h263parse: Add all always present fields to the src template caps

13 years agodiracparse: Add all fields to the src template caps
Sebastian Dröge [Wed, 7 Sep 2011 09:24:14 +0000 (11:24 +0200)]
diracparse: Add all fields to the src template caps

13 years agovideoparsers: Allow parsed streams to be passed through the parsers
Sebastian Dröge [Wed, 7 Sep 2011 09:22:07 +0000 (11:22 +0200)]
videoparsers: Allow parsed streams to be passed through the parsers

13 years agoh264parse: Put stream-format and alignment into the src template caps
Sebastian Dröge [Wed, 7 Sep 2011 09:17:54 +0000 (11:17 +0200)]
h264parse: Put stream-format and alignment into the src template caps

Also remove the parsed field from the sink template caps.

13 years agoh264parse: Change rank to PRIMARY+1 to have h264parse autoplugged
Sebastian Dröge [Wed, 7 Sep 2011 09:14:36 +0000 (11:14 +0200)]
h264parse: Change rank to PRIMARY+1 to have h264parse autoplugged

13 years agodiracparse: Constify some arrays
Sebastian Dröge [Wed, 7 Sep 2011 09:06:46 +0000 (11:06 +0200)]
diracparse: Constify some arrays

13 years agocamerabin2: preview: No need for appsink to store the last buffer
Thiago Santos [Tue, 6 Sep 2011 13:49:04 +0000 (10:49 -0300)]
camerabin2: preview: No need for appsink to store the last buffer

There is no need for preview's appsink to keep a reference to the last buffer,
so disable the property to make it unref the buffers sooner.

13 years agoUpdate for latest new plugins and moved plugins
Christian Fredrik Kalager Schaller [Tue, 6 Sep 2011 21:48:30 +0000 (22:48 +0100)]
Update for latest new plugins and moved plugins

13 years agoAutomatic update of common submodule
Stefan Sauer [Tue, 6 Sep 2011 19:25:00 +0000 (21:25 +0200)]
Automatic update of common submodule

From a39eb83 to 11f0cd5

13 years agoAutomatic update of common submodule
Stefan Sauer [Tue, 6 Sep 2011 13:40:46 +0000 (15:40 +0200)]
Automatic update of common submodule

From 605cd9a to a39eb83

13 years agoh264parse: Add Converter to classification
Sebastian Dröge [Tue, 6 Sep 2011 11:20:43 +0000 (13:20 +0200)]
h264parse: Add Converter to classification

This parser can convert between different alignments and stream-formats.

13 years agocamerabin2: preview: Appsink doesn't need to sync
Thiago Santos [Mon, 5 Sep 2011 17:39:20 +0000 (14:39 -0300)]
camerabin2: preview: Appsink doesn't need to sync

Set sync to false on preview's pipeline appsink

13 years agocamerabin2: preview: Remove one of the two colorspace converters
Thiago Santos [Mon, 5 Sep 2011 12:55:53 +0000 (09:55 -0300)]
camerabin2: preview: Remove one of the two colorspace converters

The preview pipeline doesn't need 2 colorspace converters, remove
one to speed up caps negotiation and reduce the delay on getting
the first preview buffer out of the preview pipeline.

It shouldn't cause problems as videoscale and ffmpegcolorspace seems
to handle the same caps, so no conversion should be needed for
videoscale. Additionally, camerabin1 has been working with a similar
pipeline with a single ffmpegcolorspace and no bugs have been open about it
so far.

13 years agocamerabin2: preview: Disable appsrc signals emission
Thiago Santos [Thu, 1 Sep 2011 23:01:09 +0000 (20:01 -0300)]
camerabin2: preview: Disable appsrc signals emission

Camerabin2 doesn't check appsrc's signals, so disable them

13 years agobasevideoenc: do not try to calculate latency from an unknown framerate
Vincent Penquerc'h [Fri, 26 Aug 2011 09:50:38 +0000 (10:50 +0100)]
basevideoenc: do not try to calculate latency from an unknown framerate

It'll divide by zero, and latency is unknown for an unknown framerate.

Fixes an assert in the schroenc test.

https://bugzilla.gnome.org/show_bug.cgi?id=657419

13 years agohlsdemux: If a fetch fails, decrement the sequence number
Youness Alaoui [Sat, 3 Sep 2011 00:36:34 +0000 (00:36 +0000)]
hlsdemux: If a fetch fails, decrement the sequence number

13 years agohlsdemux: make hlsdemux lock/unlock the m3u8 client when it accesses it
Youness Alaoui [Fri, 2 Sep 2011 23:49:38 +0000 (23:49 +0000)]
hlsdemux: make hlsdemux lock/unlock the m3u8 client when it accesses it

13 years agohlsdemux: Add a get_target_duration and get_current_uri api to m3u8_client
Youness Alaoui [Fri, 2 Sep 2011 23:48:22 +0000 (23:48 +0000)]
hlsdemux: Add a get_target_duration and get_current_uri api to m3u8_client

13 years agohlsdemux: Make the m3u8 client thread-safe
Youness Alaoui [Fri, 2 Sep 2011 23:48:03 +0000 (23:48 +0000)]
hlsdemux: Make the m3u8 client thread-safe

13 years agohlsdemux: Reset in PAUSED->READY after chaining up to the parent class
Sebastian Dröge [Fri, 2 Sep 2011 17:44:31 +0000 (19:44 +0200)]
hlsdemux: Reset in PAUSED->READY after chaining up to the parent class

Otherwise there are possible race conditions with the streaming thread.

13 years agodocs: minor fixes
Edward Hervey [Fri, 2 Sep 2011 14:08:59 +0000 (16:08 +0200)]
docs: minor fixes

13 years agogst-libs: Remove play directory. Was unported since 0.8
Edward Hervey [Wed, 31 Aug 2011 08:12:44 +0000 (10:12 +0200)]
gst-libs: Remove play directory. Was unported since 0.8

13 years agodocs: Add more sections to the gst-libs API docs
Edward Hervey [Wed, 31 Aug 2011 08:10:21 +0000 (10:10 +0200)]
docs: Add more sections to the gst-libs API docs

13 years agocodecparsers: h264: fix PPS parser
Gwenole Beauchesne [Mon, 29 Aug 2011 11:52:17 +0000 (13:52 +0200)]
codecparsers: h264: fix PPS parser

Fix transform_8x8_mode_flag and second_chroma_qp_index_offset

13 years agocodecparsers: h264: fix NAL reader initialization.
Gwenole Beauchesne [Mon, 29 Aug 2011 11:52:16 +0000 (13:52 +0200)]
codecparsers: h264: fix NAL reader initialization.

13 years agocodecparsers: h264: handle end-of-seq and end-of-stream NAL gracefully.
Gwenole Beauchesne [Fri, 26 Aug 2011 13:54:51 +0000 (15:54 +0200)]
codecparsers: h264: handle end-of-seq and end-of-stream NAL gracefully.

13 years agocodecparsers: h264: fix ref_pic_list_reordering().
Gwenole Beauchesne [Fri, 26 Aug 2011 14:05:58 +0000 (16:05 +0200)]
codecparsers: h264: fix ref_pic_list_reordering().

13 years agocodecparsers: h264: fix pred_weight_table() parsing.
Gwenole Beauchesne [Tue, 16 Aug 2011 14:19:02 +0000 (16:19 +0200)]
codecparsers: h264: fix pred_weight_table() parsing.

Use 16-bit signed integer values for weight values because valid
ranges are -128 to +128 inclusive.

13 years agocodecparsers: h264: record slice_header() size in bits.
Gwenole Beauchesne [Tue, 16 Aug 2011 13:21:18 +0000 (15:21 +0200)]
codecparsers: h264: record slice_header() size in bits.

13 years agocodecparsers: h264: fix slice_header() parsing.
Gwenole Beauchesne [Tue, 16 Aug 2011 13:14:03 +0000 (15:14 +0200)]
codecparsers: h264: fix slice_header() parsing.

We were not parsing the slice headers until the actual end, we are now
parsing until the end.

13 years agocodecparsers: h264: fix filler-data NAL unit type name.
Gwenole Beauchesne [Fri, 12 Aug 2011 11:19:29 +0000 (13:19 +0200)]
codecparsers: h264: fix filler-data NAL unit type name.

13 years agocodecparsers: add pkgconfig file.
Gwenole Beauchesne [Fri, 12 Aug 2011 08:17:47 +0000 (10:17 +0200)]
codecparsers: add pkgconfig file.

Adds a new pkgconfig file for codecparsers. They don't have
any specific dependency on gst-plugins-bad and they could quite be
independent bitstream parsers.

13 years agocodecparsers: mpeg: add user-data packet start code definition.
Gwenole Beauchesne [Fri, 5 Aug 2011 09:21:05 +0000 (11:21 +0200)]
codecparsers: mpeg: add user-data packet start code definition.

13 years agocodecparsers: mpeg: fix Picture_Coding_Extension parser.
Gwenole Beauchesne [Wed, 3 Aug 2011 15:31:55 +0000 (17:31 +0200)]
codecparsers: mpeg: fix Picture_Coding_Extension parser.

13 years agoh264parse: Port to the new h.264 parsing library
Thibault Saunier [Tue, 2 Aug 2011 10:37:02 +0000 (12:37 +0200)]
h264parse: Port to the new h.264 parsing library

13 years agocodecparsers: h264: Add an h.264 bitstream parsing library
Thibault Saunier [Fri, 29 Jul 2011 08:56:15 +0000 (10:56 +0200)]
codecparsers: h264: Add an h.264 bitstream parsing library

13 years agompegvideoparse: Port to the new mpeg parsing library
Thibault Saunier [Thu, 23 Jun 2011 16:54:43 +0000 (12:54 -0400)]
mpegvideoparse: Port to the new mpeg parsing library

13 years agocodecparsers: mpeg: Add an mpeg video bitstream parsing library
Thibault Saunier [Tue, 19 Jul 2011 11:52:01 +0000 (13:52 +0200)]
codecparsers: mpeg: Add an mpeg video bitstream parsing library

Create a new codecparsers library

13 years agodocs: Build documentations for libraries
Thibault Saunier [Tue, 30 Aug 2011 17:50:41 +0000 (14:50 -0300)]
docs: Build documentations for libraries

13 years agohlsdemux: Remove obsolete FIXME comment
Sebastian Dröge [Fri, 2 Sep 2011 13:00:58 +0000 (15:00 +0200)]
hlsdemux: Remove obsolete FIXME comment

13 years agohlsdemux: Reset the internal state going to/from READY instead of NULL
Sebastian Dröge [Fri, 2 Sep 2011 12:55:45 +0000 (14:55 +0200)]
hlsdemux: Reset the internal state going to/from READY instead of NULL

This theoretically allows reuse of the element when setting it to READY
instead of NULL.

13 years agohlsdemux: Set playlist parsing helper object to NULL after freeing
Sebastian Dröge [Fri, 2 Sep 2011 12:54:56 +0000 (14:54 +0200)]
hlsdemux: Set playlist parsing helper object to NULL after freeing

13 years agohlsdemux: Allow up to 3 consecutive failed downloads before erroring
Youness Alaoui [Thu, 1 Sep 2011 23:18:51 +0000 (23:18 +0000)]
hlsdemux: Allow up to 3 consecutive failed downloads before erroring

In some networks, especiall in 3G, a fragment download or playlist
update may fail. We allow for up to 3 consecutive failures, while using
the rfc's specs for retry delays before considering that there was an
error on the stream.

13 years agohlsdemux: remove unused argument
Youness Alaoui [Thu, 1 Sep 2011 22:46:19 +0000 (22:46 +0000)]
hlsdemux: remove unused argument

13 years agovp8enc: fix drop-frame property
David Schleef [Mon, 22 Aug 2011 03:15:25 +0000 (20:15 -0700)]
vp8enc: fix drop-frame property

Fixes #656929.

13 years agocamerabin2: Add a flag to disable image conversion elements
Robert Swain [Thu, 1 Sep 2011 13:57:50 +0000 (15:57 +0200)]
camerabin2: Add a flag to disable image conversion elements

If we know that our camera source element produces buffers at the same
resolution and appropriate colourspace for the output, we don't need any
of the generic conversion elements in encodebin. This reduces caps
negotiation overheads among other things.

13 years agomms: fix warnings related to the format on the debug lines.
Josep Torra [Wed, 31 Aug 2011 21:06:53 +0000 (23:06 +0200)]
mms: fix warnings related to the format on the debug lines.

13 years agocamerabin2: Micro optimize encodebin pads linking
Thiago Santos [Wed, 31 Aug 2011 16:29:17 +0000 (13:29 -0300)]
camerabin2: Micro optimize encodebin pads linking

Reduce one check in encodebin pads linking on camerabin2 by
using the link_pads_full version

13 years agobasecamerabinsrc: Optimize preview pipeline startup
Thiago Santos [Wed, 31 Aug 2011 16:28:25 +0000 (13:28 -0300)]
basecamerabinsrc: Optimize preview pipeline startup

Use gst_element_link_pads_full to optimize preview pipeline startup,
leading to a faster camerabin2 startup as well.

13 years agocamerabin2: Use explicit boolean literals to make gobject happy
Thiago Santos [Wed, 31 Aug 2011 18:33:10 +0000 (15:33 -0300)]
camerabin2: Use explicit boolean literals to make gobject happy

gobject boolean properties are strict and only accept 1 or 0, otherwise
they throw an assertion.

13 years agohlsdemux: Just join the task
Sebastian Dröge [Wed, 31 Aug 2011 12:48:47 +0000 (14:48 +0200)]
hlsdemux: Just join the task

gst_task_join() will already stop the task, no need
to stop it additionally.

13 years agohlsdemux: Join the task instead of just stopping it and not waiting for it to finish
Sebastian Dröge [Wed, 31 Aug 2011 12:47:40 +0000 (14:47 +0200)]
hlsdemux: Join the task instead of just stopping it and not waiting for it to finish

Fixes interesting race conditions that cause crashes in decodebin2
because pads are added/removed from child elements although they
should be in READY state already.

13 years agohlsdemux: Remove unused variable
Sebastian Dröge [Wed, 31 Aug 2011 10:12:59 +0000 (12:12 +0200)]
hlsdemux: Remove unused variable

13 years agohlsdemux: switch to higher/lower bitrate by more than one step at a time
Youness Alaoui [Wed, 31 Aug 2011 03:47:52 +0000 (03:47 +0000)]
hlsdemux: switch to higher/lower bitrate by more than one step at a time

13 years agohlsdemux: remove spamming GST_LOG_OBJECT
Youness Alaoui [Wed, 31 Aug 2011 03:20:48 +0000 (03:20 +0000)]
hlsdemux: remove spamming GST_LOG_OBJECT

13 years agohlsdemux: If paused, do not cache fragments until out of memory error!
Youness Alaoui [Wed, 31 Aug 2011 03:07:48 +0000 (03:07 +0000)]
hlsdemux: If paused, do not cache fragments until out of memory error!

We should stop the update thread in PAUSED state and avoid fetching
new fragments when the queue is not empty. The queue should always be
empty since we push data into a queue. Also, in totem, if we seek and
pause the stream while it's buffering, then the state will stay playing
for some reason, so it's best not to continue fetching fragments forever.

13 years agohlsdemux: start/stop update thread and keep track of status
Youness Alaoui [Wed, 31 Aug 2011 03:01:58 +0000 (03:01 +0000)]
hlsdemux: start/stop update thread and keep track of status

13 years agohlsdemux: Post duration message when we parse the playlist
Youness Alaoui [Wed, 31 Aug 2011 02:05:08 +0000 (02:05 +0000)]
hlsdemux: Post duration message when we parse the playlist

13 years agohlsdemux: Send a flush-stop when switching pads. Fixes A->A/V sync issue.
Youness Alaoui [Wed, 31 Aug 2011 01:08:39 +0000 (01:08 +0000)]
hlsdemux: Send a flush-stop when switching pads. Fixes A->A/V sync issue.

This is to ensure that we reset the accumulate segment on the sinks
so if we start with audio only then switch to audio+video, then both
sinks will have the same segments and will be synchronized.

13 years agohlsdemux: seems there was a typo, free the adapter if it contains data, not if alread...
Youness Alaoui [Fri, 26 Aug 2011 22:48:01 +0000 (22:48 +0000)]
hlsdemux: seems there was a typo, free the adapter if it contains data, not if already empty

13 years agohlsdemux: oups, we need to empty the queue after freeing its content
Youness Alaoui [Fri, 26 Aug 2011 22:47:35 +0000 (22:47 +0000)]
hlsdemux: oups, we need to empty the queue after freeing its content

13 years agotsdemux: Need to flush all streams when we receive a flush-start
Youness Alaoui [Fri, 26 Aug 2011 22:44:08 +0000 (22:44 +0000)]
tsdemux: Need to flush all streams when we receive a flush-start

13 years agohlsdemux: do not broadcast the fetcher_cond on bus error if we're cancelled
Youness Alaoui [Fri, 26 Aug 2011 21:46:06 +0000 (21:46 +0000)]
hlsdemux: do not broadcast the fetcher_cond on bus error if we're cancelled

If we cancel the fetch and call the stop_fetcher, which holds the lock,
when it sets the fetcher's state to NULL, it might send an error
on the bus. In that case, we must ignore it, otherwise it will try
to take the lock and will block forever.

13 years agohlsdemux: Split playslist on \r\n not just \n
Jonas Larsson [Fri, 26 Aug 2011 19:29:06 +0000 (19:29 +0000)]
hlsdemux: Split playslist on \r\n not just \n

13 years agohlsdemux: Do not succeed a download if the fetcher had an error
Youness Alaoui [Fri, 26 Aug 2011 19:10:58 +0000 (19:10 +0000)]
hlsdemux: Do not succeed a download if the fetcher had an error

13 years agohlsdemux: Rename functions that need to be called with a locked mutex
Youness Alaoui [Fri, 26 Aug 2011 19:10:39 +0000 (19:10 +0000)]
hlsdemux: Rename functions that need to be called with a locked mutex

13 years agohlsdemux: We must signal the fetcher_cond while holding its lock
Youness Alaoui [Fri, 26 Aug 2011 19:09:54 +0000 (19:09 +0000)]
hlsdemux: We must signal the fetcher_cond while holding its lock