platform/upstream/gstreamer.git
8 years agortsp-media: Add API to directly configure a clock on the media pipelines
Sebastian Dröge [Wed, 30 Dec 2015 14:31:13 +0000 (16:31 +0200)]
rtsp-media: Add API to directly configure a clock on the media pipelines

8 years agortsp-media: Fix typo in docs gst_rtsp_media_set_latncy() -> latency()
Sebastian Dröge [Wed, 30 Dec 2015 14:43:17 +0000 (16:43 +0200)]
rtsp-media: Fix typo in docs gst_rtsp_media_set_latncy() -> latency()

8 years agortsp-media-factory: Add FIXME for 2.0
Sebastian Dröge [Wed, 30 Dec 2015 14:30:38 +0000 (16:30 +0200)]
rtsp-media-factory: Add FIXME for 2.0

8 years agortsp-stream: Fix indentation
Sebastian Dröge [Wed, 30 Dec 2015 14:29:45 +0000 (16:29 +0200)]
rtsp-stream: Fix indentation

8 years agortsp-media: Do not prepare media after media times out
Sebastian Rasmussen [Tue, 22 Dec 2015 11:08:02 +0000 (12:08 +0100)]
rtsp-media: Do not prepare media after media times out

Deferred calls to start_prepare() can be deferred past the point until
which wait_preroll() and by proxy gst_rtsp_media_get_status() is
prepared to wait. Previously there was no lock and no check for this
situation. This meant that a media could be prepared and unprepared
simultaneously by two different threads. Now a lock is in place and a
suitable check is done.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=759773

8 years agortsp-media: Add property to decide if sending media should be stopped when a client...
Sebastian Dröge [Wed, 9 Dec 2015 16:24:24 +0000 (18:24 +0200)]
rtsp-media: Add property to decide if sending media should be stopped when a client disconnects without TEARDOWN

Without TEARDOWN it might be desireable to keep the media running and continue
sending data to the client, even if the RTSP connection itself is
disconnected.

Only do this for session medias that have only UDP transports. If there's at
least on TCP transport, it will stop working and cause problems when the
connection is disconnected.

https://bugzilla.gnome.org/show_bug.cgi?id=758999

8 years agoBack to development
Sebastian Dröge [Thu, 24 Dec 2015 14:29:33 +0000 (15:29 +0100)]
Back to development

8 years agoRelease 1.7.1
Sebastian Dröge [Thu, 24 Dec 2015 13:54:06 +0000 (14:54 +0100)]
Release 1.7.1

8 years agoconfigure: Make -Bsymbolic check work with clang.
Koop Mast [Sun, 20 Dec 2015 23:43:49 +0000 (00:43 +0100)]
configure: Make -Bsymbolic check work with clang.

Update the -Bsymbolic check with the version glib has. This version
works with clang.

https://bugzilla.gnome.org/show_bug.cgi?id=759713

8 years agortsp-session-pool: Avoid dollar sign ($) in session ids
Olivier Crête [Wed, 18 Nov 2015 03:30:54 +0000 (22:30 -0500)]
rtsp-session-pool: Avoid dollar sign ($) in session ids

Live555 in VLC strips off dollar signs and then gets very confused,
we don't loose too much entropy by just skipping it.

8 years agortsp-server: Add g_autoptr() support to all types
Xavier Claessens [Tue, 10 Nov 2015 19:17:18 +0000 (14:17 -0500)]
rtsp-server: Add g_autoptr() support to all types

https://bugzilla.gnome.org/show_bug.cgi?id=754464

8 years agortsp-stream: fixed valgrind error
Srimanta Panda [Tue, 8 Dec 2015 07:27:20 +0000 (08:27 +0100)]
rtsp-stream: fixed valgrind error

Fixed the valgrind error in unit test. The UDP source created during
gst_rtsp_stream_join_bin() was not released while destroying the rtp
bin.

https://bugzilla.gnome.org/show_bug.cgi?id=759010

8 years agoAutomatic update of common submodule
Nicolas Dufresne [Mon, 7 Dec 2015 14:11:35 +0000 (09:11 -0500)]
Automatic update of common submodule

From b319909 to 86e4663

8 years agortsp-client: suspend media during setup request
Srimanta Panda [Wed, 18 Nov 2015 10:14:39 +0000 (11:14 +0100)]
rtsp-client: suspend media during setup request

SETUP request from clients needs to suspend the media to clear the
prerolled buffers. Otherwise it will not affect the prerolled buffer
and the prerolled buffers will be incorrect (for example block-size
from setup request will not affect the prerolled buffer unless the
media is suspended).

https://bugzilla.gnome.org/show_bug.cgi?id=758268

8 years agortsp-stream: create stream pipeline based on transport
Srimanta Panda [Fri, 4 Dec 2015 07:01:37 +0000 (08:01 +0100)]
rtsp-stream: create stream pipeline based on transport

Based on the protocol, create the rtsp stream pipeline. If only TCP or
only UDP is set as the transport protocol, it will not add the extra tee
or queue element to the pipeline. Both these elements will be added, if
it supports both TCP and UDP protocols. This improves the pipeline
performance when one protocol is present.

https://bugzilla.gnome.org/show_bug.cgi?id=758179

8 years agortsp-stream: Only create RTP sending/receiving rtpbin pads if needed
Sebastian Dröge [Thu, 19 Nov 2015 13:01:16 +0000 (15:01 +0200)]
rtsp-stream: Only create RTP sending/receiving rtpbin pads if needed

Adding them when not needed will start some logic inside rtpbin that might be
problematic. Also if e.g. for a sender media we suddenly receive RTP data, we
would start up a rtpjitterbuffer and behave in weird ways.

We still set up the UDP sources for RTP receiving for a sender media to be
able to receive any packets sent by the client for NAT traversal. They will
all go to a fakesink though.

Having an rtpjitterbuffer in the media pipeline will cause the pipeline to be
NO_PREROLL, which will cause deadlocks when seeking the media as it will never
receive ASYNC_DONE after a seek.

https://bugzilla.gnome.org/show_bug.cgi?id=758319

8 years agortsp-stream: Disable multicast loopback for the multicast udp sources too
Sebastian Dröge [Tue, 17 Nov 2015 10:44:38 +0000 (12:44 +0200)]
rtsp-stream: Disable multicast loopback for the multicast udp sources too

On POSIX this setting is for sender sockets, on Windows for receiver sockets.
Previously we were only setting this for sender sockets, which caused looped
back packets to be received on Windows if a multicast transport was used.

8 years agoexamples: Actually use the provided port in the record examples
Jan Schmidt [Mon, 16 Nov 2015 14:12:28 +0000 (01:12 +1100)]
examples: Actually use the provided port in the record examples

8 years agotest-record-auth: Add the option to build in TLS support
Jan Schmidt [Mon, 16 Nov 2015 14:12:28 +0000 (01:12 +1100)]
test-record-auth: Add the option to build in TLS support

8 years agotest-auth: Use an 'anonymous' user for unauthenticated default
Jan Schmidt [Mon, 16 Nov 2015 14:12:28 +0000 (01:12 +1100)]
test-auth: Use an 'anonymous' user for unauthenticated default

There's a comment on one of the resources that 'user' and 'admin'
shouldn't even be able to see it, but they can if the default
token is 'admin2', since that gives them access anyway.

8 years agoAdd test-record-auth example
Jan Schmidt [Mon, 16 Nov 2015 14:12:28 +0000 (01:12 +1100)]
Add test-record-auth example

8 years agortsp-client: Report RECORD and ANNOUNCE as supported in the OPTIONS
Jan Schmidt [Mon, 16 Nov 2015 14:12:28 +0000 (01:12 +1100)]
rtsp-client: Report RECORD and ANNOUNCE as supported in the OPTIONS

8 years agortsp-server: Change the logic so we don't pop a NULL context
Marcus Prebble [Wed, 11 Nov 2015 13:58:33 +0000 (14:58 +0100)]
rtsp-server: Change the logic so we don't pop a NULL context

When doing a port scan (e.g. with nmap) the call to GST_RTSP_CHECK()
will sometimes fail. This call is made before any context is pushed
resulting in an attempt to pop a NULL context.

https://bugzilla.gnome.org/show_bug.cgi?id=757949

8 years agortspserver: Add udp-mcast transport SETUP test
David Svensson Fors [Thu, 22 Oct 2015 12:32:30 +0000 (14:32 +0200)]
rtspserver: Add udp-mcast transport SETUP test

Refactor utility functions in the test file so they can handle
more than UDP and TCP as lower transport.

https://bugzilla.gnome.org/show_bug.cgi?id=756969

8 years agortsp-stream: Always unref return value of gst_object_get_parent()
David Svensson Fors [Thu, 22 Oct 2015 07:15:21 +0000 (09:15 +0200)]
rtsp-stream: Always unref return value of gst_object_get_parent()

Fixes a leak of a GstBin in the udp-mcast case.

https://bugzilla.gnome.org/show_bug.cgi?id=756968

8 years agoAutomatic update of common submodule
Tim-Philipp Müller [Wed, 21 Oct 2015 13:37:19 +0000 (14:37 +0100)]
Automatic update of common submodule

From b99800a to b319909

8 years agoUse new GST_ENABLE_EXTRA_CHECKS #define
Sebastian Dröge [Tue, 20 Oct 2015 14:29:42 +0000 (17:29 +0300)]
Use new GST_ENABLE_EXTRA_CHECKS #define

https://bugzilla.gnome.org/show_bug.cgi?id=756870

8 years agoAutomatic update of common submodule
Sebastian Dröge [Wed, 21 Oct 2015 11:28:47 +0000 (14:28 +0300)]
Automatic update of common submodule

From 6babecd to b99800a

8 years agoUpdate GLib dependency to 2.40.0
Sebastian Dröge [Fri, 2 Oct 2015 19:25:47 +0000 (22:25 +0300)]
Update GLib dependency to 2.40.0

8 years agostream: listen to sender ssrc signals
Hyunjun Ko [Fri, 2 Oct 2015 07:11:05 +0000 (16:11 +0900)]
stream: listen to sender ssrc signals

https://bugzilla.gnome.org/show_bug.cgi?id=746747

8 years agocommon: update for new suppression
Tim-Philipp Müller [Tue, 29 Sep 2015 12:00:51 +0000 (13:00 +0100)]
common: update for new suppression

Makes check-valgrind pass with glib 2.46

8 years agortsp-media: Take reference to media that will be prepared
Sebastian Rasmussen [Mon, 28 Sep 2015 15:40:59 +0000 (17:40 +0200)]
rtsp-media: Take reference to media that will be prepared

default_prepare() takes a transfer-none reference GstRTSPMedia object.
Later on a g_idle_source_new() is created and a pointer to the media
object is passed as user data. If the media is freed before the idle
source is dispatched the media object pointer is invalid, but the idle
source callback expects it to still be valid. To fix this a reference to
the media object is taken when registering the source callback function
and a corresponding release of the reference is done when the souce is
destroyed.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=755748

8 years agortsp-server: Fix memory leaks when context parse fails
Vineeth TM [Thu, 20 Aug 2015 08:01:24 +0000 (17:01 +0900)]
rtsp-server: Fix memory leaks when context parse fails

When g_option_context_parse fails, context and error variables are not getting free'd
which results in memory leaks. Free'ing the same.

And replacing g_error_free with g_clear_error, which checks if the error being passed
is not NULL and sets the variable to NULL on free'ing.

https://bugzilla.gnome.org/show_bug.cgi?id=753863

8 years agoBack to development
Sebastian Dröge [Fri, 25 Sep 2015 21:51:17 +0000 (23:51 +0200)]
Back to development

8 years agoRelease 1.6.0
Sebastian Dröge [Fri, 25 Sep 2015 21:32:52 +0000 (23:32 +0200)]
Release 1.6.0

8 years agoRelease 1.5.91
Sebastian Dröge [Fri, 18 Sep 2015 18:12:06 +0000 (20:12 +0200)]
Release 1.5.91

8 years agostream: fix docs for recently-added get/set_buffer_size API
Tim-Philipp Müller [Thu, 17 Sep 2015 19:07:34 +0000 (20:07 +0100)]
stream: fix docs for recently-added get/set_buffer_size API

https://bugzilla.gnome.org/show_bug.cgi?id=749095

8 years agortsp-media: Don't crash on encrypted RTX SDP
Jan Schmidt [Fri, 4 Sep 2015 01:23:43 +0000 (11:23 +1000)]
rtsp-media: Don't crash on encrypted RTX SDP

In parse_keymgmt(), don't mutate the input string that's been passed
as const, especially since we might need the original value again if
the same key info applies to multiple streams (RTX, for example).

https://bugzilla.gnome.org/show_bug.cgi?id=754753

8 years agotest-mp4: Support filenames with spaces in them. Error out on too few arguments
Jan Schmidt [Sat, 22 Aug 2015 10:59:40 +0000 (20:59 +1000)]
test-mp4: Support filenames with spaces in them. Error out on too few arguments

8 years agotest-record: Check parameter count and print out help
Jan Schmidt [Sun, 16 Aug 2015 16:36:31 +0000 (02:36 +1000)]
test-record: Check parameter count and print out help

If no launch pipeline was supplied, print out some help

8 years agortsp-stream: Implement UDP buffer size setting.
Jan Schmidt [Mon, 31 Aug 2015 12:48:34 +0000 (22:48 +1000)]
rtsp-stream: Implement UDP buffer size setting.

Add gst_rtsp_stream_(get|set)_buffer_size and use it to configure the
UDP TX buffer size.

Incorporates a patch by Hyunjun Ko <zzoon.ko@samsung.com>
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=749095

8 years agortsp-media: Fix small typo causing gtk-doc to complain
Jan Schmidt [Mon, 31 Aug 2015 12:47:45 +0000 (22:47 +1000)]
rtsp-media: Fix small typo causing gtk-doc to complain

8 years agoRelease 1.5.90
Sebastian Dröge [Wed, 19 Aug 2015 11:15:23 +0000 (14:15 +0300)]
Release 1.5.90

8 years agomedia-factory: get port number through gst_rtsp_url_get_port
Hyunjun Ko [Wed, 12 Aug 2015 05:33:44 +0000 (14:33 +0900)]
media-factory: get port number through gst_rtsp_url_get_port

https://bugzilla.gnome.org/show_bug.cgi?id=753473

8 years agomedia-test: Removing unnecessary assertion
Francisco Velazquez [Thu, 13 Aug 2015 09:24:10 +0000 (11:24 +0200)]
media-test: Removing unnecessary assertion

https://bugzilla.gnome.org/show_bug.cgi?id=753385

8 years agoDocument that source keeps a ref on server until it's destroyed
Xavier Claessens [Thu, 23 Jul 2015 18:50:30 +0000 (14:50 -0400)]
Document that source keeps a ref on server until it's destroyed

https://bugzilla.gnome.org/show_bug.cgi?id=749227

8 years agomedia-test: Test for multiple dynamic payload
Nicolas Dufresne [Sat, 8 Aug 2015 15:09:57 +0000 (11:09 -0400)]
media-test: Test for multiple dynamic payload

https://bugzilla.gnome.org/show_bug.cgi?id=753385

8 years agomedia: Only add fakesink once per pipeline
Nicolas Dufresne [Sat, 8 Aug 2015 13:40:09 +0000 (09:40 -0400)]
media: Only add fakesink once per pipeline

The intention is to prevent going PLAYING state before pads are created.
If there was mutilple dynamic payload, it would leak few fakesink and
actually prevent from ever reaching playing state.

https://bugzilla.gnome.org/show_bug.cgi?id=753385

8 years agoRevert "rtsp-media: Only add 1 fakesink per pipeline"
Nicolas Dufresne [Sat, 8 Aug 2015 13:08:37 +0000 (09:08 -0400)]
Revert "rtsp-media: Only add 1 fakesink per pipeline"

This reverts commit 22bf61f16c1210bb458fc3f53642179a0211104f.

8 years agortsp-media: Only add 1 fakesink per pipeline
Nicolas Dufresne [Fri, 7 Aug 2015 13:21:36 +0000 (09:21 -0400)]
rtsp-media: Only add 1 fakesink per pipeline

There should be only one fakesink per pipeline, not per dynpay. This
would lead to element naming clash.

8 years agortsp-media: assertion error due to wrong condition check
Vineeth TM [Thu, 30 Jul 2015 06:32:43 +0000 (15:32 +0900)]
rtsp-media: assertion error due to wrong condition check

In media to caps function, reserved_keys array is being used for variable i,
leading to GLib-CRITICAL **: g_ascii_strcasecmp: assertion 's1 != NULL' failed
changed it to variable j

https://bugzilla.gnome.org/show_bug.cgi?id=753009

8 years agortsp-media: Strip keys from the fmtp that we use internally in our caps
Sebastian Dröge [Wed, 29 Jul 2015 10:27:05 +0000 (11:27 +0100)]
rtsp-media: Strip keys from the fmtp that we use internally in our caps

Skip keys from the fmtp, which we already use ourselves for the
caps. Some software is adding random things like clock-rate into
the fmtp, and we would otherwise here set a string-typed clock-rate
in the caps... and thus fail to create valid RTP caps

https://bugzilla.gnome.org/show_bug.cgi?id=753009

8 years agothreadpool: Fix possible warning in gst_rtsp_thread_pool_cleanup()
Xavier Claessens [Mon, 20 Jul 2015 20:37:44 +0000 (16:37 -0400)]
threadpool: Fix possible warning in gst_rtsp_thread_pool_cleanup()

https://bugzilla.gnome.org/show_bug.cgi?id=752640

8 years agoAutomatic update of common submodule
Stefan Sauer [Fri, 3 Jul 2015 20:00:00 +0000 (22:00 +0200)]
Automatic update of common submodule

From f74b2df to 9aed1d7

8 years agoBack to development
Sebastian Dröge [Wed, 24 Jun 2015 22:04:28 +0000 (00:04 +0200)]
Back to development

8 years agoRelease 1.5.2
Sebastian Dröge [Wed, 24 Jun 2015 21:44:37 +0000 (23:44 +0200)]
Release 1.5.2

8 years agortsp-client: allow application to decide what requirements are supported
Ognyan Tonchev [Thu, 18 Jun 2015 11:12:04 +0000 (13:12 +0200)]
rtsp-client: allow application to decide what requirements are supported

Add "check-requirements" signal and vfunc to allow application
(and subclasses) to check the requirements.

Based on patch from Hyunjun Ko <zzoon.ko@samsung.com>

https://bugzilla.gnome.org/show_bug.cgi?id=749417

8 years agoAutomatic update of common submodule
Nicolas Dufresne [Tue, 16 Jun 2015 21:50:26 +0000 (17:50 -0400)]
Automatic update of common submodule

From 6015d26 to f74b2df

8 years agortsp-media: Always use real payloader when creating streams
Ognyan Tonchev [Thu, 11 Jun 2015 15:39:00 +0000 (17:39 +0200)]
rtsp-media: Always use real payloader when creating streams

A bin that contains the real payloader might be used as payloader. In this
case we have to get the real payloader for the various properties it provides.

Example use cases for this are bins that payload some media and then have
additional elements that add metadata or RTP extension headers to the stream.

https://bugzilla.gnome.org/show_bug.cgi?id=750800

9 years agotest-netclock: Use gst_pipeline_set_latency() to set a high-enough, equal latency...
Sebastian Dröge [Sat, 13 Jun 2015 15:14:43 +0000 (17:14 +0200)]
test-netclock: Use gst_pipeline_set_latency() to set a high-enough, equal latency for all receivers

9 years agotest-netclock: Use new ntp-time-source property on rtpbin
Sebastian Dröge [Fri, 12 Jun 2015 21:35:32 +0000 (23:35 +0200)]
test-netclock: Use new ntp-time-source property on rtpbin

Select the clock time to be used as NTP time source. This allows proper
synchronization between receivers, independent of sharing base times, and just
requires them to use the same clock.

9 years agotest-netclock: Setting the same base time on sender and receiver is not necessary
Sebastian Dröge [Thu, 11 Jun 2015 18:41:31 +0000 (20:41 +0200)]
test-netclock: Setting the same base time on sender and receiver is not necessary

It's going to be fixed up by rtpbin when using ntp-sync=TRUE

9 years agortsp-stream: add description for gst_rtsp_stream_request_aux_sender
Hyunjun Ko [Thu, 11 Jun 2015 08:38:52 +0000 (17:38 +0900)]
rtsp-stream: add description for gst_rtsp_stream_request_aux_sender

https://bugzilla.gnome.org/show_bug.cgi?id=750764

9 years agodocs: add missing types
Hyunjun Ko [Thu, 11 Jun 2015 09:10:12 +0000 (18:10 +0900)]
docs: add missing types

https://bugzilla.gnome.org/show_bug.cgi?id=750764

9 years agodocs: add missing apis
Hyunjun Ko [Thu, 11 Jun 2015 08:37:25 +0000 (17:37 +0900)]
docs: add missing apis

https://bugzilla.gnome.org/show_bug.cgi?id=750764

9 years agotest-netclock-client: Use ntp-sync=TRUE and buffer-mode=SYNC for proper synchronization
Sebastian Dröge [Wed, 10 Jun 2015 15:14:18 +0000 (17:14 +0200)]
test-netclock-client: Use ntp-sync=TRUE and buffer-mode=SYNC for proper synchronization

9 years agoGstRTSPAuth: Add client certificate authentication support
Xavier Claessens [Sat, 6 Jun 2015 02:35:39 +0000 (22:35 -0400)]
GstRTSPAuth: Add client certificate authentication support

https://bugzilla.gnome.org/show_bug.cgi?id=750471

9 years agotest-netclock-client: Use new GstClock API to wait for clock synchronization
Sebastian Dröge [Tue, 9 Jun 2015 11:53:47 +0000 (13:53 +0200)]
test-netclock-client: Use new GstClock API to wait for clock synchronization

9 years agotest-netclock-client: Use a GMainLoop and playbin's source-setup signal
Sebastian Dröge [Tue, 9 Jun 2015 11:51:02 +0000 (13:51 +0200)]
test-netclock-client: Use a GMainLoop and playbin's source-setup signal

A mainloop is needed to get glimagesink to display something on OSX, and
the source-setup signal just makes things a little bit easier.

9 years agoAutomatic update of common submodule
Edward Hervey [Tue, 9 Jun 2015 09:30:54 +0000 (11:30 +0200)]
Automatic update of common submodule

From d9a3353 to 6015d26

9 years agoAutomatic update of common submodule
Stefan Sauer [Mon, 8 Jun 2015 21:08:34 +0000 (23:08 +0200)]
Automatic update of common submodule

From d37af32 to d9a3353

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 21:07:31 +0000 (23:07 +0200)]
Automatic update of common submodule

From 21ba2e5 to d37af32

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 15:32:29 +0000 (17:32 +0200)]
Automatic update of common submodule

From c408583 to 21ba2e5

9 years agodocs: remove variables that we define in the snippet from common
Stefan Sauer [Sun, 7 Jun 2015 15:06:40 +0000 (17:06 +0200)]
docs: remove variables that we define in the snippet from common

This is syncing our Makefile.am with upstream gtkdoc.

9 years agoAutomatic update of common submodule
Stefan Sauer [Sun, 7 Jun 2015 15:16:47 +0000 (17:16 +0200)]
Automatic update of common submodule

From 44a3517 to c408583

9 years agoBack to development
Sebastian Dröge [Sun, 7 Jun 2015 14:44:55 +0000 (16:44 +0200)]
Back to development

9 years agoRelease 1.5.1
Sebastian Dröge [Sun, 7 Jun 2015 09:20:01 +0000 (11:20 +0200)]
Release 1.5.1

9 years agortsp-client: No flush during Teardown.
Göran Jönsson [Mon, 25 May 2015 14:36:18 +0000 (16:36 +0200)]
rtsp-client: No flush during Teardown.

When calling gst_rtsp_watch_write_data in gstrtspconnection.c and
backlog is empty it can happen that just a part of a message will be
sent and rest is in backlog queue. If then flush during teardown
just a part of message will be sent.This can lead to client miss
teardown response since it expect to get the last part of message.

The flushing during teardown was introduced to fix a deadlock that now
is fixed more generally in handle_request by temporary  setting backlog
size to unlimited.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=749845

9 years agotests: Use AM_TESTS_ENVIRONMENT
Tim-Philipp Müller [Wed, 27 May 2015 16:04:41 +0000 (17:04 +0100)]
tests: Use AM_TESTS_ENVIRONMENT

Needed by the new automake test runner and the
current version of the common submodule.

9 years agortsp-server: Use single-include rtsp header to make sure we get all definitions
Sebastian Dröge [Wed, 20 May 2015 14:05:47 +0000 (17:05 +0300)]
rtsp-server: Use single-include rtsp header to make sure we get all definitions

9 years agortsp-media: Mark some more functions static
Sebastian Dröge [Tue, 5 May 2015 14:46:57 +0000 (16:46 +0200)]
rtsp-media: Mark some more functions static

9 years agortsp-media: Only unblock the media in suspend() when actually changing the state
Sebastian Dröge [Tue, 5 May 2015 14:46:19 +0000 (16:46 +0200)]
rtsp-media: Only unblock the media in suspend() when actually changing the state

Otherwise we're going to lose a few packets for live streams during DESCRIBE.

9 years agoexamples: Use AVPF profile for the RTX example
Sebastian Dröge [Mon, 4 May 2015 14:33:08 +0000 (16:33 +0200)]
examples: Use AVPF profile for the RTX example

9 years agortsp-sdp: Only add RTX to the SDP when using a feedback profile
Sebastian Dröge [Mon, 4 May 2015 14:31:20 +0000 (16:31 +0200)]
rtsp-sdp: Only add RTX to the SDP when using a feedback profile

9 years agortsp-stream: get valid clock-rate from last-sample
Hyunjun Ko [Mon, 27 Apr 2015 10:35:53 +0000 (19:35 +0900)]
rtsp-stream: get valid clock-rate from last-sample

clock-rate in last-sample's caps is integer, not unsigned.
To get this value properly, variable needs to be type-casted to int.

https://bugzilla.gnome.org/show_bug.cgi?id=747614

9 years agoautogen.sh: only run autopoint if gettext requested in configure.ac
Tim-Philipp Müller [Sun, 26 Apr 2015 14:00:05 +0000 (15:00 +0100)]
autogen.sh: only run autopoint if gettext requested in configure.ac

Not just because there happens to be a po directory.

https://bugzilla.gnome.org/show_bug.cgi?id=748058

9 years agoRevert "configure.ac: uncomment gettext version setup"
Tim-Philipp Müller [Sun, 26 Apr 2015 13:58:49 +0000 (14:58 +0100)]
Revert "configure.ac: uncomment gettext version setup"

This reverts commit 1545d8fef7065081079172ec264a0061039ac075.

We don't need a gettext setup here and there's no po
directory either, so no reason why autopoint would be
run in the first place.

See https://bugzilla.gnome.org/show_bug.cgi?id=748058

9 years agoFix timeout function signatures across tests and examples
Alistair Buxton [Thu, 23 Apr 2015 17:53:08 +0000 (18:53 +0100)]
Fix timeout function signatures across tests and examples

9 years agotests: define GST_CHECK_TEST_ENVIRONMENT_BEACON
Tim-Philipp Müller [Thu, 23 Apr 2015 16:27:40 +0000 (17:27 +0100)]
tests: define GST_CHECK_TEST_ENVIRONMENT_BEACON

Make sure the test environment is set up.

https://bugzilla.gnome.org//show_bug.cgi?id=747624

9 years agoconfigure: bump automake requirement to 1.14 and autoconf to 2.69
Tim-Philipp Müller [Thu, 23 Apr 2015 16:22:59 +0000 (17:22 +0100)]
configure: bump automake requirement to 1.14 and autoconf to 2.69

This is only required for builds from git, people can still
build tarballs if they only have older autotools.

https://bugzilla.gnome.org//show_bug.cgi?id=747624

9 years agoconfigure.ac: uncomment gettext version setup
Vincent Penquerc'h [Mon, 20 Apr 2015 07:49:57 +0000 (08:49 +0100)]
configure.ac: uncomment gettext version setup

Fixes autogen.sh. It would run autopoint, which would complain
that it could not find the gettext version in configure.ac.

https://bugzilla.gnome.org/show_bug.cgi?id=748058

9 years agotest-video-rtx: set exact payload type to PCMA payloader
Hyunjun Ko [Wed, 15 Apr 2015 01:06:30 +0000 (10:06 +0900)]
test-video-rtx: set exact payload type to PCMA payloader

Setting wrong payload type causes failure to do retransmission through audio stream

https://bugzilla.gnome.org/show_bug.cgi?id=747839

9 years agortsp-stream: fix to get valid each stream data for request-aux-sender signal
Hyunjun Ko [Wed, 15 Apr 2015 00:45:23 +0000 (09:45 +0900)]
rtsp-stream: fix to get valid each stream data for request-aux-sender signal

Because of duplicated g_signal_connect for request-aux-sender signal,
wrong stream pointer is passed to the signal handler.
Instead of passing each stream, pass stream array and get the relevant stream.

https://bugzilla.gnome.org/show_bug.cgi?id=747839

9 years agoUpdate autogen.sh to latest version from common
Tim-Philipp Müller [Mon, 6 Apr 2015 09:32:52 +0000 (10:32 +0100)]
Update autogen.sh to latest version from common

Fixes build after aclocal_check etc. helpers have been removed.

9 years agoAutomatic update of common submodule
Tim-Philipp Müller [Fri, 3 Apr 2015 17:58:26 +0000 (18:58 +0100)]
Automatic update of common submodule

From bc76a8b to c8fb372

9 years agortsp-stream: Limit the queues to 1 buffer
Sebastian Dröge [Mon, 23 Mar 2015 20:03:20 +0000 (21:03 +0100)]
rtsp-stream: Limit the queues to 1 buffer

We only need them to be able to pre-roll, queueing up more data here
is only going to harm latency and memory usage.

9 years agortsp-stream: Update comment and ASCII art to the latest code
Sebastian Dröge [Mon, 23 Mar 2015 19:59:52 +0000 (20:59 +0100)]
rtsp-stream: Update comment and ASCII art to the latest code

We have a queue in front of the udpsink too to prevent the pipeline from
locking up.

9 years agortsp-media: Properly return first rtptime
Nicolas Dufresne [Sat, 21 Mar 2015 15:04:05 +0000 (11:04 -0400)]
rtsp-media: Properly return first rtptime

Instead we where returning first GstBuffer timestamp. This would result
in clock skew and unwanted behaviour in RTSP playback.

https://bugzilla.gnome.org/show_bug.cgi?id=746479

9 years agortsp-stream: Don't leave buffer mapped
Nicolas Dufresne [Wed, 18 Mar 2015 20:44:19 +0000 (16:44 -0400)]
rtsp-stream: Don't leave buffer mapped

If the seq is NULL, the RTP buffer was left mapped. We should always
unmap the buffer.

9 years agoFix typo in README
Sebastian Dröge [Sun, 15 Mar 2015 12:27:39 +0000 (12:27 +0000)]
Fix typo in README