platform/upstream/gstreamer.git
7 years agortsp-client: Fix handling of keep-alive GET_PARAMETER/SET_PARAMETER
Sebastian Dröge [Thu, 19 Jan 2017 12:24:07 +0000 (14:24 +0200)]
rtsp-client: Fix handling of keep-alive GET_PARAMETER/SET_PARAMETER

While they logically have 0 bytes length, GstRTSPConnection is appending
a '\0' to everything making the size be 1 instead.

7 years agomeson: bump version
Tim-Philipp Müller [Fri, 13 Jan 2017 12:39:36 +0000 (12:39 +0000)]
meson: bump version

7 years agortsp-session: Only remove deprecated API if requested to do so, not just when disabling
Sebastian Dröge [Thu, 12 Jan 2017 17:04:23 +0000 (19:04 +0200)]
rtsp-session: Only remove deprecated API if requested to do so, not just when disabling

gst_rtsp_session_is_expired() and gst_rtsp_session_next_timeout() were
affected.

7 years agoBack to development
Sebastian Dröge [Thu, 12 Jan 2017 14:32:59 +0000 (16:32 +0200)]
Back to development

7 years agoRelease 1.11.1
Sebastian Dröge [Thu, 12 Jan 2017 14:14:46 +0000 (16:14 +0200)]
Release 1.11.1

7 years agortsp-stream: corrected if-statement in _get_server_port()
Patricia Muscalu [Tue, 10 Jan 2017 07:34:50 +0000 (08:34 +0100)]
rtsp-stream: corrected if-statement in _get_server_port()

This bug was accidentally introduced while fixing a segfault
in _get_server_port() function.

https://bugzilla.gnome.org/show_bug.cgi?id=776345

7 years agortsp-stream: fixed segmenation fault in _get_server_port()
Patricia Muscalu [Mon, 9 Jan 2017 13:12:05 +0000 (14:12 +0100)]
rtsp-stream: fixed segmenation fault in _get_server_port()

Calling function gst_rtsp_stream_get_server_port() results in
segmenation fault in the RTP/RTSP/TCP case.
Port that the server will use to receive RTCP makes only
sense in the UDP case, however the function should handle
the TCP case in a nicer way.

https://bugzilla.gnome.org/show_bug.cgi?id=776345

7 years agodosc: Fix a little typo
Aleksandr Slobodeniuk [Mon, 9 Jan 2017 09:22:40 +0000 (12:22 +0300)]
dosc: Fix a little typo

https://bugzilla.gnome.org/show_bug.cgi?id=777037

7 years agomeson: generate pkg-config -uninstalled pc files
Guillaume Desmottes [Wed, 4 Jan 2017 15:20:54 +0000 (16:20 +0100)]
meson: generate pkg-config -uninstalled pc files

Generating those files is useful for users building the GStreamer stack
using meson and having to link it to another project which is still
using the autotools.

https://bugzilla.gnome.org/show_bug.cgi?id=776810

7 years agopkgconfig: fix -uninstalled pc file
Guillaume Desmottes [Wed, 4 Jan 2017 15:11:08 +0000 (16:11 +0100)]
pkgconfig: fix -uninstalled pc file

pcfiledir was never defined so the paths were wrong.

https://bugzilla.gnome.org/show_bug.cgi?id=776867

7 years agortsp-stream: Fixed TCP transport case
Patricia Muscalu [Wed, 21 Dec 2016 12:41:50 +0000 (13:41 +0100)]
rtsp-stream: Fixed TCP transport case

Make sure that the appsink element is actually added to
the bin before trying to link it with the elements in it.

https://bugzilla.gnome.org/show_bug.cgi?id=776343

7 years agoRemove generated .spec file
Tim-Philipp Müller [Fri, 16 Dec 2016 17:26:04 +0000 (17:26 +0000)]
Remove generated .spec file

Likely extremely bitrotten, and we should not ship this anyway.

7 years agoAutomatic update of common submodule
Edward Hervey [Sat, 3 Dec 2016 07:21:02 +0000 (08:21 +0100)]
Automatic update of common submodule

From f980fd9 to 39ac2f5

7 years agomedia: Fix pt map caps
Edward Hervey [Fri, 2 Dec 2016 14:40:09 +0000 (15:40 +0100)]
media: Fix pt map caps

Since decryption is handled within rtpbin, all outcoming stream
caps will be application/x-rtp (i.e. regular rtp)

Fixes RECORD with SRTP streams

7 years agomedia-factory: Create media objects with the proper transport mode
Edward Hervey [Fri, 2 Dec 2016 14:38:04 +0000 (15:38 +0100)]
media-factory: Create media objects with the proper transport mode

The function called immediately afterwards (collect_streams()) will
need it to work properly

7 years agortsp-auth: Don't remove digest-auth nonces that already/still have a client connected
Sebastian Dröge [Fri, 2 Dec 2016 12:36:50 +0000 (14:36 +0200)]
rtsp-auth: Don't remove digest-auth nonces that already/still have a client connected

7 years agortsp-media-factory: Don't create a pipeline for the media pipeline string
Sebastian Dröge [Thu, 1 Dec 2016 16:04:34 +0000 (18:04 +0200)]
rtsp-media-factory: Don't create a pipeline for the media pipeline string

We're going to put a pipeline into a pipeline otherwise, which is not
exactly ideal.

7 years agomedia: Fix race condition around finish_unprepare() if called multiple time
Kseniia Vasilchuk [Tue, 25 Oct 2016 12:41:28 +0000 (15:41 +0300)]
media: Fix race condition around finish_unprepare() if called multiple time

https://bugzilla.gnome.org/show_bug.cgi?id=755329

7 years agortspclientsink: Don't leave stale pointer after unref
Jan Schmidt [Wed, 30 Nov 2016 03:06:36 +0000 (14:06 +1100)]
rtspclientsink: Don't leave stale pointer after unref

Fix a warning on shutdown - don't keep a pointer to an
alread-unreffed object.

7 years agocommon: use https protocol for common submodule
Tim-Philipp Müller [Sat, 26 Nov 2016 11:24:50 +0000 (11:24 +0000)]
common: use https protocol for common submodule

https://bugzilla.gnome.org/show_bug.cgi?id=775110

7 years agostream: block the output of rtpbin instead of the source pipeline
Matthew Waters [Mon, 21 Nov 2016 12:29:56 +0000 (23:29 +1100)]
stream: block the output of rtpbin instead of the source pipeline

85c52e194bcb81928b96614be0ae47d59eccb1ce introduced a more correct
detection of the srtp rollover counter to add to the SDP.

Unfortunately, it was incomplete for live pipelines where the logic
blocks the source bin before creating the SDP and thus would never have
the necessary informaiton to create a correct SDP with srtp encryption.

Move the pad blocks to rtpbin's output pads instead so that the
necessary information can be created before we need the information for
the SDP.

https://bugzilla.gnome.org/show_bug.cgi?id=770239

7 years agortsp-client: add IDLE timeout, before session exists
Dag Gullberg [Mon, 21 Nov 2016 15:02:39 +0000 (16:02 +0100)]
rtsp-client: add IDLE timeout, before session exists

The RTSP server will not timeout an idle RTSP connection
(note this is different from doing timeout on a RTSP
session).

At least for Apache this is a problem when running RTSP over
HTTPS since it uses one of the threads (there is a rather
limited number) that are available for handling requests.

https://bugzilla.gnome.org/show_bug.cgi?id=771830

7 years ago.gitignore more
Tim-Philipp Müller [Wed, 23 Nov 2016 09:45:08 +0000 (09:45 +0000)]
.gitignore more

7 years agortsp-stream: Set close-socket FALSE on UDP src:es
Göran Jönsson [Mon, 21 Nov 2016 12:05:50 +0000 (13:05 +0100)]
rtsp-stream: Set close-socket FALSE on UDP src:es

With this RTSP server can use the sockets independent on the udpsrc
state.
When the udp src is finalized it will unref socket and when g_socket
is finalized the socket will be closed.

https://bugzilla.gnome.org/show_bug.cgi?id=765673

7 years agortspclientsink: Move to new helper function to parse authentication responses
Sebastian Dröge [Fri, 18 Nov 2016 15:47:13 +0000 (17:47 +0200)]
rtspclientsink: Move to new helper function to parse authentication responses

https://bugzilla.gnome.org/show_bug.cgi?id=774416

7 years agortsp-auth: Add support for Digest authentication
Sebastian Dröge [Wed, 16 Nov 2016 06:42:24 +0000 (08:42 +0200)]
rtsp-auth: Add support for Digest authentication

https://bugzilla.gnome.org/show_bug.cgi?id=774416

7 years agoEnable building with MSVC
Scott D Phillips [Thu, 17 Nov 2016 17:41:53 +0000 (09:41 -0800)]
Enable building with MSVC

https://bugzilla.gnome.org/show_bug.cgi?id=774640

7 years agomeson: gstreamer gst_check_dep does not exist on windows
Thibault Saunier [Fri, 18 Nov 2016 23:23:14 +0000 (20:23 -0300)]
meson: gstreamer gst_check_dep does not exist on windows

7 years agoclient: update do_send_message to match type GstRTSPClientSendFunc
Scott D Phillips [Thu, 17 Nov 2016 17:43:37 +0000 (09:43 -0800)]
client: update do_send_message to match type GstRTSPClientSendFunc

This type mismatch fails building with MSVC

https://bugzilla.gnome.org/show_bug.cgi?id=774640

8 years agortsp-sdp: Fix indentation
Sebastian Dröge [Fri, 11 Nov 2016 12:42:08 +0000 (14:42 +0200)]
rtsp-sdp: Fix indentation

8 years agortsp-media: Only signal "new-state" if the state has actually changed
Neha Arora [Thu, 10 Nov 2016 05:16:00 +0000 (05:16 +0000)]
rtsp-media: Only signal "new-state" if the state has actually changed

https://bugzilla.gnome.org/show_bug.cgi?id=774173

8 years agoclient: emit signal in the beginning of each rtsp request
Branko Subasic [Wed, 24 Aug 2016 09:39:13 +0000 (11:39 +0200)]
client: emit signal in the beginning of each rtsp request

These signals let the application validate the requests, configure the
media/stream in a certain way and also generate error status code in
case of error or bad request.

https://bugzilla.gnome.org/show_bug.cgi?id=758062

8 years agomeson: update version
Tim-Philipp Müller [Tue, 1 Nov 2016 18:10:35 +0000 (18:10 +0000)]
meson: update version

8 years agoBack to development
Sebastian Dröge [Tue, 1 Nov 2016 16:53:15 +0000 (18:53 +0200)]
Back to development

8 years agoRelease 1.10.0
Sebastian Dröge [Tue, 1 Nov 2016 16:06:46 +0000 (18:06 +0200)]
Release 1.10.0

8 years agotests: try to avoid using the same ports in different tests
Tim-Philipp Müller [Fri, 28 Oct 2016 17:38:01 +0000 (18:38 +0100)]
tests: try to avoid using the same ports in different tests

Causes problems with client multicast tests otherwise if
tests are run in parallel.

https://bugzilla.gnome.org/show_bug.cgi?id=773640

8 years agotests: client: use fail_unless_equals_foo() for better failure reporting
Tim-Philipp Müller [Fri, 28 Oct 2016 16:50:59 +0000 (17:50 +0100)]
tests: client: use fail_unless_equals_foo() for better failure reporting

8 years agortsp-client: Session filter in unwatch session
Göran Jönsson [Mon, 26 Sep 2016 09:16:04 +0000 (11:16 +0200)]
rtsp-client: Session filter in unwatch session

Call session filter with filter_session_media as paramer in
client_unwatch_session if using drop_backlog = FALSE.

In client_unwatch_session its allowed to grow the watchs backlog.
If using drop_backlog = FALSE and the backlog is full it will cause
a deadlock when setting session media state to NULL
if the backlog is not allowed to grow.

https://bugzilla.gnome.org/show_bug.cgi?id=771983

8 years agomeson: add fallbacks for gst modules
Tim-Philipp Müller [Thu, 20 Oct 2016 20:40:18 +0000 (21:40 +0100)]
meson: add fallbacks for gst modules

For gst-all.

8 years agortsp-client: Fix factory leaking in find_media() in error cases
Nikita Bobkov [Wed, 14 Sep 2016 14:48:39 +0000 (17:48 +0300)]
rtsp-client: Fix factory leaking in find_media() in error cases

https://bugzilla.gnome.org/show_bug.cgi?id=771488

8 years agostream: Fix randomly missing streams from SDP with dynamic elements
Xavier Claessens [Thu, 6 Oct 2016 15:47:50 +0000 (11:47 -0400)]
stream: Fix randomly missing streams from SDP with dynamic elements

When using dynamic elements, gst_rtsp_stream_join_bin() is called from
"pad-added" signal. In that case priv->srcpad could already have its caps,
and they'll be sent to priv->send_src[0] pad. That means that when it
connects "notify::caps" signal, that pad could already have received its
caps and the signal won't be emitted anymore.

In that case priv->caps stay to NULL and when building the SDP that stream
gets ignored. Leading to missing video or audio when playing in client side.

https://bugzilla.gnome.org/show_bug.cgi?id=772478

8 years agomeson: update version
Tim-Philipp Müller [Fri, 30 Sep 2016 10:42:08 +0000 (11:42 +0100)]
meson: update version

8 years agoRelease 1.9.90
Sebastian Dröge [Fri, 30 Sep 2016 10:04:12 +0000 (13:04 +0300)]
Release 1.9.90

8 years agortsp-server: Hint that set_multicast_iface expects the name of the interface
Ian Jamison [Sat, 17 Sep 2016 12:17:19 +0000 (13:17 +0100)]
rtsp-server: Hint that set_multicast_iface expects the name of the interface

To prevent any possibly confusion with IPs or anything else.

https://bugzilla.gnome.org/show_bug.cgi?id=771530

8 years agortsp-media: Call g_free() instead of g_object_unref() on multicast-iface strings
Sebastian Dröge [Sun, 18 Sep 2016 13:58:55 +0000 (09:58 -0400)]
rtsp-media: Call g_free() instead of g_object_unref() on multicast-iface strings

https://bugzilla.gnome.org/show_bug.cgi?id=763000#c5

8 years agoconfigure: Depend on gstreamer 1.9.2.1
Sebastian Dröge [Wed, 14 Sep 2016 09:31:15 +0000 (11:31 +0200)]
configure: Depend on gstreamer 1.9.2.1

8 years agoAutomatic update of common submodule
Jan Schmidt [Sat, 10 Sep 2016 10:52:31 +0000 (20:52 +1000)]
Automatic update of common submodule

From b18d820 to f980fd9

8 years agoAutomatic update of common submodule
Jan Schmidt [Fri, 9 Sep 2016 23:58:31 +0000 (09:58 +1000)]
Automatic update of common submodule

From 6f2d209 to b18d820

8 years agortsp-stream: Remove unused _locked() variant of a function
Sebastian Dröge [Wed, 7 Sep 2016 15:44:34 +0000 (18:44 +0300)]
rtsp-stream: Remove unused _locked() variant of a function

It was added during refactoring.

8 years agostream: cosmetic cleanup
Xavier Claessens [Wed, 7 Sep 2016 14:21:09 +0000 (10:21 -0400)]
stream: cosmetic cleanup

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: Compare IP addresses case insensitive in more places
Xavier Claessens [Wed, 7 Sep 2016 14:16:19 +0000 (10:16 -0400)]
stream: Compare IP addresses case insensitive in more places

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: Fix leaked joined_bin
Xavier Claessens [Wed, 7 Sep 2016 14:12:18 +0000 (10:12 -0400)]
stream: Fix leaked joined_bin

There is no need to keep a strong ref on it, and _leave_bin() was
setting it to NULL before calling g_clear_object() so it was leaked.

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agortsp-stream: Compare IP address strings case insensitive
Sebastian Dröge [Tue, 6 Sep 2016 16:15:23 +0000 (19:15 +0300)]
rtsp-stream: Compare IP address strings case insensitive

Otherwise IPv6 addresses might fail this comparision.

8 years agortsp-stream: Bind multicast sockets to ANY as before
Sebastian Dröge [Tue, 6 Sep 2016 16:10:21 +0000 (19:10 +0300)]
rtsp-stream: Bind multicast sockets to ANY as before

https://bugzilla.gnome.org/show_bug.cgi?id=766612#c48

8 years agortsp-session: Fix segfault when doing keep-alive after removing the session
Kseniia [Mon, 5 Sep 2016 15:31:36 +0000 (18:31 +0300)]
rtsp-session: Fix segfault when doing keep-alive after removing the session

If keep-alive happens after removing the session but before finalizing the
stream transport, we would segfault.

https://bugzilla.gnome.org/show_bug.cgi?id=750544

8 years agortsp-stream: Always create multicast UDP elements if the protocol flag is set
Sebastian Dröge [Mon, 5 Sep 2016 15:04:50 +0000 (18:04 +0300)]
rtsp-stream: Always create multicast UDP elements if the protocol flag is set

Adding them later will cause deadlocks due to
1) pre-rolling and staying in PAUSED with the unicast/TCP sinks
2) adding the multicast sink
3) waiting for it to get data to preroll again

3) never happens because the queues after the tee are full.

8 years agortsp-stream: Fix up various multicast related issues
Sebastian Dröge [Mon, 5 Sep 2016 13:32:57 +0000 (16:32 +0300)]
rtsp-stream: Fix up various multicast related issues

8 years agotests: Fix compilation
Sebastian Dröge [Mon, 5 Sep 2016 10:40:59 +0000 (13:40 +0300)]
tests: Fix compilation

8 years agostream: revert back to create udpsrc/udpsink on DESCRIBE for unicast
Xavier Claessens [Thu, 28 Jul 2016 19:33:05 +0000 (15:33 -0400)]
stream: revert back to create udpsrc/udpsink on DESCRIBE for unicast

This is basically reverting changes introduced in commit f62a9a7,
because it was introducing various regressions:

- It introduces a leak of udpsrc elements that got wrongly fixed by adding
  an hash table in commit cba045e. We should have at most 4 udpsrc for unicast:
  ipv4/ipv6, rtp/rtcp. They can be reused for all unicast clients.
- If a mcast client connects, it creates a new socket in SETUP to try to respect
  the destination/port given by the client in the transport, and overrides the
  socket already set on the udpsink element. That means that if we already had a
  client connected, the source address on the udp packets it receives suddenly
  changes.
- If a 2nd mcast client connects, the destination/port in its transport is
  ignored but its transport wasn't updated.

What this patch does:

- Revert back to create udpsrc/udpsink for unicast clients on DESCRIBE.
- Always have a tee+queue when udp is enabled. This could be optimized
  again in a later patch, but is more complicated. If no unicast clients
  connects then those elements are useless, this could be also optimized
  in a later patch.
- When mcast transport is added, it creates a new set of udpsrc/udpsink,
  seperated from those for unicast clients. Since we already support only
  one mcast address, we also create only one set of elements.

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: factor our plug_src function
Xavier Claessens [Thu, 28 Jul 2016 19:20:31 +0000 (15:20 -0400)]
stream: factor our plug_src function

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: factor out plug_sink function
Xavier Claessens [Fri, 22 Jul 2016 01:46:16 +0000 (21:46 -0400)]
stream: factor out plug_sink function

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: small documentation clarification
Xavier Claessens [Thu, 21 Jul 2016 03:05:09 +0000 (23:05 -0400)]
stream: small documentation clarification

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: rename addr_v4/6 to mcast_addr_v4/6 for clarity
Xavier Claessens [Wed, 20 Jul 2016 19:35:44 +0000 (15:35 -0400)]
stream: rename addr_v4/6 to mcast_addr_v4/6 for clarity

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: Keep a ref on joined bin
Xavier Claessens [Thu, 14 Jul 2016 15:10:31 +0000 (11:10 -0400)]
stream: Keep a ref on joined bin

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: code cleanup
Xavier Claessens [Wed, 20 Jul 2016 19:11:32 +0000 (15:11 -0400)]
stream: code cleanup

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agostream: small fix in error code path
Xavier Claessens [Thu, 21 Jul 2016 03:18:23 +0000 (23:18 -0400)]
stream: small fix in error code path

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agoRevert "rtsp-stream: Fix crash on cleanup with shared media and multiple udpsrc"
Xavier Claessens [Thu, 21 Jul 2016 00:09:57 +0000 (20:09 -0400)]
Revert "rtsp-stream: Fix crash on cleanup with shared media and multiple udpsrc"

This partly reverts commit cba045e1b19fad6e689e10206f57903e15f1229a,
but keeps unit tests.

https://bugzilla.gnome.org/show_bug.cgi?id=766612

8 years agoBack to development
Sebastian Dröge [Thu, 1 Sep 2016 09:33:00 +0000 (12:33 +0300)]
Back to development

8 years agoRelease 1.9.2
Sebastian Dröge [Thu, 1 Sep 2016 09:32:51 +0000 (12:32 +0300)]
Release 1.9.2

8 years agoAdd support for Meson as alternative/parallel build system
Tim-Philipp Müller [Wed, 27 Jan 2016 01:03:52 +0000 (01:03 +0000)]
Add support for Meson as alternative/parallel build system

https://github.com/mesonbuild/meson

8 years agobuild: silence error about pthread for 'make check' in osx
Josep Torra [Fri, 26 Aug 2016 19:56:13 +0000 (21:56 +0200)]
build: silence error about pthread for 'make check' in osx

Fixes "clang: error: argument unused during compilation: '-pthread'"

8 years agortsp-client: Fix leaking of media in error cases
Nikita Bobkov [Fri, 25 Sep 2015 15:04:00 +0000 (15:04 +0000)]
rtsp-client: Fix leaking of media in error cases

With additional fixes by Kseniya Vasilchuk <vasilchukkseniia@gmail.com>
and myself to make the media refcounting a bit easier to follow.

https://bugzilla.gnome.org/show_bug.cgi?id=755632

8 years agortsp-client: Fix leaking of session in error cases
Sebastian Dröge [Tue, 2 Aug 2016 12:08:22 +0000 (15:08 +0300)]
rtsp-client: Fix leaking of session in error cases

https://bugzilla.gnome.org/show_bug.cgi?id=755632

8 years agoAutomatic update of common submodule
Stefan Sauer [Mon, 11 Jul 2016 19:16:04 +0000 (21:16 +0200)]
Automatic update of common submodule

From f363b32 to f49c55e

8 years agoBack to development
Sebastian Dröge [Wed, 6 Jul 2016 10:51:15 +0000 (13:51 +0300)]
Back to development

8 years agoRelease 1.9.1
Sebastian Dröge [Wed, 6 Jul 2016 10:28:12 +0000 (13:28 +0300)]
Release 1.9.1

8 years agoconfigure: Need to add -DGST_STATIC_COMPILATION when building only statically
Nirbheek Chauhan [Thu, 23 Jun 2016 20:32:20 +0000 (02:02 +0530)]
configure: Need to add -DGST_STATIC_COMPILATION when building only statically

https://bugzilla.gnome.org/show_bug.cgi?id=767463

8 years agoAutomatic update of common submodule
Nicolas Dufresne [Tue, 21 Jun 2016 15:49:02 +0000 (11:49 -0400)]
Automatic update of common submodule

From ac2f647 to f363b32

8 years agosdp: add rollover counters for all sender SSRC
Aleix Conchillo Flaqué [Fri, 15 Apr 2016 05:56:11 +0000 (22:56 -0700)]
sdp: add rollover counters for all sender SSRC

We add different crypto sessions in MIKEY, one for each sender
SSRC. Currently, all of them will have the same security policy, 0.

The rollover counters are obtained from the srtpenc element using the
"stats" property.

https://bugzilla.gnome.org/show_bug.cgi?id=730539

8 years agodocs: fix some typos
Tim-Philipp Müller [Tue, 7 Jun 2016 19:44:42 +0000 (20:44 +0100)]
docs: fix some typos

8 years agog-i: pass compiler env to g-ir-scanner
Tim-Philipp Müller [Wed, 25 May 2016 09:28:43 +0000 (10:28 +0100)]
g-i: pass compiler env to g-ir-scanner

It's what introspection.mak does as well. Should
fix spurious build failures on gnome-continuous
(caused by g-ir-scanner getting compiler details
via python which is broken in some environments
so passing the compiler details bypasses that).

8 years agortsp-session: RFC2326 does not allow a space between ; and timeout in the Session...
Ian [Wed, 18 May 2016 15:48:44 +0000 (16:48 +0100)]
rtsp-session: RFC2326 does not allow a space between ; and timeout in the Session header

This works with rtspsrc and live555, but fails with e.g. ffmpeg.

https://bugzilla.gnome.org/show_bug.cgi?id=766619

8 years agortspclientsink: Check return value of sscanf
Edward Hervey [Mon, 7 Mar 2016 13:48:38 +0000 (14:48 +0100)]
rtspclientsink: Check return value of sscanf

And just make sure we always have 0/0 if we have an error

CID #1352031

8 years agortsp-stream: Fix crash on cleanup with shared media and multiple udpsrc
Jake Foytik [Mon, 25 Apr 2016 12:55:25 +0000 (08:55 -0400)]
rtsp-stream: Fix crash on cleanup with shared media and multiple udpsrc

 - Unicast udpsrcs are now managed in a hash table. This allows for proper cleanup in with shared streams and fixes a memory leak.
 - Unicast udpsrcs are now properly cleaned up when shared connections exit. See the update_transport() function.
 - Create unit test for shared media.

https://bugzilla.gnome.org/show_bug.cgi?id=764744

8 years agortsp-stream: Always bind to ANY when address is a multicast address and not only...
Sebastian Dröge [Mon, 11 Apr 2016 07:55:23 +0000 (10:55 +0300)]
rtsp-stream: Always bind to ANY when address is a multicast address and not only on Windows

For IPv6 addresses, binding to a multicast group does not work on Linux
either. Always bind to ANY and then later join the multicast group.

https://bugzilla.gnome.org/show_bug.cgi?id=764679

8 years agoAutomatic update of common submodule
Julien Isorce [Thu, 14 Apr 2016 09:05:02 +0000 (10:05 +0100)]
Automatic update of common submodule

From 6f2d209 to ac2f647

8 years agortsp-thread-pool: explained why GSource is a part of ThreadImpl
Patricia Muscalu [Wed, 6 Apr 2016 08:09:46 +0000 (10:09 +0200)]
rtsp-thread-pool: explained why GSource is a part of ThreadImpl

Clarified why it is necessary to add source information to
GstRTSPThreadImpl. See the reported bug in GLib:
https://bugzilla.gnome.org/show_bug.cgi?id=720186
for more information.

https://bugzilla.gnome.org/show_bug.cgi?id=761702

8 years agoexamples: Clean up CFLAGS/LDADD even more
Sebastian Dröge [Mon, 4 Apr 2016 09:58:38 +0000 (12:58 +0300)]
examples: Clean up CFLAGS/LDADD even more

The internal .la should come first and is part of LDADD, as is
GST_CFLAGS/LIBS.

8 years agoexamples: Clean up CFLAGS/LDADD to link with the correct versions of all libraries
Sebastian Dröge [Mon, 4 Apr 2016 09:39:39 +0000 (12:39 +0300)]
examples: Clean up CFLAGS/LDADD to link with the correct versions of all libraries

8 years agortsp-server: Use $(GST_NET_LIBS) / $(GST_NET_CFLAGS)
Sebastian Dröge [Sun, 3 Apr 2016 09:06:29 +0000 (12:06 +0300)]
rtsp-server: Use $(GST_NET_LIBS) / $(GST_NET_CFLAGS)

8 years agortsp-server: Implement clock signalling according to RFC7273
Sebastian Dröge [Wed, 30 Dec 2015 16:39:05 +0000 (18:39 +0200)]
rtsp-server: Implement clock signalling according to RFC7273

For NTP and PTP clocks we signal the actual clock that is used and signal
the direct media clock offset.

For all other clocks we at least signal that it's the local sender clock.

This allows receivers to know which clock was used to generate the media and
its RTP timestamps. Receivers can then implement network synchronization,
either absolute or at least relative by getting the sender clock rate directly
via NTP/PTP instead of estimating it from RTP timestamps and packet receive
times.

https://bugzilla.gnome.org/show_bug.cgi?id=760005

8 years agortspclientsink: Add support for setting the multicast interface
Sebastian Dröge [Wed, 2 Mar 2016 17:42:58 +0000 (19:42 +0200)]
rtspclientsink: Add support for setting the multicast interface

https://bugzilla.gnome.org/show_bug.cgi?id=763000

8 years agortsp-media: Add support for setting the multicast interface
Sebastian Dröge [Wed, 2 Mar 2016 17:42:13 +0000 (19:42 +0200)]
rtsp-media: Add support for setting the multicast interface

https://bugzilla.gnome.org/show_bug.cgi?id=763000

8 years agortspclientsink: use new gst_element_class_add_static_pad_template()
Vineeth TM [Sun, 6 Mar 2016 23:50:01 +0000 (08:50 +0900)]
rtspclientsink: use new gst_element_class_add_static_pad_template()

https://bugzilla.gnome.org/show_bug.cgi?id=763196

8 years agoBack to development
Sebastian Dröge [Thu, 24 Mar 2016 11:33:43 +0000 (13:33 +0200)]
Back to development

8 years agoRelease 1.8.0
Sebastian Dröge [Thu, 24 Mar 2016 11:00:35 +0000 (13:00 +0200)]
Release 1.8.0

8 years agortsp-stream: Don't set the state of the appsrc from PLAYING to PAUSED again during...
Sebastian Dröge [Wed, 16 Mar 2016 21:35:09 +0000 (23:35 +0200)]
rtsp-stream: Don't set the state of the appsrc from PLAYING to PAUSED again during setup

This would get us NO_PREROLL in the bin again and break seeking.
Thanks to Carlos Rafael Giani for helping to debug this!

https://bugzilla.gnome.org/show_bug.cgi?id=740509

8 years agoRelease 1.7.91
Sebastian Dröge [Tue, 15 Mar 2016 10:26:13 +0000 (12:26 +0200)]
Release 1.7.91

8 years agortsp-stream: Ensure that the pipeline is live and later-added udpsrcs are syncing...
Sebastian Dröge [Thu, 10 Mar 2016 11:54:38 +0000 (13:54 +0200)]
rtsp-stream: Ensure that the pipeline is live and later-added udpsrcs are syncing the state with the parent bin

Without this, RECORD pipelines are broken because
a) we wait for ASYNC_DONE which never happens anymore because udpsrc would be
added later. Previously it was there earlier and due to NO_PREROLL caused the
pipeline to preroll immediately
b) the udpsrc for the pipeline is added later and never set to PLAYING state,
as the corresponding code previously was only for PLAY pipelines.

https://bugzilla.gnome.org/show_bug.cgi?id=763281

8 years agortsp-stream: Fix typo in the docstring
Jan Schmidt [Thu, 10 Mar 2016 14:22:54 +0000 (01:22 +1100)]
rtsp-stream: Fix typo in the docstring

gst_rtsp_stream_set_client_side -> gst_rtsp_stream_is_client_side