Alexander Schrab [Mon, 27 May 2013 10:58:07 +0000 (12:58 +0200)]
rtsp-server: Do not crash on nmapping of server
* generate error when gst_rtsp_connection_accept fails
* do not stop accepting incoming connections because
accepting a client fails
https://bugzilla.gnome.org/show_bug.cgi?id=701072
Alexander Schrab [Fri, 24 May 2013 11:39:50 +0000 (13:39 +0200)]
rtsp-client: ipv4 adress should not be marked ipv6 even if socket is ipv6
https://bugzilla.gnome.org/show_bug.cgi?id=700953
Sebastian Rasmussen [Wed, 22 May 2013 01:29:38 +0000 (03:29 +0200)]
rtsp-sdp: Parse framerate caps field and set SDP attribute
The SDP attribute and its format is described in RFC4566.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=700747
Sebastian Rasmussen [Wed, 22 May 2013 01:29:30 +0000 (03:29 +0200)]
rtsp-sdp: Parse width/height from caps and set SDP attribute
The SDP attribute and its format is described in RFC6064.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=700747
Patricia Muscalu [Mon, 29 Apr 2013 12:46:30 +0000 (14:46 +0200)]
rtsp-sdp: add bandwidth line
https://bugzilla.gnome.org/show_bug.cgi?id=699220
Sebastian Dröge [Wed, 15 May 2013 08:55:09 +0000 (10:55 +0200)]
Automatic update of common submodule
From 5edcd85 to 098c0d7
Ognyan Tonchev [Tue, 23 Apr 2013 09:28:39 +0000 (11:28 +0200)]
tests: add dynamic payloader prepare/unprepare check
Wim Taymans [Tue, 23 Apr 2013 08:27:35 +0000 (10:27 +0200)]
media: release lock when removing fakesink
Wim Taymans [Tue, 23 Apr 2013 08:16:17 +0000 (10:16 +0200)]
stream: set elements to NULL before removing
When removing a stream, set the elements to NULL first. This avoids
element-is-not-in-NULL-state errors when we dispose the elements.
Tim-Philipp Müller [Mon, 22 Apr 2013 22:55:48 +0000 (23:55 +0100)]
Automatic update of common submodule
From 3cb3d3c to 5edcd85
Wim Taymans [Mon, 22 Apr 2013 15:34:37 +0000 (17:34 +0200)]
media: listen to pad-removed signals
Listen to the pad-removed signal and remove the stream associated with the
removed pad.
Add signal to be notified of the removed pad.
Remove the fakesink in unprepare()
Fix signatures of the signal methods
Wim Taymans [Mon, 22 Apr 2013 15:33:30 +0000 (17:33 +0200)]
tests: add example of reusable pipelines
Ognyan Tonchev [Mon, 22 Apr 2013 15:32:31 +0000 (17:32 +0200)]
stream: add method to get the srcpad
Ognyan Tonchev [Mon, 22 Apr 2013 14:49:39 +0000 (16:49 +0200)]
check: add media prepare/unprepare test
See https://bugzilla.gnome.org/show_bug.cgi?id=698376
Ognyan Tonchev [Mon, 22 Apr 2013 14:40:48 +0000 (16:40 +0200)]
media: disconnect from signal handlers in unprepare()
We connected to the pad-added and no-more-pads signals in prepare() so
we need to disconnect from them in unprepare().
See https://bugzilla.gnome.org/show_bug.cgi?id=698376
Ognyan Tonchev [Mon, 22 Apr 2013 14:25:17 +0000 (16:25 +0200)]
media: don't free streams array
Don't free the streams array in the unprepare() method, they were not
added in prepare().
See https://bugzilla.gnome.org/show_bug.cgi?id=698376
Ognyan Tonchev [Mon, 22 Apr 2013 14:19:35 +0000 (16:19 +0200)]
media: don't unref the pipeline in unprepare
Unprepare() should undo what prepare() does. Because the pipeline is
not created in prepare(), we should not unref it in unprepare()
Ognyan Tonchev [Mon, 22 Apr 2013 14:09:22 +0000 (16:09 +0200)]
stream: clear session and caps for reuse
Set the session and caps to NULL after unref otherwise we might unref
them again later.
See https://bugzilla.gnome.org/show_bug.cgi?id=698376
David Svensson Fors [Mon, 15 Apr 2013 10:21:54 +0000 (12:21 +0200)]
client: send out teardown signal before tearing down
The advantage is that in the signal handler you get direct access to
information about what streams are about to get torn down (in the
GstRTSPClientState).
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=697686
David Svensson Fors [Mon, 15 Apr 2013 10:17:34 +0000 (12:17 +0200)]
client: expose connection
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=697546
Tim-Philipp Müller [Sun, 14 Apr 2013 16:58:22 +0000 (17:58 +0100)]
Automatic update of common submodule
From aed87ae to 3cb3d3c
Wim Taymans [Fri, 12 Apr 2013 10:34:38 +0000 (11:34 +0100)]
media: add method to get the base_time of the pipeline
Together with a shared clock, this base-time could eventually be sent to
the client so that it can reconstruct the exact running-time of the clock
on the server.
Wim Taymans [Tue, 9 Apr 2013 20:35:28 +0000 (22:35 +0200)]
media: add GstNetTimeProvider support
Add a property to let the media provide a GstNetTimeProvider for its clock.
Make methods to get the clock and nettimeprovider
Add a x-gst-clock property to the SDP with the IP and port number of the nettime
provider and also the current time of the clock. This should make it possible
for (GStreamer) clients to slave their clock to the server clock.
Stefan Sauer [Tue, 9 Apr 2013 19:02:47 +0000 (21:02 +0200)]
Automatic update of common submodule
From 04c7a1e to aed87ae
Wim Taymans [Tue, 9 Apr 2013 18:39:58 +0000 (20:39 +0200)]
media: wait for buffering to complete
Wait for buffering to complete before changing the state to the target state.
Wim Taymans [Tue, 9 Apr 2013 18:11:35 +0000 (20:11 +0200)]
media: small cleanup
David Svensson Fors [Wed, 20 Mar 2013 11:33:54 +0000 (12:33 +0100)]
tests: remove extra unref in test_setup_non_existing_stream
The unref is not needed anymore, teardown runs without it.
https://bugzilla.gnome.org/show_bug.cgi?id=696542
David Svensson Fors [Wed, 20 Mar 2013 10:28:11 +0000 (11:28 +0100)]
tests: GSocketService cleanup in test_bind_already_in_use
Use g_socket_service_stop so the rtspserver test stops listening for
incoming connections in test_bind_already_in_use.
https://bugzilla.gnome.org/show_bug.cgi?id=696541
Olivier Crête [Fri, 22 Mar 2013 22:25:07 +0000 (18:25 -0400)]
rtsp-media-factory: g_signal_connect_object is not thread safe, can't use it here
Instead use a GWeakRef which is safe to use
This is a known GLib bug, see:
https://bugzilla.gnome.org/show_bug.cgi?id=667145
Olivier Crête [Fri, 22 Feb 2013 19:17:29 +0000 (14:17 -0500)]
rtsp-media/client: Reply to PLAY request with same type of Range
Remember the type of Range from the PLAY request and use the same type for
the reply.
Patricia Muscalu [Mon, 18 Mar 2013 08:25:54 +0000 (09:25 +0100)]
rtsp-client: expose uri
Olivier Crête [Wed, 13 Mar 2013 21:46:58 +0000 (17:46 -0400)]
tests: Hold ref while creating second media
To test if the media aren't shared, make sure we keep the first one while creating a second
otherwise the same memory address may be reused.
Tim-Philipp Müller [Tue, 12 Mar 2013 00:10:18 +0000 (00:10 +0000)]
configure: remove out-of-date comment
Tim-Philipp Müller [Tue, 12 Mar 2013 00:05:49 +0000 (00:05 +0000)]
.gitignore: ignore more build files
Tim-Philipp Müller [Tue, 12 Mar 2013 00:03:36 +0000 (00:03 +0000)]
tests: use right _LIBS variable for gst-plugins-base libs
Wim Taymans [Mon, 11 Mar 2013 10:35:14 +0000 (11:35 +0100)]
check: add librtp to libs
Olivier Crête [Thu, 21 Feb 2013 00:37:51 +0000 (19:37 -0500)]
tests: Add test to check selecting a port the server will send from
Olivier Crête [Wed, 20 Feb 2013 23:30:01 +0000 (18:30 -0500)]
tests: Make sure packets are actually received
Olivier Crête [Tue, 19 Feb 2013 23:27:20 +0000 (18:27 -0500)]
stream: Select unicast address from pool if appropriate
Olivier Crête [Tue, 19 Feb 2013 21:43:08 +0000 (16:43 -0500)]
stream: Properties are always there in Gst 1.0
Olivier Crête [Tue, 19 Feb 2013 21:36:20 +0000 (16:36 -0500)]
tests: Add tests for unicast addresses in pool
Olivier Crête [Wed, 20 Feb 2013 19:26:03 +0000 (14:26 -0500)]
address-pool: Verify that multicast addresses are used for multicast and vice-versa
Olivier Crête [Tue, 19 Feb 2013 21:34:16 +0000 (16:34 -0500)]
address-pool: Add unicast addresses
Olivier Crête [Tue, 19 Feb 2013 18:19:41 +0000 (13:19 -0500)]
rtsp-server: Limit the number of threads per server instance
If we exceed the maximum, just round robin the clients over the existing
threads.
Olivier Crête [Tue, 19 Feb 2013 17:31:23 +0000 (12:31 -0500)]
rtsp-server: No need to store the GMainContext in the client context
Olivier Crête [Tue, 19 Feb 2013 01:22:18 +0000 (20:22 -0500)]
tests: Add test for client disconnection
Olivier Crête [Tue, 19 Feb 2013 01:15:41 +0000 (20:15 -0500)]
tests: Test client and session timeouts with multiple threads
Olivier Crête [Mon, 18 Feb 2013 19:59:58 +0000 (14:59 -0500)]
Document locking and its order
Olivier Crête [Sat, 16 Feb 2013 01:02:31 +0000 (20:02 -0500)]
tests: Test that slow DESCRIBE don't block other clients
Olivier Crête [Fri, 15 Feb 2013 00:52:09 +0000 (19:52 -0500)]
tests: Add tests for client-requested multicast address
Olivier Crête [Thu, 14 Feb 2013 18:44:54 +0000 (13:44 -0500)]
docs: Put the various functions in the right sections
Olivier Crête [Thu, 14 Feb 2013 18:38:07 +0000 (13:38 -0500)]
docs: Generate docs for GstRTSPAddressPool
Olivier Crête [Wed, 13 Feb 2013 23:32:20 +0000 (18:32 -0500)]
client: Check client provided addresses against the address pool
Olivier Crête [Wed, 13 Feb 2013 23:01:43 +0000 (18:01 -0500)]
address-pool: Add API to request a specific address from the pool
Also add relevant unit tests.
Olivier Crête [Wed, 13 Feb 2013 00:34:24 +0000 (19:34 -0500)]
tests: Check the passing around of a RTSPAddressPool
Make sure the RTSPAddressPool is propagated from the MediaFactory all the
way down to the stream.
Olivier Crête [Tue, 12 Feb 2013 21:34:37 +0000 (16:34 -0500)]
tests: Add more tests for the address pool
Olivier Crête [Tue, 12 Feb 2013 21:29:25 +0000 (16:29 -0500)]
address-pool: Fix off by one error
When splitting a port range, the port after a skip is not part of range.
Tim-Philipp Müller [Thu, 7 Mar 2013 00:04:19 +0000 (00:04 +0000)]
Automatic update of common submodule
From 2de221c to 04c7a1e
George McCollister [Thu, 7 Feb 2013 22:18:08 +0000 (16:18 -0600)]
configure: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS
AM_CONFIG_HEADER was removed in automake 1.13
https://bugzilla.gnome.org/show_bug.cgi?id=693368
Stefan Sauer [Mon, 28 Jan 2013 19:45:44 +0000 (20:45 +0100)]
Automatic update of common submodule
From a942293 to 2de221c
Wim Taymans [Mon, 28 Jan 2013 09:31:50 +0000 (10:31 +0100)]
client: make sure the watch exists while sending data
Protect the send_func with a lock. This allows us to wait for sending
to complete before changing the send_func and user_data. We add an
extra ref to the watch to make sure that it remains valid during
sending.
When closing the connection, set the send_func to NULL
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=692433
Tim-Philipp Müller [Wed, 16 Jan 2013 12:16:32 +0000 (12:16 +0000)]
tests: use GST_*_1_0 environment variables everywhere
The _1_0 suffixed environment variables override the
non-suffixed ones, so if we're in an environment that
sets the _1_0 suffixed ones, such as jhbuild, we need
to set those to make sure ours actually always get
used.
Tim-Philipp Müller [Tue, 15 Jan 2013 15:09:24 +0000 (15:09 +0000)]
Automatic update of common submodule
From acb04d9 to a942293
Wim Taymans [Fri, 14 Dec 2012 10:58:29 +0000 (11:58 +0100)]
rtsp-client: set the client backlog
Set the client backlog to a reasonable default
Ognyan Tonchev [Tue, 4 Dec 2012 08:47:35 +0000 (09:47 +0100)]
rtsp-media: Make the element a constructor parameter
https://bugzilla.gnome.org/show_bug.cgi?id=689594
Sebastian Rasmussen [Tue, 4 Dec 2012 00:05:31 +0000 (01:05 +0100)]
docs: Link with gcov library when gcov is enabled
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=689583
Wim Taymans [Fri, 30 Nov 2012 14:03:15 +0000 (15:03 +0100)]
media: match prepare with unprepare
Really unprepare when there were an equal amount of prepare calls.
Wim Taymans [Fri, 30 Nov 2012 13:58:46 +0000 (14:58 +0100)]
media: media has to be unprepared in finalize
Because unprepare takes away the last ref on the media.
Wim Taymans [Fri, 30 Nov 2012 13:36:30 +0000 (14:36 +0100)]
Revert "client: never call gst_rtsp_media_unprepare, let gst_rtsp_media_finalize do it"
This reverts commit
ba5b78ff2ff223049188eb456e228c709ccd3e05.
We can't use the refcount to trigger unprepare because it is the unprepare call
that removes the last refcount after all messages are consumed. What we should
probably do is make a prepared refcount and only unprepare when the refcount
reaches 0.
Wim Taymans [Fri, 30 Nov 2012 12:35:05 +0000 (13:35 +0100)]
media: let the source unref the last media ref
the last ref to the media is held by the source so we don't need to add more ref
and unrefs, we simply destroy the media when the source is gone.
Wim Taymans [Fri, 30 Nov 2012 11:54:10 +0000 (12:54 +0100)]
media: improve debug
Wim Taymans [Fri, 30 Nov 2012 11:53:02 +0000 (12:53 +0100)]
media: check state
Make sure we are in the right state when collecting the position and duration.
Only make ourselves PREPARED when we were previously PREPARING.
Wim Taymans [Fri, 30 Nov 2012 09:05:48 +0000 (10:05 +0100)]
media: use g_object_ref/unref for GObjects
Alessandro Decina [Fri, 30 Nov 2012 06:05:25 +0000 (07:05 +0100)]
client: never call gst_rtsp_media_unprepare, let gst_rtsp_media_finalize do it
Calling gst_rtsp_media_unprepare breaks shared medias. Just unref
GstRTSPMedia instances and let gst_rtsp_media_finalize unprepare when a media
isn't being used anymore.
Alessandro Decina [Fri, 30 Nov 2012 05:17:46 +0000 (06:17 +0100)]
Fix compiler warning
Alessandro Decina [Fri, 30 Nov 2012 05:14:49 +0000 (06:14 +0100)]
Add missing g_type_class_add_private in GstRTSPMediaFactoryURI
Wim Taymans [Thu, 29 Nov 2012 16:21:12 +0000 (17:21 +0100)]
small cleanup
Wim Taymans [Thu, 29 Nov 2012 16:20:56 +0000 (17:20 +0100)]
media: avoid element leak
Wim Taymans [Thu, 29 Nov 2012 16:20:26 +0000 (17:20 +0100)]
media: require an element in media constructor
Wim Taymans [Thu, 29 Nov 2012 16:07:30 +0000 (17:07 +0100)]
Revert "client: TEARDOWN brings that state to Init again"
This reverts commit
4b61fdad85a3ca84752bf074fdb2fa203954b32e.
The object is already disposed, there is no point in setting the state.
Wim Taymans [Thu, 29 Nov 2012 11:30:20 +0000 (12:30 +0100)]
client: TEARDOWN brings that state to Init again
Wim Taymans [Thu, 29 Nov 2012 10:11:05 +0000 (11:11 +0100)]
rtsp: make object details private
Make all object details private
Add methods to access private bits
Wim Taymans [Wed, 28 Nov 2012 13:50:47 +0000 (14:50 +0100)]
tests: add media tests
Wim Taymans [Wed, 28 Nov 2012 13:45:30 +0000 (14:45 +0100)]
media: check if prepared for some methods
Check that the media object is prepared before doing seek and getting the
current position etc.
Add some g_return checks.
Wim Taymans [Wed, 28 Nov 2012 11:40:46 +0000 (12:40 +0100)]
tests: add mediafactory test
Wim Taymans [Wed, 28 Nov 2012 11:40:18 +0000 (12:40 +0100)]
stream: improve debug
Wim Taymans [Wed, 28 Nov 2012 11:39:37 +0000 (12:39 +0100)]
media: unref pipeline in finalize to avoid leaking it
Wim Taymans [Wed, 28 Nov 2012 11:10:47 +0000 (12:10 +0100)]
rtsp: use gst_object_unref on GstObjects
Wim Taymans [Wed, 28 Nov 2012 11:10:14 +0000 (12:10 +0100)]
media-factory: require an url
Wim Taymans [Wed, 28 Nov 2012 10:40:33 +0000 (11:40 +0100)]
examples: fix include
Wim Taymans [Wed, 28 Nov 2012 10:17:27 +0000 (11:17 +0100)]
server: remove unused include
Wim Taymans [Wed, 28 Nov 2012 10:07:57 +0000 (11:07 +0100)]
tests: add test for mountpoints
Wim Taymans [Wed, 28 Nov 2012 10:05:08 +0000 (11:05 +0100)]
client: fix factory leak
Keep the factory in the state object only for authorization checks and make
sure we unref it on failure. Also don't keep invalid objects in the state
object.
Wim Taymans [Wed, 28 Nov 2012 09:40:14 +0000 (10:40 +0100)]
mounts: add g_return_if guards
Wim Taymans [Tue, 27 Nov 2012 11:51:55 +0000 (12:51 +0100)]
tests: add more tests
Wim Taymans [Tue, 27 Nov 2012 11:33:02 +0000 (12:33 +0100)]
client: improve debug
Wim Taymans [Tue, 27 Nov 2012 11:24:21 +0000 (12:24 +0100)]
client: improve debug and fix leaks
Cleanup the uri and session when there is a bad request.
Wim Taymans [Tue, 27 Nov 2012 11:17:05 +0000 (12:17 +0100)]
update common
Wim Taymans [Tue, 27 Nov 2012 11:13:59 +0000 (12:13 +0100)]
test: add test for session in options request
Wim Taymans [Tue, 27 Nov 2012 11:11:41 +0000 (12:11 +0100)]
client: use 454 when session can't be found
We should use 454 when a session can't be found because there was no session
pool configured in the server. This is not a server configuration problem
because the server on which the request is done might not be the same one that
will keep the sessions for us and so it does not need to support sessions.