Commit graph

105 commits

Author SHA1 Message Date
Thiago Santos
0d6195686b decodebin2: error when all streams have no buffers
In some cases (all buffers dropped by a parser) a decodebin2
chain might receive an EOS before it gets enough data to
expose a decoded pad. In the case that no streams can expose
a pad we should error out instead of hang.

Fixes #542758
2009-11-19 14:51:33 -03:00
Thiago Santos
e3e7ba0d1a decodebin2: set to buffer less on no-more-pads
When a decodebin2 receives no-more-pads of a group it
can set that group's multiqueue buffering thresholds to
'playing' buffering method, avoiding that it buffers
too long and cause problems when using with queue2.
See the associated bug for details.

Fixes #600787
2009-11-18 15:16:18 -03:00
Sebastian Dröge
1da5a3f7d3 playback: Update factories list on every access if the registry has changed
This makes application's simpler because the element doesn't need to
go to NULL first to make use of newly installed plugins.

Fixes bug #601480.
2009-11-11 14:00:26 +01:00
Sebastian Dröge
ab96265c57 playback: When going from NULL->READY check if the registry has new features
This makes it possible to use newly installed plugins after going back
to NULL instead of requiring a new instance.

Fixes bug #599266.
2009-11-10 18:30:46 +01:00
Sebastian Dröge
6c15d9e8d4 decodebin2: Add property to disable/enable posting of stream-topology messages
Most people don't need this messages and generating them is quite
expensive.
2009-11-06 17:01:04 +01:00
Sebastian Dröge
5798b543df decodebin2: Protect subtitle elements and subtitle encoding by a new mutex
Using the object lock here can and will lead to deadlocks because
of deep-notifies of property changes: the deep-notify handler will
get the parent of objects, which will take the object lock again.

Fixes bug #600479.
2009-11-06 15:15:06 +01:00
Stefan Kost
a78c8bf3ed pad: rename new api from _refed to _reffed.
Due to popular demand rename the new api as we still can.
2009-11-05 13:00:27 +02:00
Wim Taymans
89f02fb269 decodebin2: use new getcaps function to avoid copies
Use the gst_pad_get_caps_refed() to avoid some caps copy functions.
2009-11-04 18:31:09 +01:00
Sebastian Dröge
0285d77d96 decodebin2: Use static caps for the default raw caps and put them into a separate header
This way we can use the same default raw caps everywhere.
2009-11-03 13:03:23 +01:00
Tim-Philipp Müller
6f4c1ac583 Remove GST_DEBUG_FUNCPTR where they're pointless
There's not much point in using GST_DEBUG_FUNCPTR with GObject
virtual functions such as get_property, set_propery, finalize and
dispose, since they'll never be used by anyone anyway. Saves a
few bytes and possibly a sixteenth of a polar bear.
2009-10-28 00:59:35 +00:00
Wim Taymans
0b00e25b54 decodebin2: implement low/high watermark property 2009-10-27 15:22:22 +01:00
Wim Taymans
f998858192 decodebin2: buffering is implemented now 2009-10-24 16:09:59 -04:00
Wim Taymans
3d2b3dd268 decodebin2: configure use-buffering on multiqueue 2009-10-24 16:09:59 -04:00
Wim Taymans
baecd335b2 decodebin2: refactor queue size configuration.
Refactor the queue size configuration into a new method.
Use the same queue values for buffering as for preroll.
2009-10-24 16:07:36 -04:00
Wim Taymans
cffe4d7bd3 decodebin2: move error path down 2009-10-24 16:07:36 -04:00
Wim Taymans
1c982d0dbe decodebin2: implement max queue size properties 2009-10-24 16:07:36 -04:00
Wim Taymans
3fffb0e2dd decodebin2: add properties for buffering
Add properties that can be used to configure the multiqueue buffers and
buffering methods
2009-10-24 16:07:36 -04:00
Edward Hervey
891c54f6f8 gst/decodebin2: Ensure we get fixed caps for topology message
There are some corner cases (like with dvdemux amongst others) where
the caps won't be negotiated, but the pad has fixed caps.
2009-10-20 10:15:57 +02:00
Edward Hervey
64c8b1d5d9 gst/decodebin2: Don't expose chains if we're shutting down.
This avoids adding flushing pads to ourself
2009-10-20 10:15:48 +02:00
Sebastian Dröge
efcca84bac decodebin2: Post a element message on the bus with the stream topology
Fixes bug #598533.
2009-10-15 13:35:29 +02:00
Sebastian Dröge
50fdbcd9ea decodebin2: Store the "endcaps" of a chain
This are the caps that either resulted in a deadend if
no plugin for them could be found or raw caps.
2009-10-15 13:35:29 +02:00
Sebastian Dröge
366aaae825 decodebin2: Store for every chain, which pad resulted in its creation 2009-10-15 13:35:28 +02:00
Sebastian Dröge
d40246ff7d decodebin2: Ignore no-more-pads from non-demuxer elements
instead of printing an error that no corresponding group could
be found. no-more-pads from non-demuxer elements doesn't give
any additional information because there can only be a single srcpad.

Fixes bug #598288.
2009-10-13 16:52:43 +02:00
Jan Schmidt
592b8ecb09 decodebin2: Fix type-punning warning 2009-10-08 00:17:21 +01:00
Sebastian Dröge
9bd6fe41cb decodebin2: Chains with an exposed endpad are complete too
This allows partial group changes, i.e. demuxer2 in the example below
goes EOS but has a next group and audio2 stays the same.

          /-- >demuxer2---->video
demuxer---             \--->audio1
          \--->audio2
2009-10-07 17:46:30 +02:00
Sebastian Dröge
bf7cd0ed81 decodebin2: Use the iterate internal links function instead of string magic to get multiqueue srcpads 2009-10-07 17:46:29 +02:00
Sebastian Dröge
cf9c6a2271 decodebin2: Rewrite autoplugging and how groups of pads are exposed
This now keeps track of everything that is going on, creates
a tree of chains and groups to allow "demuxer after demuxer" scenarios
and allows chained Oggs with multiple streams (needs oggdemux or playbin2 fixes).

Also document everything in detail and give a general overview of what
decodebin2 is doing at the top of the sources.

Fixes bug #596183, #563828 and #591677.
2009-10-07 17:46:28 +02:00
Sebastian Dröge
999483b454 decodebin2: Don't set the external ghostpads blocked but only their targets
Pad blocks should never be done on external pads as outside elements
might want to use their own pad blocks on them and this will lead to
conflicts and deadlocks.
2009-10-07 07:23:20 +02:00
Sebastian Dröge
0dcc0857aa decodebin2: Only use the object lock for protecting the subtitle elements
Using the decodebin lock will result in deadlocks if the subtitle encoding
is accessed from a pad-added handler.
2009-10-07 07:23:19 +02:00
Sebastian Dröge
2194166e05 decodebin2: Post missing plugin messages before any error messages 2009-08-31 11:10:55 +02:00
Mark Nauwelaerts
188d698449 decodebin2: avoid assertion failure on empty/NULL caps 2009-08-12 13:39:12 +02:00
Edward Hervey
8cd1b5209b gst: Remove dead assignments and resulting unused variables 2009-08-08 15:54:02 +02:00
Sebastian Dröge
62f43a1c52 decodebin2: Free iterator after removing all groups 2009-06-17 07:24:53 +02:00
Wim Taymans
ef1030ee6e decodebin2: make sure varargs are of right type
Explicitly cast the variables to g_object_set to their right types.
2009-06-05 18:13:25 +02:00
Wim Taymans
f444f0edce decodebin2: increase stream probing queues
When we are probing for streams, we want to set the queue size in such a way
that we can scan a maximum amount of data without consuming too much memory.
Therefore, remove the time limit on the queue and only stop scanning after 2MB
of data.

See #584104.
2009-06-05 16:49:58 +02:00
Jan Schmidt
47d7464b10 decodebin/uridecodebin: Recognise subpicture/x-pgs pads and output them. 2009-05-28 22:07:30 +01:00
Wim Taymans
650215bcfd decodebin2: remove leftover elements
Remove all of the elements inside decodebin2 when goint to READY and NULL.
Makes decodebin2 reusable.
Fixes #583750
2009-05-27 18:12:10 +02:00
Wim Taymans
e685a9e86d decodebin2: avoid some _caps_ref in some cases
Only mess with the caps refcount when we configure different caps.
2009-05-19 15:34:38 +02:00
Wim Taymans
a0525fadb4 decodebin2: make subpictures a raw output format
Subpictures are a raw format, we want those pads exposed so that playbin2 can do
the subpicture mixing.
2009-05-12 10:40:11 +02:00
Edward Hervey
71a372c847 decodebin2: Don't reduce max-size-time of exposed groups. Fixes #579912
The 2s limit is way too small for a lot of files (which have an interleave
in time of between 3 and 5s). Instead, leave it to the initial 5s value
and reduce the other limits (allowing us to stay memory-efficient).
2009-04-23 11:54:55 +02:00
Wim Taymans
4265511b70 decodebin2: fix up the debugs and warnings
Use _OBJECT variants because we can. Go over some log statements and put them in
the right category.
Fixes #567740.
2009-04-14 11:34:49 +02:00
Wim Taymans
5ec2d48f0a decodebin2: do some more cleanup
Free the groups when we go to READY.
Allow for NO_PREROLL elements.
2009-03-25 19:01:45 +01:00
Wim Taymans
554b3aafe4 decodebin2: a pad starts out being not drained.
Mark a new pad as not drained until we get EOS on it.
2009-03-20 15:47:47 +01:00
Wim Taymans
600a810236 decodebin2: recover from failing to add a pad
When we cannot add a pad to the decodebin2 for some reason, print a warning but
continue adding the remaining pads.
2009-03-19 20:31:01 +01:00
Wim Taymans
20468a22c9 decodebin2: more cleanups and docs.
Add some more comments and use g_list_prepend().
2009-03-19 19:35:15 +01:00
Wim Taymans
17e7948325 decodebin2: refactoring and race fixes
Refactor some code so that we can take the right locks and in the right order.
Fixes quite a bit of races already.
2009-03-19 19:19:38 +01:00
Wim Taymans
563db0fdca decodebin2: add extra dynamic ref for demuxers
When we make a group connected to a demuxer, keep an extra dynamic refcount for
the group which is only decremented when no_more_pads or a multiqueue overrun is
detected. This way we avoid a race between exposing the group while more dynamic
refs are added from new pads.

Fixes #575588.
2009-03-17 19:09:02 +01:00
Wim Taymans
72533ecccc decodebin2: only remove pads that were added
Flag pads that were added so that we can see if we need to remove them later or
not.
2009-03-09 15:46:21 +01:00
Mark Nauwelaerts
b7ea2a9105 Unblock blocked ghostpads when shutting down. Fixes #574293. 2009-03-09 13:32:21 +01:00
Michael Smith
e9e9d82fbe decodebin2: don't stay connected to notify::caps after negotiation
Disconnect the notify::caps signal in our callback (it'll be re-added
if we're not, in fact, finished getting complete caps). Ensures that
caps changes mid-stream (e.g. from an mp3 that changes from
stereo->mono mid-file) don't cause us to try to add a new pad.
2009-03-05 15:44:17 -08:00