mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2024-09-23 04:20:17 +00:00
Code Issues 1 Projects Releases Wiki Activity
1787 commits 44 branches 1508 tags 427 MiB
Commit graph

170 commits

Author SHA1 Message Date
Thibault Saunier
59d83f1a93 ges: Always create a project to back a timeline 
And fix all the tests as we need to wait for the project to be loaded
to check the reference count of the timeline (as we keep a ref on the
timeline in project to later emit "loaded" on idle).
 2013-11-09 10:35:12 -03:00
Mathieu Duponchelle
fb6aabfe4d timeline: pass the correct argument to disconnect_by_func. 
fixes #709205
 2013-10-02 21:17:56 +02:00
Thibault Saunier
002977cd85 timeline: Add a signal to know when it is commited 2013-09-28 15:42:20 +02:00
Thibault Saunier
9c77d52f30 timeline: Make sure we do not move object when only trimming 
We were missing a few checks so that we do not move objects when their
duration is equal to the max duration, or 0
 2013-09-08 19:19:24 -03:00
Thibault Saunier
c47c115dd1 timeline: Avoid setting duration > max_duration when rippling 
We should use the trimming method to set duration to make sure to avoid
going over the max duration.

Also avoid computing when setting duration to the same old value.
 2013-09-07 12:59:17 -04:00
Thibault Saunier
c4c26f8748 ges: Handle trimming in groups 
This was broken, clips where moving all around, make it behave properly.
 2013-09-07 12:55:58 -04:00
Thibault Saunier
c74aee5fdc timeline: Make ripple start be trimming 
This is a more natural behaviour as a user instead of doing nothing at
all.
 2013-09-07 02:10:12 -04:00
Thibault Saunier
f7ac18c138 timeline: No autotrans between elements in same toplevel container 
This makes no sense, we ended up creating/removing tons of transition
while moving groups
 2013-08-30 18:47:24 -04:00
Mathieu Duponchelle
09405c0080 timeline: when there are no objects anymore, set duration to 0. 2013-08-21 08:06:59 -04:00
Thibault Saunier
0671b69d34 timeline: Add support for group-id in the stream-start event 2013-08-02 14:47:24 +02:00
Thibault Saunier
5c8e2737b4 timeline: Add methods to get and set the snapping distance 2013-07-24 22:37:06 -04:00
Mathieu Duponchelle
4addde8d74 timeline: create_transitions_on_layer *before* actually commiting 
Everything need to be in place before commiting, otherwize it makes no
sense at all.
 2013-07-12 18:17:00 -04:00
Mathieu Duponchelle
96204ac1d2 container: Add a 'recursive' argument to the get_children method 
API:
  - ges_container_get_children (GESContainer *container);
  + ges_container_get_children (GESContainer *container, gboolean recurse);
 2013-07-10 23:25:12 -04:00
Thibault Saunier
b0202ed22b timeline: Append missing layers when moving groups between layers 
This was a missing feature of the newly added groups
 2013-07-09 21:30:59 -04:00
Thibault Saunier
a933d9540d ges: Implement a GESGroup class, subclass of GESContainer 
The GESGroup class is used to group various GESContainer
together, it can contain either GESClips or GESGroup or both.
 2013-07-09 16:47:00 -04:00
Thibault Saunier
55b7628382 ges: Make space in the GESTracks to be able to add mixing elements later 
And update the tests
 2013-06-23 18:28:52 -04:00
Thibault Saunier
b57f55121b timeline-element: Make it possible to reset the timeline property to NULL 
+ Add some debug symbol
 2013-06-23 17:39:05 -04:00
Thibault Saunier
f6038e3ad2 ges: Port to the new commit based API in GNL 
The GNL API changed to go from a model where user could
enable/disable updates in the composition, which leaded to races
in many places, to a model where any positioning change in the
composition is not directly done but 'cached' and then the user
has to commit those changes so they become effective in the media
processing stack.

The new API in GES is pretty similare and is basically copy
pasting this new design.

We still need to see if in some context it would make sense to add
a mode where we would commit any changes ourself at the end of our
operation for basic use cases.

Removed APIs:
  ges_timeline_enable_update
  ges_timeline_is_updating
  ges_track_enable_update
  ges_track_is_updating

New APIs:
  ges_track_commit
  ges_timeline_commit
 2013-06-23 16:33:33 -04:00
Thibault Saunier
284648e536 timeline: Sync track enable_update property with parent 
When we add a track to a timeline, we want it "enable update" property
to be set to the timeline's
 2013-06-12 11:16:37 -04:00
Thibault Saunier
7e121ff167 ges: Protect from Gst dynamic callbacks 
The pad-added and no-more-pad signal can be emited from any thread
so we have to protect our code from that
 2013-05-29 17:14:31 -04:00
Anton Belka
7d999c5ab8 timeline: add auto-transition 
API:
 ges_timeline_get_auto_transition
 ges_timeline_set_auto_transition
 GESTimeline::auto-transition
 2013-05-09 18:54:56 -04:00
Thibault Saunier
b2b96c743d Rename GESTimelineLayer to GESLayer 2013-04-23 20:22:31 -03:00
Thibault Saunier
b5c52a17ec layer: Remove the "rate" property of ges_timeline_layer_add_asset 
API:
 - ges_timeline_layer_add_asset (layer, asset,  start,  inpoint, duration, rate, track_types);
 + ges_timeline_layer_add_asset (layer, asset,  start,  inpoint, duration, track_types);
 2013-04-23 20:22:27 -03:00
Thibault Saunier
5ab40ce5cb ges: Fix compilation with clang 2013-04-21 21:11:52 -03:00
Thibault Saunier
cdd00ed207 ges-clip: Remove the unlocked TrackElement APIs 
Remove APIs:
  ges_track_element_set_locked
  ges_track_element_is_locked

Those APIs where really not nice to use and were causing more issues
than solving them. If 2 time related properties of TimelineElement must
be different, then those element can *not* have the same parent.

Plus, with the new ges_container_group () API, we will recreate 1
GESClip containing the proper GESTimelineElements if it is the thing
to do.
 2013-04-15 00:18:05 -03:00
Thibault Saunier
4039468b57 timeline: call sync_state_with_parent when adding a child 2013-04-09 00:11:18 -03:00
Thibault Saunier
40376f9219 timeline: Do no unref the timeline before returning it 2013-03-30 13:34:36 +01:00
Thibault Saunier
f114b6eb3b timeline: Comment some variables goals 2013-03-29 15:50:12 +01:00
Thibault Saunier
7313916ae8 timeline: Remove TrackElement from its container on GESTimelineLayer::"clip-removed" 
.... when the Track is NULL
 2013-03-28 11:16:41 +01:00
Thibault Saunier
5a5b07297c Use gst_object_ref_sink instead of g_object_ref_sink when appropriate 
Making refcount issue debugging simpler
 2013-03-18 12:48:42 -03:00
Thibault Saunier
0767f08c7a Always prefer gst_object_(un)ref over g_object_(un)ref 
Making the refcount issue debugging easier
 2013-03-18 12:48:41 -03:00
Thibault Saunier
aa740d86bd Remove GESTrackElements from GESTracks when removing from a GESClip 
... Not the other way round.

  + Add and enhance debugging info on the way

The user should not be responsible for removing the GESTrackElements from
GESTracks, instead, removing it from a GESClip should imply removing
it from any GESTrack it is in.

This patch changes sensibly the behaviour when we remove a
GESTrackElement from a GESTrack, not remoing it from the GESClip it is
in. *But*, users should never remove a GESTrackElement from a GESTrack
anyway. The testsuite has been updated to that new behaviour.
 2013-03-15 11:17:06 -03:00
Thibault Saunier
7943bb510a ges: Make GESTimeline responsible for adding GESTrackElement to GESTrack 
+ Fix tests as necessary (Do not use agingtv as it can be "applied" on any TrackType
    	   and is not representative of what happens IRL)

We already had the infrastructure so the user can have the control over where to add
the elements (through the "select-track-for-object" signal). We now make use of that
signal everytime a GESClip is added to a GESTimelineLayer. This make user's life easier,
and object responsability clearer.
 2013-03-15 00:13:09 -03:00
Thibault Saunier
795b8df1b6 Implement GESContainer 
+ Fix unit tests
  + Minor enhancement in unit tests

API changes:
-----------
  * ges_track_element_get_clip     -> ges_timeline_element_get_parent
  * ges_clip_add_track_element     -> ges_container_add
  * ges_clip_release_track_element -> ges_container_remove
  * ges_clip_get_track_elements    -> ges_container_get_children
                                      (or GES_CONTAINER_CHILDREN)
 2013-03-15 00:13:08 -03:00
Thibault Saunier
0888e5e25b Rename object/tobj/trobj to clip or track_element as necessary 
Not really complete but it is a good start!
 2013-02-15 16:26:30 -03:00
Thibault Saunier
7f3c952737 Finish renaming tck_obj and derivate to track_element 2013-02-15 15:23:16 -03:00
Thibault Saunier
d21791a8d0 Rename GESTimelineLayer.xxx_object to GESTimelineLayer.xxx_clip 2013-02-15 15:23:15 -03:00
Thibault Saunier
ec1db99266 track: Rename all GESTrack.xxx_object to GESTrack.xxx_element 2013-02-15 15:23:15 -03:00
Thibault Saunier
fc8f06eedb Properly rename object-added to clip-added 2013-02-15 15:23:15 -03:00
Thibault Saunier
4ddcfb642d Finish renaming timeline object to clip 2013-02-15 15:23:15 -03:00
Thibault Saunier
b4e1131b70 Finish renaming track object to track element 2013-02-15 15:23:15 -03:00
Thibault Saunier
3f97a05c11 Rename GESTrackTransition to GESTransition 2013-02-15 14:42:06 -03:00
Thibault Saunier
6363d0fdad Rename GESTrackSource to GESSource 2013-02-15 14:42:06 -03:00
Thibault Saunier
d69964fd2a Rename GESTrackObject to GESTrackElement 2013-02-15 14:42:05 -03:00
Thibault Saunier
6f4315b75d Rename GESStandardTransitionClip to GESTransitionClip 2013-02-15 14:42:04 -03:00
Thibault Saunier
d7693b4014 Rename GESTimelineStandardTransition to GESStandardTransitionClip 2013-02-15 14:42:03 -03:00
Thibault Saunier
b73cf7c3d7 Rename GESTimelineTransition to GESTransitionClip 2013-02-15 14:42:03 -03:00
Thibault Saunier
04a3c49ebd Rename GESTimelineObject to GESClip 2013-02-15 14:42:02 -03:00
Thibault Saunier
f0359481c0 Add a GESTimelineElement base class 
+ Port GESTrackObject and GESTimelineObject to the new baseclass
 2013-02-15 14:42:02 -03:00
Thibault Saunier
02652902f5 Reimplement the auto-transition feature 
+ Actually implement unit tests
 2013-01-10 18:53:15 -03:00