Tim-Philipp Müller [Thu, 14 Jun 2012 22:36:29 +0000 (23:36 +0100)]
taglist: no need to store the size of the tag list structure inside the structure
Wim Taymans [Fri, 15 Jun 2012 09:24:46 +0000 (11:24 +0200)]
miniobject: expand docs a little
Add blurb about qdata and weak refs.
Wim Taymans [Fri, 15 Jun 2012 08:44:24 +0000 (10:44 +0200)]
miniobject: add qdata
Keep track of qdata for miniobjects. Reuse the weak ref array for this because
we can.
Wim Taymans [Fri, 15 Jun 2012 08:56:13 +0000 (10:56 +0200)]
miniobject: fix error in the weak ref handling
When 2 weak refs are added, the array is not resized big enough.
Simplify the weak ref handling code.
Free memory when we remove all weak refs.
Allow installing the same weak ref multiple times, like in gobject.
Wim Taymans [Thu, 14 Jun 2012 15:11:11 +0000 (17:11 +0200)]
miniobject: remove the size field
The size field is used by subclasses to store the total allocated size of the
memory for this miniobject. Because miniobject doesn't really do anything with
this field we can move it to the subclasses.
Wim Taymans [Thu, 14 Jun 2012 14:30:12 +0000 (16:30 +0200)]
memory: make GstMemory a miniobject
Wim Taymans [Thu, 14 Jun 2012 14:27:26 +0000 (16:27 +0200)]
trace: always print miniobject refcount
Wim Taymans [Thu, 14 Jun 2012 13:40:43 +0000 (15:40 +0200)]
memory: fix copy function
Make the copy function map to ref because we can't safely copy the user_data.
Wim Taymans [Thu, 14 Jun 2012 13:33:50 +0000 (15:33 +0200)]
memory: make GstAllocator a miniobject
David Svensson Fors [Tue, 12 Jun 2012 11:26:35 +0000 (13:26 +0200)]
gstfunnel: avoid access of freed pad
Save the value of the pad's got_eos in gst_funnel_release_pad,
before calling gst_element_remove_pad. This is because
gst_element_remove_pad may free the pad.
https://bugzilla.gnome.org/show_bug.cgi?id=678017
Vincent Penquerc'h [Thu, 14 Jun 2012 13:05:15 +0000 (14:05 +0100)]
element: fix pad transfer annotation from none to full
since the pad will be unreffed.
Wim Taymans [Wed, 13 Jun 2012 08:52:48 +0000 (10:52 +0200)]
message: add the running-time to the async-done message
Add the running-time of the buffer that caused the async operation to complete
to the async-done message.
Update bin to handle the new async-done message.
Wim Taymans [Wed, 13 Jun 2012 08:51:04 +0000 (10:51 +0200)]
pipeline: use reset_time message to reset the start time
Use the new RESET_TIME message to reset the start-time of the pipeline to the
requested time.
Make basesink request a new running-time when the flush-stop message tells it to
insteasd of waiting for preroll.
Wim Taymans [Wed, 13 Jun 2012 08:16:27 +0000 (10:16 +0200)]
message: add a new message to reset time
Add a new message to reset the pipeline running_time. Currently reseting the
pipeline can only be requested in the async_done message which means that the
pipeline needs to be prerolled. It is better to move this to a separate message.
Wim Taymans [Tue, 12 Jun 2012 15:11:51 +0000 (17:11 +0200)]
bin: always recurse into bins when doing state changes
Never skip the state change of a bin because it needs to update the base time of
its children when needed.
Tim-Philipp Müller [Tue, 12 Jun 2012 23:30:48 +0000 (00:30 +0100)]
docs: update for new datetime api
Tim-Philipp Müller [Tue, 12 Jun 2012 23:28:00 +0000 (00:28 +0100)]
win32: update .def file for latest API
Tim-Philipp Müller [Tue, 12 Jun 2012 23:25:24 +0000 (00:25 +0100)]
docs: add new datetime API
Tim-Philipp Müller [Tue, 12 Jun 2012 23:21:32 +0000 (00:21 +0100)]
tests: add some basic unit tests for partial date time fields
Tim-Philipp Müller [Tue, 12 Jun 2012 22:52:02 +0000 (23:52 +0100)]
datetime: clean-ups and new API adjustments
Remove constructors we don't want:
gst_date_time_new_ymd_h() because we don't want to
support hour-only for now;
gst_date_time_new_ymd_hm() because we don't want to
add constructors with time info where the caller doesn't
have to think about what timezone the time is in.
Lots of compulsive clean-up. Docs fixes. Replace
has_minute() and has_hour() with has_time().
Oleksij Rempel [Tue, 12 Jun 2012 20:35:42 +0000 (22:35 +0200)]
datetime: allow GstDateTime where not all fields are set
In order to deserialise and re-serialise dates and date times
from tags properly, we need to be able to express partial
dates (e.g. YYYY or YYYY-MM) and date times.
We only support partial date times where all the more
significant fields above the first unset field are set
(e.g. YYYY-00-DD is not supported).
Calling _get_foo() when foo is not set is not allowed
any more, callers need to check which fields are set
first.
https://bugzilla.gnome.org/show_bug.cgi?id=677757
Tim-Philipp Müller [Tue, 12 Jun 2012 21:45:09 +0000 (22:45 +0100)]
message: fix up minor inconsistency in structure name of state-changed message
Evan Nemerson [Tue, 12 Jun 2012 18:42:30 +0000 (11:42 -0700)]
introspection: add some missing annotations
Wim Taymans [Tue, 12 Jun 2012 12:24:27 +0000 (14:24 +0200)]
basesrc: handle flush events on the element as well
Handle flush-start and flush-stop sent on the element as well and send them
downstream. Make sure to send a segment event after the flush stop.
Stefan Sauer [Tue, 12 Jun 2012 09:05:05 +0000 (11:05 +0200)]
childproxy: add a few more comments
Stefan Sauer [Mon, 11 Jun 2012 18:34:00 +0000 (20:34 +0200)]
childproxy: fix signal handler signatures in class
When adding the name parameter, we forgot to add it here too.
Stefan Sauer [Mon, 11 Jun 2012 08:59:49 +0000 (10:59 +0200)]
childproxy: use GstChildProxy instead of GObject on the public api
Fix usage and also cleanup gst_object api use on gobjects.
Wim Taymans [Mon, 11 Jun 2012 13:49:10 +0000 (15:49 +0200)]
element: remove unused UNPARENTING flag
Wim Taymans [Mon, 11 Jun 2012 13:41:58 +0000 (15:41 +0200)]
bin: reorganize _remove_func to avoid races
Make the gst_bin_remove_func more like the add_func. Check if the element we try
to remove from the bin has the bin as the parent and set the parent flag to NULL
immediately, this allows us to avoid concurrent remove operations without using
the UNPARENTING element flag. After we unparented the element from the bin, we
update the bin state and remove the element from the list. Finally we unlink
all the pads.
This avoids a race condition where the element could still claim to have the
bin as the parent while the bin didn't have a pointer to the element anymore.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=647759
Matej Knopp [Sun, 10 Jun 2012 16:48:00 +0000 (12:48 -0400)]
typefindelement: Only send caps when pad is being activated
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=677819
Matej Knopp [Sun, 10 Jun 2012 16:41:12 +0000 (12:41 -0400)]
gstelement: Start over if subclass removed the next pad too
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=677818
Tim-Philipp Müller [Sat, 9 Jun 2012 17:05:14 +0000 (18:05 +0100)]
datetime: remove fallback code for old GLibs
Tim-Philipp Müller [Sat, 9 Jun 2012 16:13:04 +0000 (17:13 +0100)]
scripts: add create-uninstalled-setup script
Little script that sets up things in ~/gst and clones
the main modules and prints some instructions.
From http://gstreamer.freedesktop.org/wiki/UninstalledSetup
Sebastian Dröge [Fri, 8 Jun 2012 13:45:38 +0000 (15:45 +0200)]
elements: Use gst_pad_set_caps() and don't ignore its return value
Sebastian Dröge [Fri, 8 Jun 2012 13:41:17 +0000 (15:41 +0200)]
basesrc: Don't ignore the return value of gst_pad_set_caps() and call it after the vfunc
Sebastian Dröge [Fri, 8 Jun 2012 13:36:40 +0000 (15:36 +0200)]
basesink: Use gst_pad_set_caps() instead of the manual event fiddling
Sebastian Dröge [Fri, 8 Jun 2012 13:32:37 +0000 (15:32 +0200)]
basetransform: Don't return the return value of gst_pad_set_caps()
e.g. it returns FALSE if incompatible caps are set on the pad.
Edward Hervey [Wed, 6 Jun 2012 17:02:00 +0000 (19:02 +0200)]
gstutils: Faster read macros
On platforms that can do unaligned read/write, we can read/write much faster
by just casting.
https://bugzilla.gnome.org/show_bug.cgi?id=599546
Edward Hervey [Thu, 7 Jun 2012 10:49:10 +0000 (12:49 +0200)]
check: Add a test for GST_READ_* macros
Edward Hervey [Fri, 8 Jun 2012 12:49:51 +0000 (14:49 +0200)]
Update common submodule
Wim Taymans [Thu, 7 Jun 2012 15:58:02 +0000 (17:58 +0200)]
basetransform: fix reconfigure
Use the pad methods to set and check the reconfigure flags
Clear the reconfigure flag before we negotiate so that we don't miss any
reconfigure events while negotiating
Sebastian Dröge [Thu, 7 Jun 2012 13:56:44 +0000 (15:56 +0200)]
Back to development
Sebastian Dröge [Thu, 7 Jun 2012 13:56:17 +0000 (15:56 +0200)]
Release 0.11.92
Sebastian Dröge [Thu, 7 Jun 2012 13:53:49 +0000 (15:53 +0200)]
Update .po files
Wim Taymans [Thu, 7 Jun 2012 13:28:39 +0000 (15:28 +0200)]
basesrc: release the object lock sooner
Release the object lock before we get the time of the clock because that code
might take other locks.
Fix potential clock refcount error because we released the object lock but
didn't ref the clock.
Wim Taymans [Thu, 7 Jun 2012 08:34:46 +0000 (10:34 +0200)]
basesrc: remove 0.11 fixme
We always require elements to have an unlock_stop vmethod.
Edward Hervey [Wed, 6 Jun 2012 16:11:13 +0000 (18:11 +0200)]
registry: We name the registry after the target cpu
And not the host cpu
Conflicts:
gst/gstregistry.c
Edward Hervey [Wed, 6 Jun 2012 16:18:18 +0000 (18:18 +0200)]
Automatic update of common submodule
From
1fab359 to
03a0e57
Wim Taymans [Wed, 6 Jun 2012 13:45:11 +0000 (15:45 +0200)]
tests: fix unit test after event change
Someone forgot to run make check before pushing...
Sreerenj Balachandran [Wed, 6 Jun 2012 08:06:32 +0000 (11:06 +0300)]
gstadapter: Align the comment description with public api instead of internal one.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=677536
Wim Taymans [Wed, 6 Jun 2012 13:29:49 +0000 (15:29 +0200)]
element: fix pad cleanup in dispose
In the dispose handler we first need to release all the request pads and then
remove the remaining pads. This is because it is possible that releasing the
request pad might also cleanly remove some of the other dynamic pads, like
what rtpsession does.
https://bugzilla.gnome.org/show_bug.cgi?id=677436
Sebastian Dröge [Wed, 6 Jun 2012 12:14:31 +0000 (14:14 +0200)]
event: Don't make the TOC event a multi-sticky event
Elements are supposed to merge upstream events.
Havard Graff [Tue, 13 Oct 2009 15:24:34 +0000 (17:24 +0200)]
Make sure that unlinked pads do not cause a return false on latency events.
Context: Latency configuration should not be
messed up because of not-linked pads. In general,
one return FALSE on latency distribution causes
the "overall" pipeline latency configuration to
fail. This shows up as noise in logs (warning).
Conflicts:
gst/gstpad.c
Wim Taymans [Wed, 6 Jun 2012 10:52:09 +0000 (12:52 +0200)]
event: add name to sticky_multi events
The name of the event is used to store multiple sticky events of a certain type
on a pad.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=676859
Sebastian Dröge [Wed, 6 Jun 2012 07:59:55 +0000 (09:59 +0200)]
design: Also mention that the order of the filter caps is important
Wim Taymans [Wed, 6 Jun 2012 07:15:04 +0000 (09:15 +0200)]
query: improve docs
Wim Taymans [Wed, 6 Jun 2012 07:13:11 +0000 (09:13 +0200)]
pad: only serialized events can't pass after EOS
Only serialized events can't be sent on pads that are EOS. Otherwise a seek
event would be refused as well.
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=677520
Wim Taymans [Tue, 5 Jun 2012 12:38:25 +0000 (14:38 +0200)]
docs: talk about the filter caps
Tim-Philipp Müller [Sat, 2 Jun 2012 15:44:59 +0000 (16:44 +0100)]
tests: add unit test for tag list writability
Tim-Philipp Müller [Sat, 2 Jun 2012 15:38:35 +0000 (16:38 +0100)]
gst_tag_list_free -> gst_tag_list_unref
Tim-Philipp Müller [Sat, 2 Jun 2012 15:29:45 +0000 (16:29 +0100)]
docs: expand taglist section in porting-to-0.11 docs a bit
Wim Taymans [Tue, 5 Jun 2012 09:28:38 +0000 (11:28 +0200)]
docs: update negotiation docs
Mention that the acceptcaps query does not have to be recursive
Wim Taymans [Tue, 5 Jun 2012 07:40:14 +0000 (09:40 +0200)]
queue2: remove obsolete caps code
Wim Taymans [Tue, 5 Jun 2012 07:39:41 +0000 (09:39 +0200)]
utils: improve debug
Wim Taymans [Tue, 5 Jun 2012 07:21:10 +0000 (09:21 +0200)]
tests: fix unit test
Before we can change the caps on a sinkpad with fixed caps we need to unfix the
pad caps.
Wim Taymans [Tue, 5 Jun 2012 07:10:05 +0000 (09:10 +0200)]
pad: don't pause task on EOS
Elements should not rely on core to pause tasks on EOS.
Wim Taymans [Tue, 5 Jun 2012 07:00:01 +0000 (09:00 +0200)]
pad: fix event type check
Wim Taymans [Mon, 4 Jun 2012 14:19:26 +0000 (16:19 +0200)]
pad: fix 'res' may be used uninitialized in this function
Sebastian Dröge [Mon, 4 Jun 2012 11:00:53 +0000 (13:00 +0200)]
funnel: Fix unit test
Sebastian Dröge [Mon, 4 Jun 2012 10:57:59 +0000 (12:57 +0200)]
valve: Fix unit test
Sebastian Dröge [Mon, 4 Jun 2012 09:46:42 +0000 (11:46 +0200)]
pad: Don't accept any buffers or events after EOS
Mark Nauwelaerts [Mon, 4 Jun 2012 09:13:02 +0000 (11:13 +0200)]
baseparse: also perform state processing upon non-OK return
... since processing might still continue (if e.g. NOT_LINKED)
and then proper state (e.g. offset) needs to be maintained
(e.g. to arrange for a new frame setup).
Sebastian Dröge [Mon, 4 Jun 2012 09:25:47 +0000 (11:25 +0200)]
pad: Always return errors for EOS events immediately
For non-EOS events things will error out later during data
flow but after EOS events no data flow is happening.
See bug #677340.
Sebastian Dröge [Mon, 4 Jun 2012 07:27:35 +0000 (09:27 +0200)]
pad: Only forward caps events to a pad if it accepts the caps
Fixes bug #677335.
Sebastian Dröge [Sat, 2 Jun 2012 18:01:01 +0000 (20:01 +0200)]
Revert "pad: Return FALSE if pushing of sticky events failed"
This reverts commit
0f924b922c712059d7752fc15b832551745ff27e.
Sticky events should always return TRUE when pushing and will
only cause failures during data flow later.
Tim-Philipp Müller [Sat, 2 Jun 2012 15:18:46 +0000 (16:18 +0100)]
pad: fix variable-set-but-not-used compiler warning
Sebastian Dröge [Sat, 2 Jun 2012 14:55:07 +0000 (16:55 +0200)]
pad: If pushing a sticky event failed, make sure to at least push any pending EOS events
Otherwise a pipeline where one sticky event fails to be sent will
never forward EOS events downstream. This can cause pipelines to
wait forever for EOS on errors.
Sebastian Dröge [Sat, 2 Jun 2012 14:02:07 +0000 (16:02 +0200)]
pad: Return FALSE if pushing of sticky events failed
Instead of just ignoring failure of pushing sticky events and
returning TRUE as if everything is fine.
Andre Moreira Magalhaes (andrunko) [Fri, 1 Jun 2012 14:34:16 +0000 (16:34 +0200)]
inputselector: Correctly get current running time when syncing to the segment information
Fixes bug #677263.
Edward Hervey [Fri, 1 Jun 2012 08:28:30 +0000 (10:28 +0200)]
Automatic update of common submodule
From
f1b5a96 to
1fab359
Mike Ruprecht [Sat, 26 May 2012 03:58:57 +0000 (22:58 -0500)]
tests: Add funnel test to cover EOS event handling
Ported from
f3b2dd6f in the 0.10 branch
Mike Ruprecht [Sat, 26 May 2012 03:52:33 +0000 (22:52 -0500)]
funnel: Only emit EOS event if all sinkpads have received one
If multiple sources are plugged into the funnel and one of the
sources emits an EOS, that event is propogated through the funnel
even though other sources connected to the funnel may still be
pushing data. This patch waits to send an EOS event until the
funnel has received an EOS event on each sinkpad.
Ported from
d397ea97 in 0.10 branch.
Mike Ruprecht [Wed, 30 May 2012 00:24:25 +0000 (19:24 -0500)]
tests: Fix invalid read when releasing request pads in funnel tests
Mike Ruprecht [Wed, 30 May 2012 00:23:07 +0000 (19:23 -0500)]
funnel: Fix buffer leak
Andre Moreira Magalhaes (andrunko) [Thu, 31 May 2012 15:45:29 +0000 (17:45 +0200)]
inputselector: Don't try to sync on the segment if it has no TIME format
...and wait until it is actually configured and has a format before
trying to sync.
Sebastian Dröge [Thu, 31 May 2012 15:03:54 +0000 (17:03 +0200)]
inputselector: No need to broadcast the signal in flush-stop
Everything stopped at this point already.
Conflicts:
plugins/elements/gstinputselector.c
Sebastian Dröge [Thu, 31 May 2012 11:07:49 +0000 (13:07 +0200)]
Automatic update of common submodule
From
92b7266 to
f1b5a96
Bastien Nocera [Thu, 31 May 2012 09:10:41 +0000 (10:10 +0100)]
queue2: Fix property name in the docs
temp-template, not temp-tmpl
https://bugzilla.gnome.org/show_bug.cgi?id=677170
Andre Moreira Magalhaes (andrunko) [Mon, 28 May 2012 17:29:00 +0000 (14:29 -0300)]
inputselector: Properly sync when changing streams
This adds properties to use the clock time for deciding when
to drop buffers for inactive pads and a property to buffer all
not rendered buffers for the active pad to allow pad switching
without losing any buffers at all.
Conflicts:
plugins/elements/gstinputselector.c
Sebastian Dröge [Wed, 30 May 2012 10:44:56 +0000 (12:44 +0200)]
Automatic update of common submodule
From
ec1c4a8 to
92b7266
Sebastian Dröge [Wed, 30 May 2012 09:18:39 +0000 (11:18 +0200)]
Automatic update of common submodule
From
3429ba6 to
ec1c4a8
Tim-Philipp Müller [Tue, 29 May 2012 07:48:36 +0000 (08:48 +0100)]
taglist: add guards to make sure taglist is writable when modifying it
Now that taglists are refcounted we need to check if they're
writable before modifying them.
Tim-Philipp Müller [Mon, 28 May 2012 22:54:10 +0000 (23:54 +0100)]
taglist: avoid unnecessary string copying when registering tags
Add gst_tag_register_static() - no need to copy all those
string constants, whether translated or not.
API: gst_tag_register_static()
Tim-Philipp Müller [Sun, 27 May 2012 23:08:56 +0000 (00:08 +0100)]
check: check for GLib-GIO criticals as well
Tim-Philipp Müller [Sun, 27 May 2012 23:08:18 +0000 (00:08 +0100)]
gst_tag_list_free() -> gst_tag_list_unref()
Tim-Philipp Müller [Sun, 27 May 2012 22:58:27 +0000 (23:58 +0100)]
taglist: make GstTagList a GstMiniObject
Which adds refcounting support, and other things.
Tim-Philipp Müller [Sun, 27 May 2012 19:31:30 +0000 (20:31 +0100)]
caps: log freeing of caps at same log level as creation, i.e. TRACE
Tim-Philipp Müller [Sat, 26 May 2012 10:37:49 +0000 (11:37 +0100)]
message, event: update for tag lists not being structures any more
Tim-Philipp Müller [Sun, 20 May 2012 23:31:29 +0000 (00:31 +0100)]
taglist: make proper struct not just a GstStructure
Josep Torra [Fri, 25 May 2012 08:28:40 +0000 (10:28 +0200)]
debugutils: Fix static linking on OS X
The linking behaviour of external variables that are not initialized
in the compilation unit where they are defined is undefined. On OS X
this causes a linking failure when statically linking GStreamer.