Occasionally this test would fail, especially if the system is under load,
because the position query would pick up the last position from the
last buffer timestamp which has a lower timestamp than what we're
looking for. The sleep is long enough, however. It's unclear to me why
exactly this happens but there seems to be some kind of scheduling
issue going on as the streaming thread floods the sink with buffers.
Let's throttle the fakesrc to 100 buffers per second and make the sink
sync to the clock to restore some sanity. It should be totally sufficient
to test what we want to test, and seems to make things reliable here.
When holding a regular ref it will cause the GstBus to never
reach 0 references and it won't be destroyed unless the application
explicitly calls gst_bus_remove_signal_watch().
Switching to weakref will allow the GstBus to be destroyed.
The application is still responsible for destroying the
GSource.
https://bugzilla.gnome.org/show_bug.cgi?id=762552
Clear the initial floating ref in the init function for
busses and clocks. These objects can be set on multiple
elements, so there's no clear parent-child relationship
here. Ideally we'd just not make them derive from
GInitiallyUnowned at all, but since we want to keep
using GstObject features for debugging, we'll just do
it like this.
This should also fix some problems with bindings, which
seem to get confused when they get floating refs from
non-constructor functions (or functions annotated to
have a 'transfer full' return type). This works now:
from gi.repository import GObject, Gst
GObject.threads_init()
Gst.init(None)
pipeline=Gst.Pipeline()
bus = pipeline.get_bus()
pipeline.set_state(Gst.State.NULL)
del pipeline;
https://bugzilla.gnome.org/show_bug.cgi?id=679286https://bugzilla.gnome.org/show_bug.cgi?id=657202
Better now than later in the cycle. These might come in handy:
sed -i -e 's/GstProbeReturn/GstPadProbeReturn/g' `git grep GstProbeReturn | sed -e 's/:.*//' | sort -u`
sed -i -e 's/GST_PROBE_/GST_PAD_PROBE_/g' `git grep GST_PROBE_ | sed -e 's/:.*//' | sort -u`
sed -i -e 's/GstProbeType/GstPadProbeType/g' `git grep GstProbeType | sed -e 's/:.*//' | sort -u`
Original commit message from CVS:
* tests/check/gst/gstpipeline.c:
Extending the test a little to verify that we also get the NULL state-
change message.
Original commit message from CVS:
* gst/gst.c: (init_post), (gst_deinit):
* tests/check/gst/gstpipeline.c: (GST_START_TEST), (pipeline_thread),
(test_concurrent_create), (gst_pipeline_suite):
Ref some more classes in gst_init() to work around thread-safety
issues in pre-2.16 GLibs, and add basic unit test.
Original commit message from CVS:
2005-11-15 Andy Wingo <wingo@pobox.com>
* gst/gstelement.c (gst_element_set_base_time): Add debugging.
* gst/gstpipeline.c (gst_pipeline_set_new_stream_time): Document
using GST_CLOCK_TIME_NONE to disable base time management.
(do_pipeline_seek, gst_pipeline_change_state): Don't reset stream
time if it was NONE before.
(gst_pipeline_change_state): Only munge the base time if
stream_time != GST_CLOCK_TIME_NONE.
* check/gst/gstpipeline.c (test_base_time): Punt around the
problem of the probe not being called, because that's not the
issue I'm looking at. Add a check that setting stream_time to NONE
disables base time management.
Original commit message from CVS:
2005-11-15 Andy Wingo <wingo@pobox.com>
* check/gst/gstpipeline.c (test_base_time): Punt around the
problem of the probe not being called, because that's not the
issue I'm looking at...
Original commit message from CVS:
2005-11-15 Andy Wingo <wingo@pobox.com>
* check/gst/gstpipeline.c (test_base_time): Add check that the
base time and stream time are reset correctly.
Original commit message from CVS:
2005-11-15 Andy Wingo <wingo@pobox.com>
* check/gst/gstpipeline.c (test_base_time): New test for buffer
timestamps with live capture.
* gst/elements/gstfakesrc.c (gst_fake_src_create): If the datarate
is 0 but we are a live source, timestamp the buffers using the
element's clock.
Original commit message from CVS:
* check/gst/gstbin.c: (GST_START_TEST):
* check/gst/gstmessage.c: (GST_START_TEST):
* check/gst/gstpipeline.c: (GST_START_TEST), (message_received):
* gst/gstbin.c: (gst_bin_class_init), (gst_bin_recalc_func),
(bin_bus_handler):
* gst/gstelement.c: (gst_element_abort_state),
(gst_element_commit_state), (gst_element_lost_state):
* gst/gstmessage.c: (gst_message_new_state_changed),
(gst_message_new_state_dirty), (gst_message_new_segment_start),
(gst_message_new_segment_done), (gst_message_new_duration),
(gst_message_parse_state_changed),
(gst_message_parse_segment_start),
(gst_message_parse_segment_done), (gst_message_parse_duration):
* gst/gstmessage.h:
* tools/gst-launch.c: (event_loop):
Seriously, this is better than a previous commit as we only need
to notify the fact that an element changed state in a streaming
thread, marking the state of the parents dirty, hence the
STATE_DIRTY message instead of abusing a boolean in a STATE_CHANGE
message.
Original commit message from CVS:
* check/gst/gstbin.c: (GST_START_TEST):
* check/gst/gstmessage.c: (GST_START_TEST):
* check/gst/gstpipeline.c: (GST_START_TEST), (message_received):
* gst/gstbin.c: (bin_bus_handler):
* gst/gstelement.c: (gst_element_commit_state),
(gst_element_lost_state):
* gst/gstmessage.c: (gst_message_new_state_changed),
(gst_message_new_segment_start), (gst_message_new_segment_done),
(gst_message_new_duration), (gst_message_parse_state_changed),
(gst_message_parse_segment_start),
(gst_message_parse_segment_done), (gst_message_parse_duration):
* gst/gstmessage.h:
* tools/gst-launch.c: (event_loop):
Make messages future proof.
state-change gets a flag if it was a message comming from the
streaming thread.
segment-start/stop can also be specified in other formats.
A message to notify an app that a pipeline changed playback
duration.
Also fix a GstMessage leak in -launch
Original commit message from CVS:
* check/gst/gstbin.c: (GST_START_TEST):
* check/gst/gstmessage.c: (GST_START_TEST):
* check/gst/gstpipeline.c: (GST_START_TEST), (message_received):
* gst/gstelement.c: (gst_element_commit_state),
(gst_element_lost_state):
* gst/gstmessage.c: (gst_message_new_state_changed),
(gst_message_parse_state_changed):
* gst/gstmessage.h:
* tools/gst-launch.c: (event_loop):
Added extra field to STATE_CHANGE message with the pending
state, which will be different from the new state soon.
Original commit message from CVS:
* check/gst/gstbin.c:
* check/states/sinks.c:
fix tests for the new warning
* check/gst/gstpipeline.c:
add a test for pipeline and bus interaction
* gst/gstelement.c:
elements should be NULL if they get disposed; add a warning if not
Original commit message from CVS:
* check/gst/gstpipeline.c: (GST_START_TEST), (gst_pipeline_suite):
add another check
* gst/gstbus.c:
add some doc
* gst/gstinfo.c: (_gst_debug_init):
slightly more readable color for refcount debugging
Original commit message from CVS:
* check/gst/gstbin.c: (pop_messages), (GST_START_TEST):
* check/gst/gstbus.c: (message_func_eos), (message_func_app),
(send_messages), (GST_START_TEST), (gstbus_suite):
* check/gst/gstpipeline.c: (GST_START_TEST):
* check/pipelines/cleanup.c: (run_pipeline):
* check/pipelines/simple_launch_lines.c: (run_pipeline),
(GST_START_TEST):
* gst/gstbus.c: (gst_bus_have_pending), (gst_bus_source_prepare),
(gst_bus_source_check), (gst_bus_source_dispatch),
(gst_bus_create_watch), (gst_bus_add_watch_full),
(gst_bus_add_watch), (poll_func), (poll_timeout), (gst_bus_poll):
* gst/gstbus.h:
* tools/gst-launch.c: (event_loop):
* tools/gst-md5sum.c: (event_loop):
GstBusHandler -> GstBusFunc, return value has the same meaning as
any other GSource (FALSE == remove source).
_add_watch() and _add_watch_full() now take a MessageType mask to
only handle specific types of messages.
_poll() returns the GstMessage instead of the message type to avoid
race conditions.
_have_pending() takes a MessageType mask now too.
Added testsuite for multiple bus watches.
Fix testsuites and applications for new bus API.
Original commit message from CVS:
* check/gst/gstpipeline.c: (GST_START_TEST):
* docs/gst/gstreamer-sections.txt:
* gst/gstutils.c: (set_state_async_thread_func),
(gst_element_set_state_async):
* gst/gstutils.h:
add a gst_element_set_state_async method that
sets the state and starts a thread to make sure the state
change completes as best as it can
Original commit message from CVS:
* check/Makefile.am:
* check/gst/.cvsignore:
* check/gst/gstpipeline.c: (pop_messages), (GST_START_TEST),
(gst_pipeline_suite), (main):
adding a test for pipelines and state changes
* gst/gstutils.c: (get_state_func):
add some debugging
* gstreamer.spec.in:
fix up spec file
