=== release 1.4.0 === 2014-10-20 Thibault Saunier * configure.ac: releasing 1.4.0 === release 1.3.90 === 2014-09-24 11:10:03 +0200 Thibault Saunier * ChangeLog: * NEWS: * RELEASE: * configure.ac: Release 1.3.90 2014-08-06 10:23:56 +0100 Tim-Philipp Müller * tests/check/gnl/gnlcomposition.c: tests: fix 'Attempt to unlock mutex that was not locked' in gnlcomposition test Locking is all a bit weird in this test, but at least it doesn't abort any longer. 2014-08-01 10:43:18 +0200 Edward Hervey * Makefile.am: * common: Makefile: Add usage of build-checks step Allows building checks without running them 2014-06-25 17:51:02 +0100 Tim-Philipp Müller * configure.ac: configure: lower GLib requirement to 2.32 We don't actually require 2.34, and 2.32 is what the other modules currently require. 2014-06-24 12:17:51 +0200 Thibault Saunier * gnl/gnlobject.c: gnlobject: Commit object in READY_TO_PAUSED Making it possible to use Gnl with gst-launch (with patch from #732149) 2014-05-19 13:12:56 +0200 Thibault Saunier * gnl/gnlcomposition.c: * tests/check/gnl/gnlcomposition.c: composition: Add a property defining deactivated elements state In the composition many elements are not used, until now, they were always set to the PAUSED state though it makes sense to be able to do gapless playback it is sometimes not ideal to preserve memory usage and it means the more children the composition has the more threads are used and it is very simple to reach the kernel hard limit number of threads. In order to avoid that to happen, we set default state of unused thread to READY and we add a property to the composition so that users can override that default behaviour. + Make sure we send "stream-start" again when changing our ghostpad to avoid "Data flow before stream-start assertion" + Fix unit tests, do not check refcount when changing state as it can vary. https://bugzilla.gnome.org/show_bug.cgi?id=596849 2014-05-21 10:53:53 +0200 Sebastian Dröge * common: Automatic update of common submodule From 211fa5f to 1f5d3c3 2014-05-07 16:18:40 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: composition: Drop buffers until we get seeked Make sure we don't let buffers pass before our entry got correctly seeked and had a segment when updating the pipeline 2014-05-06 16:37:24 +0200 Thibault Saunier * tests/check/gnl/common.c: * tests/check/gnl/common.h: * tests/check/gnl/seek.c: tests: Add a flag about whether to keep expected segments or not Most test rely on the common helper to handles that, but for seeking tests it does not make sense to handle it that way (as in gnlcomposition we send 2 seeks when we get seek and thus we can received 2 times the same segment) 2014-05-06 15:37:41 +0200 Thibault Saunier * tests/check/gnl/common.c: * tests/check/gnl/seek.c: tests: By design, we can get twice the same GstSegment out Handle it in the tests as it is not critical 2014-05-06 09:37:31 +0200 Thibault Saunier * tests/check/gnl/seek.c: tests:seek: Wait for ASYNC_DONE between seek execution 2014-05-03 10:17:48 +0200 Sebastian Dröge * common: Automatic update of common submodule From bcb1518 to 211fa5f 2014-03-31 23:52:17 +0100 Tim-Philipp Müller * .gitignore: .gitignore log files from autotools 'make check' test-driver 2014-03-31 21:12:24 +0100 Tim-Philipp Müller * tests/check/gnl/gnlcomposition.c: tests: pass 64-bit integers to 64-bit integer properties 2014-03-16 12:50:24 +0100 Thibault Saunier * configure.ac: Back to development === release 1.2.0 === 2014-03-16 12:49:23 +0100 Thibault Saunier * ChangeLog: * NEWS: * RELEASE: * configure.ac: Release 1.2.0 2014-02-28 09:36:24 +0100 Sebastian Dröge * common: Automatic update of common submodule From fe1672e to bcb1518 2014-02-26 22:15:19 +0100 Stefan Sauer * common: Automatic update of common submodule From 1a07da9 to fe1672e 2014-01-30 10:45:29 +0100 Edward Hervey * common: Automatic update of common submodule From d48bed3 to 1a07da9 2013-12-30 16:29:54 +0100 Edward Hervey * gnl/gnloperation.c: gnloperation: Remove dead code ret is only set before leaving the loop. COVERITY CID 1139661 COVERITY CID 1139662 2013-12-22 22:35:42 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From dbedaa0 to d48bed3 2013-11-29 09:50:33 +0100 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Fix typo 2013-11-05 11:22:15 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 865aa20 to dbedaa0 2013-09-28 21:06:41 +0200 Thibault Saunier * configure.ac: Back to development === release 1.1.90 === 2013-09-24 14:39:47 +0200 Thibault Saunier * ChangeLog: * configure.ac: * docs/libs/gnonlin.hierarchy: * docs/libs/gnonlin.interfaces: * docs/libs/gnonlin.prerequisites: * docs/libs/gnonlin.signals: * docs/libs/inspect-build.stamp: * docs/libs/inspect.stamp: * docs/libs/inspect/plugin-gnonlin.xml: Release 1.1.90 2013-09-24 18:34:58 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 6b03ba7 to 865aa20 2013-09-20 16:18:30 +0200 Edward Hervey * common: Automatic update of common submodule From b613661 to 6b03ba7 2013-09-19 18:45:49 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 74a6857 to b613661 2013-09-19 17:38:45 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 01a7a46 to 74a6857 2013-08-28 14:48:58 -0400 Thibault Saunier * gnl/gnlcomposition.c: composition: Do not update the pipeline if we are in NULL state And always update start/duration when removing an object from the composition 2013-08-26 16:26:31 -0400 Thibault Saunier * gnl/gnlcomposition.c: composition: Release objects lock while forwarding an event There is no reason to keep it and in some rare cases it creates deadlocks as followed: t1: → Composition receives a flushing seek, it takes the OBJECTS_LOCK and fowards the flushing seek event upstream → adder receives the seek and set its collectpad to flushing This implies tacking STREAM_LOCK (collectpad) t2: → Collectpad has buffers ready, and has the STREAM_LOCK (collectpad) and is EOS, so it sends it downstream → The composition receives EOS, and needs to check if it is the actual EOS or not, thus need to take the OBJECTS_LOCK This create a deadlock, and in the first stage, we did not need the OBJECTS_LOCK to forward downstream the flushing seek, so do not take it. https://bugzilla.gnome.org/show_bug.cgi?id=706831 2013-08-24 22:15:15 -0400 Thibault Saunier * gnl/gnlcomposition.c: composition: Validate stack right after it has been re computed 2013-08-15 21:57:49 +0200 Thibault Saunier * gnl/gnlcomposition.c: composition: Implement a separate probe to drop data. It ensures that all data that still flows while we update the pipeline is dropped, otherwise we have races where we could get buffer from the old pipeline flowing into elements that are already fully flushed (and have no segment caps, etc... set) Also validate the stack a little later. This commits finnishes what had been started with a24a3be13b81c5d0080c061edf61daece6341f95 2013-08-13 18:20:27 +0200 Thibault Saunier * gnl/gnlcomposition.c: gnlcomposition: Drop any buffer flowing from a source when updating pipeline There is a race where we can have an EOS from the element which srcpad is our ghost pad but other elements upstream are still outputing data and we flush the element so its pads have no segment info anymore, and we end up having warning about flow before segement event. The solution here is that we do not let any data flowing when we are updating the pipeline. 2013-08-02 14:19:28 +0200 Lubosz Sarnecki * configure.ac: build: add subdir-objects to AM_INIT_AUTOMAKE Fixes warnings with automake 1.14 https://bugzilla.gnome.org/show_bug.cgi?id=705350 2013-01-22 14:40:15 +0000 Tim-Philipp Müller * gnl/gnloperation.c: gnloperation: check for request pads on the element class not the factory When checking for request pads as fallback, just look at the element's class, not the factory, since there might not be a factory (in case of python elements) or the factory might be the wrong one (in case of a GstBin sub-class) and doesn't have complete information. https://bugzilla.gnome.org/show_bug.cgi?id=582244 2013-06-05 15:14:54 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: gnlcomposition: grammar (childs -> children) https://bugzilla.gnome.org/show_bug.cgi?id=701647 2013-07-09 23:16:53 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: composition: Check if we have an entry before trying to get its probeid The object can be removed at the time the pad is removed, avoid segfaulting when that happens. 2013-06-26 17:25:37 -0400 Thibault Saunier * gnl/gnlcomposition.c: composition: Add entry to the the hashtable before connecting to pad-added Avoiding races where the pad would be added right between the connection and inserting to the hashtable 2013-06-13 20:33:00 -0400 Thibault Saunier * gnl/gnlcomposition.c: * tests/check/gnl/complex.c: gnlcomposition: Remove support for gaps The way we were handling gap was weird and not natural at all for users, remove it for now. In the long term, we should have a proper way of filling gaps in the API and for now only emit an STREAM_ERROR message on the bus when a gap is detected. The user is now responsible for filling gaps in the composition https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-06-13 19:39:55 -0400 Thibault Saunier * gnl/gnlcomposition.c: gnlcomposition: Remove uneeded indents Going from if (something) { do_something(); } else { } to if (!something) return; do_something (); 2013-06-13 19:33:46 -0400 Thibault Saunier * gnl/gnlcomposition.c: gnlcomposition: Make the pipeline update protected from any child changes When calling the gnl_object_commit method, we need to make sure that no object can be changed and, it should be true until the pipeline has fully been udpated. We now need the update_pipeline function to be called with the COMP_OBJECTS_LOCK taken, and we do not unlock that lock during the whole pipeline updating process. https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-06-13 18:51:15 -0400 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlobject.c: gnl: Setting priority == MAXUINT32 does not mean "default object" This was how expandables worked prior to the expandable property. Using MAXUINT32 has some limitation and which are fully covered by the expandable property. Drop this support, as we will plan on implementing an automatic way of filling gaps. https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-06-07 18:58:24 -0400 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlobject.c: * gnl/gnlobject.h: * tests/check/gnl/common.c: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: gnl: First implementation of the commit based API Currently, the only safe way to modify a composition, or an object, is to pause the pipeline first. We think that this limitation eliminates many use cases for GNonLin. Few video editors let you create a draft of your video by applying multi-cam technic. Also people could want to eventually consider doing live video production, which obviously require being able to change the composition in playing state (and increase compositon duration as we go). This commit is a first step making GNL fully commit base. Objects, Operations and Compositon keep a copy of modified values, and only apply those changes when a commit is executed. This potentially allows making the changes (or group of changes) fully 'atomic' without having to pause the pipeline. Added API: ---------- GnlObject::commit action signal Removed API: ------------ GnlComposition:update property + Fix the tests https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-06-07 18:17:25 -0400 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlobject.c: gnl: Remove trailling whitespaces and tabs in .c files 2013-06-07 11:39:17 -0400 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlobject.c: * gnl/gnlobject.h: * tests/check/gnl/common.c: * tests/check/gnl/common.h: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: gnl: Remove the notion of media-duration Rational described at https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-05-30 17:48:47 -0400 Thibault Saunier * docs/libs/gnonlin.args: * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlobject.c: * gnl/gnlobject.h: * tests/check/gnl/common.c: * tests/check/gnl/common.h: * tests/check/gnl/gnlcomposition.c: gnl: Rename media-start as inpoint Rational described at https://bugzilla.gnome.org/show_bug.cgi?id=701287 2013-06-11 18:22:11 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: composition: update operations base time even when we don't update the pipeline. 2013-05-29 16:55:34 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: gnlcomposition: forward non time-resetting flush stops outside the composition, except those consecutive to a seek event. The 'reset-time' parameter has been added in 1.0, and we need it to be FALSE outside the composition. Related to https://bugzilla.gnome.org/show_bug.cgi?id=701146 2013-05-28 19:09:23 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: gnlcomposition: Don't flush_start / flush_stop according to user's flags. We do it ourselves when needed, doing so led the pipeline to lose state, which was updating the base time. fixes https://bugzilla.gnome.org/show_bug.cgi?id=701146 2013-06-10 17:46:51 -0400 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnloperation.c: * gnl/gnloperation.h: gnl: Properly set operations basetime For operations (especially mixers) to be able to do proper synchronization they need the segment from different sources to be in the same running time context. This commit makes sure that the basetime of the incoming segments of operations will be the same, and equal to the duration of the already "played" time of the operation at the position at which the new portion of the composition starts. 2013-06-05 15:17:37 +0200 Sebastian Dröge * common: Automatic update of common submodule From 098c0d7 to 01a7a46 2013-05-27 20:52:36 +0200 Mathieu Duponchelle * tests/check/Makefile.am: * tests/check/gnl/common.c: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: tests: sanitization. + Create a no_install libtestutils. + Remove code from common.h to put it in common.c + Abstract away bus polling from test_simplest # TODO use it as much as possible. + Fix various little errors spotted thanks to new flags. 2013-05-25 03:05:40 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: gnlcomposition: When removing a ghostpad, remove the probe_handlers associated to its target. 2013-05-25 03:43:20 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: * gnl/gnlcomposition.h: gnlcomposition: replace eos_main_thread with update_pipeline_func. + This function is called in a thread of its own. 2013-05-24 16:23:26 +0200 Mathieu Duponchelle * gnl/gnlcomposition.c: gnlcomposition: Forward EOS from the streaming thread when appropriate. 2013-04-26 18:14:36 +0200 Nicolas Dufresne * .gitignore: Cleanup and fix gitignore 2012-09-11 18:35:17 -0400 Nicolas Dufresne * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: * gnl/gnloperation.c: * gnl/gnlsource.c: ghostpad: Remove unsued parameter flush_hack As a result, remove unused gnl_object_ghost_pad_full. 2013-01-25 10:35:22 -0500 Nicolas Dufresne * gnl/gnlghostpad.c: ghostpad: Remove uneeded set_caps() Linking the element should automatically resend all sticky events. 2013-01-09 11:54:36 -0500 Nicolas Dufresne * gnl/gnlghostpad.c: ghostpad: Correctly handle pad creation failure We where trying to activate the pad even if creating the ghostpad failed. Instead, warn if that fails, and return NULL as expected. 2012-09-12 12:43:07 -0400 Nicolas Dufresne * gnl/gnlsource.c: source: Split change_state into prepare/cleanup 2012-09-12 12:37:04 -0400 Nicolas Dufresne * gnl/gnlobject.c: object: Fix function declaration coding style 2012-09-11 19:28:55 -0400 Nicolas Dufresne * gnl/gnlobject.c: doc: Add missing description 2012-09-11 19:28:42 -0400 Nicolas Dufresne * gnl/gnlcomposition.c: * gnl/gnloperation.c: * gnl/gnlsource.c: * gnl/gnlurisource.c: doc: Remove duplicate short description nowadays the plugin description is used. 2012-09-11 19:09:30 -0400 Nicolas Dufresne * docs/libs/gnonlin-sections.txt: * docs/libs/gnonlin.args: * docs/libs/gnonlin.hierarchy: * docs/libs/gnonlin.interfaces: * docs/random/design: Drop remaining references to GnlFileSource 2013-05-23 21:09:58 -0400 Thibault Saunier * gnl/gnlcomposition.c: Revert "composition: Foward eos from the streaming thread when appropriate" This reverts commit 9db8b0211bf64137ff3215b78172fe76de1994a1. This should be implemented properly using our knowledge of the composition's content 2013-05-23 20:35:58 -0400 Thibault Saunier * gnl/gnlcomposition.c: composition: Foward eos from the streaming thread when appropriate 2013-05-19 12:06:40 -0400 Thibault Saunier * tests/check/gnl/common.h: tests: Use spaces instead of tabs 2013-05-16 00:16:18 -0400 Thibault Saunier * gnl/gnlcomposition.c: gnlcomposition: Only send stream-start once and for each source start We concider the output of a gnlcomposition as 1 coherente stream. 2013-05-05 11:24:45 +0100 Thibault Saunier * gnl/gnlcomposition.c: * gnl/gnlobject.c: Remove useless GLib version checks 2013-04-12 18:16:40 -0300 Thibault Saunier * tests/check/gnl/gnlcomposition.c: tests: Add a simple test with adder 2013-05-15 10:54:24 +0200 Sebastian Dröge * common: Automatic update of common submodule From 5edcd85 to 098c0d7 2013-04-22 23:55:18 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 3cb3d3c to 5edcd85 2012-10-24 12:17:28 +0200 Sebastian Dröge * configure.ac: * gnl/Makefile.am: gst: Add better support for static plugins 2013-04-14 17:57:45 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From aed87ae to 3cb3d3c 2013-04-09 21:02:21 +0200 Stefan Sauer * common: Automatic update of common submodule From 04c7a1e to aed87ae 2013-03-28 19:07:30 +0100 Edward Hervey * gnl/gnlcomposition.c: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: composition: Ensure segment base time is correct We update it based on accumulated segment time. Tests are also updated to check for valid base/time/duration 2013-03-28 17:26:52 +0000 Nicolas Dufresne * gnl/gnlcomposition.c: gnlcomposion: When seeking flush downstream before unlinking The 'update' parameter in seek_handling is used to know if flush most be performed before unlinking or linking new objects. It accidently set to TRUE d196cd3cdc850f2355bb16518e22ccfda1170078. 2013-03-28 16:26:25 +0000 Nicolas Dufresne * gnl/gnlcomposition.c: gnlcomposion: objects lock must be taken to call _set_target() gnl_composition_ghost_pad_set_target() require the objects_lock to be taken. Aslo improve protection around user_seek_flush. 2013-03-07 00:03:43 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 2de221c to 04c7a1e 2013-01-30 15:12:55 +0000 Luis de Bethencourt * configure.ac: configure: update to 1.0 dependency 2013-01-28 20:44:59 +0100 Stefan Sauer * common: Automatic update of common submodule From a942293 to 2de221c 2013-01-16 12:30:11 +0000 Tim-Philipp Müller * docs/libs/Makefile.am: * tests/check/Makefile.am: tests, docs: 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. 2013-01-15 15:08:44 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From a72faea to a942293 2012-12-30 23:19:05 +0000 Kerrick Staley * configure.ac: configure: replace deprecated AM_CONFIG_HEADER with AC_CONFIG_HEADERS AM_CONFIG_HEADER has been removed in the just-released automake 1.13: https://lists.gnu.org/archive/html/automake/2012-12/msg00038.html https://bugzilla.gnome.org/show_bug.cgi?id=690881 2012-11-28 15:59:43 +0100 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Add support for reverse playback Conflicts: gnl/gnlcomposition.c 2012-11-28 15:49:37 +0100 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Make sure to pass flushes downstream if flushing seeks were received Conflicts: gnl/gnlcomposition.c 2012-11-19 11:30:01 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 6bb6951 to a72faea 2012-11-07 12:30:51 +0100 Edward Hervey * tests/check/gnl/common.h: check: Detect stray segments better 2012-11-04 00:13:12 +0000 Tim-Philipp Müller * COPYING: * COPYING.LIB: * LICENSE: * gnl/gnl.c: * gnl/gnl.h: * gnl/gnlcomposition.c: * gnl/gnlcomposition.h: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: * gnl/gnlobject.c: * gnl/gnlobject.h: * gnl/gnloperation.c: * gnl/gnloperation.h: * gnl/gnlsource.c: * gnl/gnlsource.h: * gnl/gnltypes.h: * gnl/gnlurisource.c: * gnl/gnlurisource.h: * tests/check/gnl/gnlcomposition.c: Fix FSF address 2012-11-03 18:16:05 +0100 Edward Hervey * tests/check/gnl/seek.c: check: Re-enable more tests Event though they don't succeed due to some issues in videomixer segment handling (or lack thereof). 2012-11-03 18:15:20 +0100 Edward Hervey * tests/check/gnl/gnlcomposition.c: check: Fix usage of probe We only care about downstream data, and we only need to be called once 2012-11-03 18:13:32 +0100 Edward Hervey * gnl/gnlghostpad.c: * tests/check/gnl/common.h: misc: Add more/better debugging 2012-11-03 18:09:39 +0100 Edward Hervey * gnl/gnloperation.c: gnloperation: Only release sink pads that need to be 2012-11-03 18:08:37 +0100 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Control internal pad straight away The behaviour changed in 1.0, the internal (proxy) pad is always present for a ghostpad. We therefore need to control straight away to avoid races between the mode it is linked with the target and the moment we set custom pad functions. 2012-11-03 18:03:58 +0100 Edward Hervey * gnl/gnlcomposition.c: composition: Only care about source pads that go away sink pads are handled by operations 2012-09-25 14:22:54 +0200 Edward Hervey * gnl/gnlcomposition.c: composition: Ensure seeks always have flush/accurate Otherwise we will never be able to unblock upstream elements 2012-09-25 14:21:55 +0200 Edward Hervey * gnl/gnlghostpad.c: events: Don't leak segments and carry-over seqnums Even though we transform events, we carry-over the seqnum 2012-09-25 14:21:12 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Fix leak 2012-10-17 17:51:49 +0100 Tim-Philipp Müller * gnl/gnlcomposition.c: * gnl/gnloperation.c: * gnl/gnlsource.c: * gnl/gnlurisource.c: gnl: use gst_element_class_set_static_metadata() 2012-10-06 15:01:28 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 6c0b52c to 6bb6951 2012-09-22 16:10:51 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 4f962f7 to 6c0b52c 2012-09-03 13:05:34 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Update for API change 2012-08-22 13:32:51 +0200 Stefan Sauer * common: Automatic update of common submodule From 668acee to 4f962f7 2012-08-05 16:43:01 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 94ccf4c to 668acee 2012-07-23 08:47:43 +0200 Sebastian Dröge * common: Automatic update of common submodule From 98e386f to 94ccf4c 2012-07-05 13:24:41 +0200 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Implement segment-done event 2012-06-11 10:46:17 +0200 Edward Hervey * configure.ac: configure: Remove -Wcast-align See https://bugzilla.gnome.org/show_bug.cgi?id=615698 2012-06-08 15:06:46 +0200 Edward Hervey * common: Automatic update of common submodule From 03a0e57 to 98e386f 2012-06-06 18:20:27 +0200 Edward Hervey * common: Automatic update of common submodule From 1fab359 to 03a0e57 2012-06-01 10:30:37 +0200 Edward Hervey * common: Automatic update of common submodule From f1b5a96 to 1fab359 2012-05-31 13:10:50 +0200 Sebastian Dröge * common: Automatic update of common submodule From 92b7266 to f1b5a96 2012-05-30 12:48:11 +0200 Sebastian Dröge * common: Automatic update of common submodule From ec1c4a8 to 92b7266 2012-05-30 11:26:42 +0200 Sebastian Dröge * common: Automatic update of common submodule From 3429ba6 to ec1c4a8 2012-05-13 15:58:20 +0200 Sebastian Dröge * common: Automatic update of common submodule From dc70203 to 3429ba6 2012-04-17 17:59:41 +0400 Руслан Ижбулатов * gnl/gnlcomposition.c: gnlcomposition: fix printf format in debug message https://bugzilla.gnome.org/show_bug.cgi?id=675276 2012-04-24 15:38:44 +0200 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Use correct enum type 2012-04-16 09:11:12 +0200 Sebastian Dröge * common: Automatic update of common submodule From 6db25be to dc70203 2012-04-13 13:54:59 +0200 Sebastian Dröge * autogen.sh: * configure.ac: configure: Modernize autotools setup a bit Also we now only create tar.bz2 and tar.xz tarballs. 2012-04-13 13:39:02 +0200 Sebastian Dröge * common: Automatic update of common submodule From 464fe15 to 6db25be 2012-04-05 18:45:02 +0200 Sebastian Dröge * common: Automatic update of common submodule From 7fda524 to 464fe15 2012-04-05 17:40:34 +0200 Sebastian Dröge * gnl/gnl.c: gst: Update for GST_PLUGIN_DEFINE() API changes 2012-04-04 14:52:13 +0200 Sebastian Dröge * configure.ac: * docs/libs/gnonlin-docs.sgml: * docs/version.entities.in: * gnl/Makefile.am: * gnonlin.spec.in: gnl: Update versioning 2012-03-26 12:40:13 +0200 Wim Taymans Replace master with 0.11 2012-03-13 19:49:36 +0000 Tim-Philipp Müller Merge remote-tracking branch 'origin/master' into 0.11 Conflicts: configure.ac gnl/gnlfilesource.c gnl/gnlsource.c tests/check/gnl/common.h 2012-03-13 19:32:08 +0000 Tim-Philipp Müller * configure.ac: * gnl/.gitignore: * gnl/Makefile.am: * gnl/gnloperation.c: Remove gstmarshal.[ch] completely and use the generic marshaller 2012-03-06 15:33:49 +0100 Sebastian Dröge * gnl/gnlobject.h: gnl: Fix 'signed shift result (0x20000000000) requires 43 bits to represent, but 'int' only has 32 bits' compiler warning Seems we don't have that many flags left... 2012-01-25 18:13:35 +0100 Edward Hervey * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/seek.c: tests: don't use deprecated API 2012-01-25 18:13:15 +0100 Edward Hervey * configure.ac: * gnl/gnlcomposition.c: * gnl/gnlsource.c: gnl: Don't use deprecated GMutex/GThread API And depend on same glib version as 0.11 core 2012-01-25 14:12:03 +0100 Thomas Vander Stichele * common: Automatic update of common submodule From c463bc0 to 7fda524 2012-01-25 11:40:36 +0100 Sebastian Dröge * common: Automatic update of common submodule From 2a59016 to c463bc0 2012-01-18 16:48:22 +0100 Sebastian Dröge * common: Automatic update of common submodule From 0807187 to 2a59016 2012-01-13 12:32:29 +0000 Tim-Philipp Müller * tests/check/gnl/common.h: tests: don't redefine fail_unless_equals_int64() Fixes compiler warnings with recent GStreamer versions. 2012-01-13 12:19:31 +0000 Tim-Philipp Müller * gnl/gnlfilesource.c: gnlfilesource: unescape URIs when converting an URI into a filename URIs may contain escapes, which need to be unescaped when creating a filename (even if we pass an URI that's unescaped, we may get back an escaped one, can't really make any assumptions). 2012-01-13 12:18:14 +0000 Tim-Philipp Müller * configure.ac: * gnl/gnlfilesource.c: gnlfilesource: use gst_filename_to_uri() to create a proper file URI This will create a correct URI even for relative paths, which are allowed for "location" properties. Bump GStreamer core/base requirement to 0.10.33 for that API. 2012-01-04 19:55:27 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 0546e58 to 0807187 2011-12-30 14:16:41 +0100 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlsource.c: * tests/check/gnl/common.h: * tests/check/gnl/gnlcomposition.c: More updates for probe changes 2011-11-24 00:38:21 +0000 Tim-Philipp Müller * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlsource.c: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: gnl, tests: update for pad probe API changes But unit tests time out after pad gets blocked, so still something missing somewhere. 2011-11-01 00:49:00 +0000 Tim-Philipp Müller * gnl/gnlcomposition.c: * gnl/gnlsource.c: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: Update for pad API changes GstProbeType, GstProbeReturn and GstActivateMode -> GstPad* 2011-10-06 10:50:12 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Don't use _accept_caps to figure out compatible pads It isn't reliable due to issues with gst_caps_is_subset. Instead we just check if a pad caps intersect with the target caps Ported from 0.11 (471611). Fixes bug #662312. Conflicts: gnl/gnlsource.c 2011-10-10 17:15:30 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Set ghostpad *after* sending seek event Avoids ending up with stray EOS events being copied over. 2011-10-10 17:14:21 +0200 Edward Hervey * tests/check/gnl/seek.c: check: Use a probe to detect events ... and not a pad block 2011-10-10 13:18:43 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Make debug message more informative 2011-10-10 13:18:32 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Fix iterator/GValue usage 2011-10-10 13:18:01 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: gnlcomposition: Refactor probe handling 2011-10-10 13:15:53 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlobject.c: * gnl/gnloperation.c: * gnl/gnlsource.c: gnl: Code Cleanup 2011-10-06 10:50:12 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Don't use _accept_caps to figure out compatible pads It isn't reliable due to issues with gst_caps_is_subset. Instead we just check if a pad caps intersect with the target caps 2011-10-06 10:49:15 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Don't remove probes that don't exist 2011-10-06 10:46:46 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Ref the proper pad And not the pointer to the pad 2011-10-05 11:02:44 +0200 Edward Hervey Merge remote-tracking branch 'origin/master' into 0.11 2011-10-05 11:00:25 +0200 Edward Hervey * tests/check/Makefile.am: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: tests: Update to new 0.11 API * Use GST_UNSTABLE_API * Remove unused variables * Properly disable tests that require videomixer if not present * Clarify failures with ERROR messages * Use new probe API Compiles ... but most tests fail due to caps issues (which is also present in 0.10) 2011-10-05 10:57:24 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: * gnl/gnloperation.c: * gnl/gnlsource.c: gnl: More porting to 0.11 API Now properly compiles * Switch to new probe API for pad probes and pad blocking * Switch to new GstIterator API * Switch to new GstSegment API * Switch to new GstEvent API (for SEGMENT and SEEK events) 2011-10-05 10:56:02 +0200 Edward Hervey * common: common: Switch to head of 0.11 branch 2011-09-07 15:56:12 +0200 Stefan Sauer * docs/libs/Makefile.am: docs: cleanup makefiles Remove commented out parts that we don't need. Remove "the wingo addition" - no so useful after all. Narrow down file-globs for plugin docs. 2011-09-06 21:53:17 +0200 Stefan Sauer * common: Automatic update of common submodule From a39eb83 to 11f0cd5 2011-09-06 16:06:47 +0200 Stefan Sauer * common: Automatic update of common submodule From 605cd9a to a39eb83 2011-08-05 13:12:48 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlobject.c: * gnl/gnloperation.c: * gnl/gnlsource.c: * gnl/gnlurisource.c: gnl: Remove GST_BOILERPLATE usage doesn't compile yet 2011-08-05 12:52:14 +0200 Edward Hervey * docs/libs/gnonlin-docs.sgml: * gnl/Makefile.am: * gnl/gnl.c: * gnl/gnl.h: * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: * gnl/gnltypes.h: * gnl/gnlurisource.h: gnlfilesource: is dead, long live gnlurisource 2011-08-05 12:42:00 +0200 Edward Hervey * common: * configure.ac: * gnl/Makefile.am: 0.11 branch is open \o/ 2011-06-23 11:29:50 -0700 David Schleef * common: Automatic update of common submodule From 69b981f to 605cd9a 2011-05-19 23:00:01 +0300 Stefan Kost * common: Automatic update of common submodule From 9e5bbd5 to 69b981f 2011-05-18 16:13:29 +0300 Stefan Kost * common: Automatic update of common submodule From fd35073 to 9e5bbd5 2011-05-18 12:26:54 +0300 Stefan Kost * common: Automatic update of common submodule From 46dfcea to fd35073 2011-04-24 14:06:15 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From c3cafe1 to 46dfcea 2011-01-27 17:32:28 +0100 Alessandro Decina * Android.mk: * gnl/Makefile.am: android: make it ready for androgenizer Remove the android/ top dir Fix the Makefile.am to be androgenized To build gstreamer for android we are now using androgenizer which generates the needed Android.mk files. Androgenizer can be found here: http://git.collabora.co.uk/?p=user/derek/androgenizer.git 2011-04-04 15:59:13 +0300 Stefan Kost * common: Automatic update of common submodule From 1ccbe09 to c3cafe1 2011-03-25 22:36:45 +0100 Sebastian Dröge * common: Automatic update of common submodule From 193b717 to 1ccbe09 2011-03-25 14:57:43 +0200 Stefan Kost * common: Automatic update of common submodule From b77e2bf to 193b717 2011-03-25 09:34:19 +0100 Sebastian Dröge * common: Automatic update of common submodule From d8814b6 to b77e2bf 2011-03-25 09:10:40 +0100 Sebastian Dröge * common: Automatic update of common submodule From 6aaa286 to d8814b6 2011-03-24 18:51:06 +0200 Stefan Kost * common: Automatic update of common submodule From 6aec6b9 to 6aaa286 2011-03-18 19:34:57 +0100 Luis de Bethencourt * autogen.sh: autogen: wingo signed comment 2011-02-28 20:26:33 +0100 Mark Nauwelaerts * configure.ac: configure.ac: export plugin description more platform independent Fixes #642504. 2011-02-28 18:34:36 +0100 Mark Nauwelaerts * common: Automatic update of common submodule From 1de7f6a to 6aec6b9 2011-02-24 12:50:01 +0100 Edward Hervey * gnl/gnloperation.c: gnloperation: Handle the cases where the sink pads disappear 2011-01-31 15:40:19 +0100 Edward Hervey * gnl/gnloperation.c: gnloperation: Synchronize all pads We were previously only removing one pad instead of synchronizing back down to the number of requested pads. 2011-01-31 15:39:52 +0100 Edward Hervey * gnl/gnloperation.c: gnloperation: Cleanup pads when going from PAUSED to READY 2011-02-14 12:55:36 +0200 Stefan Kost * common: Automatic update of common submodule From f94d739 to 1de7f6a 2011-02-10 19:11:57 +0100 Edward Hervey * tests/check/gnl/gnlcomposition.c: tests: Make gnlcomposition check stop faster 2011-02-10 18:51:41 +0100 Edward Hervey * tests/check/gnl/gnloperation.c: * tests/check/gnl/seek.c: * tests/check/gnl/simple.c: tests: Add doc and refine seek check 2011-02-10 18:51:08 +0100 Edward Hervey * tests/check/gnl/common.h: tests: Small cleanup 2011-02-08 13:45:03 +0100 Edward Hervey * tests/check/gnl/seek.c: tests: Added more seeking tests involving operations 2011-02-08 12:59:10 +0100 Edward Hervey * tests/check/gnl/gnloperation.c: check: refactor 'gnloperation' tests 2011-02-08 09:12:35 +0100 Edward Hervey * tests/check/gnl/common.h: * tests/check/gnl/complex.c: tests: refactor 'complex' test 2011-02-02 18:10:52 +0100 Edward Hervey * tests/check/Makefile.am: * tests/check/gnl/seek.c: tests: First go at testing seeking Still needs more checks though 2011-01-31 16:05:34 +0100 Edward Hervey * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: check: More debugging and state checking 2011-01-31 16:05:15 +0100 Edward Hervey * tests/check/gnl/common.h: tests: Add a method for checking element states 2011-01-29 12:52:30 +0100 Edward Hervey * tests/check/gnl/complex.c: * tests/check/gnl/gnloperation.c: check: Add some comments 2011-01-29 12:52:09 +0100 Edward Hervey * tests/check/gnl/simple.c: check: Add a simple variant 2011-01-24 13:34:38 +0100 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: clarify update_pipeline internal comments 2011-01-21 15:35:59 +0100 Edward Hervey * docs/random/part-gnlcomposition.txt: docs: Explanations about scheduling in gnlcomposition 2011-01-20 13:31:51 +0100 Edward Hervey * docs/random/design: docs: Re-indent design doc to 70 column and clarify 2011-01-25 11:17:47 +0100 Edward Hervey * configure.ac: configure.ac: And back to development we go === release 0.10.17 === 2011-01-20 20:56:23 +0100 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Release 0.10.17 2011-01-13 11:29:52 +0100 Edward Hervey * configure.ac: configure.ac: 0.10.16.2 pre-release 2011-01-11 15:52:11 +0200 Stefan Kost * common: Automatic update of common submodule From e572c87 to f94d739 2011-01-10 16:38:45 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From ccbaa85 to e572c87 2011-01-10 14:55:50 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 46445ad to ccbaa85 2011-01-05 17:19:57 +0100 Edward Hervey * tests/check/gnl/gnlcomposition.c: tests: Don't use huge durations in tests 2011-01-05 17:18:38 +0100 Edward Hervey * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/simple.c: tests: Poll more often. Should lower the risks of timing out 2010-12-20 17:48:15 +0100 Edward Hervey * common: Automatic update of common submodule From 169462a to 46445ad 2010-12-19 11:10:29 +0100 Sebastian Dröge * tests/check/gnl/gnlcomposition.c: gnlcomposition: Fix memory leak in the unit test 2010-12-17 12:28:14 +0100 Edward Hervey * gnl/gnlcomposition.c: GnlComposition: Use g_object_notify_by_pspec if available 2010-12-17 12:12:15 +0100 Edward Hervey * gnl/gnlobject.c: GnlObject: Use g_object_notify_by_pspec if available 2010-12-17 12:12:01 +0100 Edward Hervey * configure.ac: configure.ac: Bump GLib requirement to 2.22 In sync with gstreamer core 2010-12-15 14:57:18 +0200 Stefan Kost * common: Automatic update of common submodule From 20742ae to 169462a 2010-12-14 00:56:53 +0000 Tim-Philipp Müller * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/simple.c: tests: give test suites different names So the output tells us which test is actually running. Also use GST_CHECK_MAIN() while we're at it. 2010-12-13 16:24:51 +0200 Stefan Kost * common: Automatic update of common submodule From 011bcc8 to 20742ae 2010-10-27 13:18:18 +0100 Jan Schmidt * common: Automatic update of common submodule From 7bbd708 to 011bcc8 2010-10-14 12:32:42 -0700 David Schleef * common: Automatic update of common submodule From 5a668bf to 7bbd708 2010-10-12 15:25:10 +0100 Tim-Philipp Müller * gnl/gnlcomposition.c: composition: fix debugging message printf format 2010-10-08 12:45:12 -0700 David Schleef * common: Automatic update of common submodule From c4a8adc to 5a668bf 2010-10-08 12:57:01 +0200 Sebastian Dröge * common: Automatic update of common submodule From 5e3c9bf to c4a8adc 2010-09-21 18:35:06 +0200 Edward Hervey * common: Automatic update of common submodule From 080e025 to 5e3c9bf 2010-09-08 10:14:40 +0100 Tim-Philipp Müller * docs/libs/.gitignore: * docs/libs/Makefile.am: docs: put inspect registry into docs/libs next to the other inspect info Instead of putting the docs registry into docs/plugins. 2010-09-08 00:44:45 +0100 Tim-Philipp Müller * configure.ac: configure: bump core/base requirement to released version Bump core requirement to released version. People should always use the same -base as core, so bump -base requirement to match core requirement. Also bump GLib requirement to GLib requirement of core. 2010-09-08 00:40:41 +0100 Tim-Philipp Müller * configure.ac: configure: another small clean-up AG_GST_CHECK_GST already sets GST_TOOLS_DIR and GST_PLUGINS_DIR, no need to do it a second time. 2010-09-08 00:35:52 +0100 Tim-Philipp Müller * configure.ac: configure: small clean-up AG_GST_CHECK_GST_PLUGINS_BASE already sets GSTPB_PLUGINS_DIR and prints it to the output, no need to do that twice. 2010-09-08 00:33:20 +0100 Tim-Philipp Müller * common: * configure.ac: * tests/check/Makefile.am: tests: find uninstalled -good plugins and use them for tests Enables some more tests / code paths in an uninstalled setup. Also remove duplicate GST_PLUGIN_DIRS. https://bugzilla.gnome.org/show_bug.cgi?id=628943 2010-09-08 00:25:42 +0100 Tim-Philipp Müller * tests/check/gnl/gnlcomposition.c: tests: skip gnlcomposition's test_no_more_pads_race test if videomixer is not available Fixes 'make check' in uninstalled setup where it currently can't find the -good plugins. https://bugzilla.gnome.org/show_bug.cgi?id=628943 2010-09-08 09:55:22 +0100 Tim-Philipp Müller * configure.ac: * docs/libs/gnonlin.hierarchy: * docs/libs/inspect/plugin-gnonlin.xml: Back to development 2010-09-07 11:43:48 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From c2e10bf to aa0d1d0 === release 0.10.16 === 2010-09-06 17:14:09 +0200 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Release 0.10.16 2010-09-05 18:59:10 -0700 David Schleef * common: Automatic update of common submodule From d3d9acf to c2e10bf 2010-09-05 12:21:26 +0200 Sebastian Dröge * common: Automatic update of common submodule From ca1c867 to d3d9acf 2010-09-03 11:05:23 +0200 Edward Hervey * configure.ac: 0.10.15.2 pre-release 2010-09-03 11:04:46 +0200 Edward Hervey * common: Update to latest common 2010-09-02 16:10:43 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use proper macro to detect sources Allows us to use composition as sources 2010-08-17 14:10:38 +0200 Alessandro Decina * gnl/gnlcomposition.c: * tests/check/gnl/gnlcomposition.c: gnlcomposition: fix a race. Fixes #626733. Fix a race between no_more_pads_object_cb and compare_relink_single_node. 2010-08-12 15:53:42 +0200 Alessandro Decina * tests/check/gnl/gnlcomposition.c: tests: add a test case for a race in gnlcomposition. See #626733. Add a test for a race between no_more_objects_cb and compare_relink_single_node in gnlcomposition. 2010-08-19 15:10:25 +0200 Alessandro Decina * gnl/gnlcomposition.c: * tests/check/gnl/gnlcomposition.c: gnlcomposition: fail early in remove_object when removing an invalid object. Make gnl_composition_remove_object return immediately FALSE when removing an object not contained in the composition. 2010-08-19 15:00:38 +0200 Alessandro Decina * gnl/gnlcomposition.c: gnlcomposition: fix thinko. In gnl_composition_remove_object initialize the return value to FALSE and not to GST_STATE_CHANGE_FAILURE. 2010-08-20 10:06:33 +0200 Edward Hervey * tests/check/gnl/gnlcomposition.c: tests: Be more flexible in tests When removing an object from a composition, there might still be some internal refcounting still present since it's not in NULL. Also, use ASYNC_DONE message to detect for pre-roll 2010-08-13 17:25:15 +0300 Stefan Kost * common: Automatic update of common submodule From 3e8db1d to ec60217 2010-08-10 10:59:55 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From bd2054b to 3e8db1d 2010-08-05 13:58:11 +0300 Stefan Kost * common: Automatic update of common submodule From a519571 to bd2054b 2010-08-04 19:32:09 +0200 Sebastian Dröge * configure.ac: configure: Check if the compiler supports ISO C89 or C99 and which parameters are required This first checks what is required for ISO C99 support and sets the relevant compiler parameters and if no C99 compiler is found, it checks for a C89 compiler. This enables us to check for and use C89/C99 functions that gcc hides from us without the correct compiler parameters. 2010-08-03 10:51:34 +0200 Edward Hervey * docs/libs/gnonlin.args: * docs/libs/inspect/plugin-gnonlin.xml: docs: update 2010-07-28 18:13:23 +0200 Sebastian Dröge * gnl/gnlcomposition.c: gnlcomposition: Fix dropping of QOS events for previous segments If the different between the internal and external segment is larger than the timestamp plus diff, the QOS event was for a previous segment and the QOS event would become invalid, i.e. it would be for a negative timestamp. 2010-07-27 17:30:42 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Don't discard QoS event for current segment. curdiff will be zero in the first segment 2010-07-27 16:33:29 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Remove a useless check for GstBin property We depend on a version of GStreamer that will contain it 2010-07-23 15:50:49 +0200 Edward Hervey * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/simple.c: tests: Don't use the expensive videotestsrc patterns We're not viewing them anyway, so might as well take the cheapest to use. 2010-07-23 15:36:04 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use GSlice instead of malloc/free 2010-07-23 15:35:32 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Use GSlice instead of malloc/free 2010-07-23 15:34:51 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Use G_TYPE private instances This ensures the private data is 'closer' to the public structures 2010-07-23 15:33:49 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use G_TYPE private instances This ensures the private data is 'closer' to the public structures 2010-07-23 14:48:32 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use GNL_OBJECT_* macros wherever possible 2010-07-23 14:46:16 +0200 Edward Hervey * gnl/gnlobject.h: gnlobject: New macros for less ugly code 2010-07-20 21:56:28 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Store static pad Avoids having to check twice for available compatible pad 2010-07-12 19:14:24 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use fast pad linking some more 2010-06-28 17:31:54 +0100 Tim-Philipp Müller * autogen.sh: * configure.ac: Bump automake requirement to 1.10 and autoconf to 2.60. 2010-06-27 13:44:55 +0200 Edward Hervey * tests/check/gnl/common.h: check: Use fast pad linking methods 2010-06-27 13:37:31 +0200 Edward Hervey * configure.ac: * gnl/gnlcomposition.c: gnlcomposition: Use fast pad linking Requires latest core 2010-06-24 15:10:17 +0300 Stefan Kost * common: Automatic update of common submodule From 35617c2 to a519571 2010-06-15 16:51:00 +0200 Edward Hervey * common: Automatic update of common submodule From 9339ccc to 35617c2 2010-06-15 16:55:27 +0300 Stefan Kost * common: Automatic update of common submodule From 5adb1ca to 9339ccc 2010-06-15 16:36:35 +0300 Stefan Kost * common: Automatic update of common submodule From 57c89b7 to 5adb1ca 2010-06-15 15:50:53 +0300 Stefan Kost * common: Automatic update of common submodule From c804988 to 57c89b7 2010-06-14 13:28:15 +0200 Sebastian Dröge * configure.ac: configure: Use GLIB_EXTRA_CFLAGS 2010-06-14 13:06:05 +0200 Sebastian Dröge * common: Automatic update of common submodule From 7a0fdf5 to c804988 2010-06-14 11:35:50 +0200 Sebastian Dröge * common: Automatic update of common submodule From 6da3bab to 7a0fdf5 2010-06-12 08:30:13 +0200 Sebastian Dröge * common: Automatic update of common submodule From 733fca9 to 6da3bab 2010-06-10 16:01:07 +0200 Edward Hervey * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnlsource.c: tests: Fix memory leaks 2010-06-10 14:07:01 +0200 Edward Hervey * docs/libs/Makefile.am: docs: Fix source/header file location 2010-06-10 13:57:59 +0200 Edward Hervey * gnl/gnlghostpad.c: GnlGhostPad: Add debug statement 2010-06-10 13:57:43 +0200 Edward Hervey * gnl/gnloperation.c: GnlOperation: Fix more leaks 2010-06-10 13:31:20 +0200 Edward Hervey * tests/check/Makefile.am: * tests/check/common.h: * tests/check/complex.c: * tests/check/gnl/common.h: * tests/check/gnl/complex.c: * tests/check/gnl/gnlcomposition.c: * tests/check/gnl/gnloperation.c: * tests/check/gnl/gnlsource.c: * tests/check/gnl/simple.c: * tests/check/gnlcomposition.c: * tests/check/gnloperation.c: * tests/check/gnlsource.c: * tests/check/simple.c: check: Move tests to a subdirectory 2010-06-09 12:40:35 -0700 David Schleef * common: Automatic update of common submodule From fad145b to 733fca9 2010-06-09 12:34:30 -0700 David Schleef * common: Automatic update of common submodule From 595defc to fad145b 2010-06-09 17:10:36 +0200 Edward Hervey * common: common: Update to current master 2010-06-09 12:58:24 +0200 Edward Hervey * gnl/gnloperation.c: GnlOperation: Properly dispose of our ghostpads. Plugs a leak 2010-06-01 23:49:48 -0700 David Schleef * common: Automatic update of common submodule From 17f89e5 to 47683c1 2010-06-01 22:55:36 -0700 David Schleef * common: Automatic update of common submodule From fd7ca04 to 17f89e5 2010-05-31 12:54:05 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Remove stray debugging statement 2010-05-28 10:22:11 +0100 Tim-Philipp Müller * .gitignore: * Makefile.am: build: copy build files into m4/ instead of common/m4/ We don't want cruft in the common submodule directory. 2010-05-26 11:56:38 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 357b0db to fd7ca04 2010-05-14 18:26:28 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From 4d67bd6 to 357b0db 2010-05-07 15:35:24 +0200 Edward Hervey * tests/check/.gitignore: tests: Ignore more files 2010-05-07 15:34:37 +0200 Edward Hervey * gnl/gnlurisource.c: gnlurisource: Set 'expose-all-streams' on uridecodebin to FALSE We only need the stream specified in the caps 2010-04-23 14:42:29 +0100 Tim-Philipp Müller * common: Automatic update of common submodule From fc85867 to 4d67bd6 2010-04-13 15:35:28 +0200 Edward Hervey * tests/check/common.h: check: Lower framerate coming out of videotestsrc. This should lower the amount of time required to run the tests 2010-04-09 11:24:04 +0200 Sebastian Dröge * common: Automatic update of common submodule From c1d07dd to fc85867 2010-04-01 12:08:11 +0200 Edward Hervey * gnl/gnlsource.c: GnlSource: Demote a WARNING to a DEBUG Not having a source pad is normal, it could be dynamic 2010-03-25 18:47:37 +0100 Edward Hervey * gnl/gnlcomposition.c: GnlComposition: Set the composition caps on the contained elements This is only done if the caps of the composition are no GST_CAPS_ANY. This is a step in allowing easy decode-less support for editing. 2010-03-25 18:44:45 +0100 Edward Hervey * gnl/gnlobject.c: * gnl/gnlobject.h: GnlObject: expose gnl_object_set_caps method 2010-03-24 18:56:21 +0100 Edward Hervey * common: Automatic update of common submodule From 55cd514 to c1d07dd 2010-03-24 11:49:52 +0100 Edward Hervey * configure.ac: * gnl/gnl.c: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: build: Add all kinds of compiler warning flags and fix the resulting warnings 2010-03-18 23:19:46 +0100 Benjamin Otte * gnl/gnlcomposition.c: * gnl/gnlfilesource.c: * gnl/gnloperation.c: * gnl/gnlsource.c: * gnl/gnlurisource.c: gst_element_class_set_details => gst_element_class_set_details_simple https://bugzilla.gnome.org/show_bug.cgi?id=613283 2010-03-19 12:13:07 +0100 Edward Hervey * gnl/gnlsource.c: GnlSource: Handle multiple pad_blocked calls. The pad blocked handler can be called by multiple threads, we therefore need to make sure we only start the ghost_seek_pad handler once. 2010-03-19 10:36:42 +0100 Edward Hervey * gnl/gnlurisource.c: GnlURISource: Set the GnlObject caps on uridecodebin. This allows getting non-raw streams. 2010-03-12 14:00:42 +0100 Edward Hervey * common: Automatic update of common submodule From e272f71 to 55cd514 2010-03-11 11:21:48 +0100 Benjamin Otte * common: Automatic update of common submodule From df8a7c8 to e272f71 2010-03-10 22:05:32 +0100 Benjamin Otte * configure.ac: Update for recent changes to common submodule This just replaces every "$ERROR_CFLAGS" usage with a usage of "$WARNING_CFLAGS $ERROR_CFLAGS" to get the same functionality as previously. Actually using that separation will happen later. 2010-03-10 21:53:06 +0100 Benjamin Otte * common: Automatic update of common submodule From 9720a7d to df8a7c8 2010-03-10 20:44:52 +0100 Benjamin Otte * common: Automatic update of common submodule From 0b6e072 to 9720a7d 2010-03-10 16:10:50 +0100 Benjamin Otte * common: Automatic update of common submodule From 7cc5eb4 to 0b6e072 2010-03-10 10:26:30 +0100 Edward Hervey * configure.ac: configure.ac: And back to development we go 2010-03-10 01:11:35 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 7aa65b5 to 7cc5eb4 2010-03-09 21:42:14 +0000 Sebastian Dröge * common: Automatic update of common submodule From 44ecce7 to 7aa65b5 === release 0.10.15 === 2010-03-09 12:37:36 +0100 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Release 0.10.15 "I missed the snow in Barcelona" 2010-03-07 19:23:46 +0100 Edward Hervey * tests/check/simple.c: tests: Adjust simple test for slight behaviour change introduced in last commit 2010-03-07 19:24:17 +0100 Edward Hervey * tests/check/gnloperation.c: tests: Add a new test for checking fix introduce by last commit. 2010-03-07 19:12:07 +0100 Alessandro Decina * gnl/gnlcomposition.c: gnlcomposition: fix a bug figuring out the segments. If we have two sources (with no default sources) laid out like this: [ source1 ] [ source2 ] We need to configure 3 different segments and therefore build 3 different stacks. One in which only source1 is active, one in which both source1 and source2 are active and finally one in which only source2 is active. https://bugzilla.gnome.org/show_bug.cgi?id=609792 2010-02-28 20:03:17 +0100 Alessandro Decina * gnl/gnlcomposition.c: composition: call update_pipeline when an expandable source is added. 2010-02-28 20:02:32 +0100 Alessandro Decina * gnl/gnlfilesource.c: Fix a compiler warning in OSX. 2010-02-26 16:41:54 +0100 Sebastian Dröge * gnl/Makefile.am: build: Make some more rules silent if requested 2010-02-26 15:50:51 +0100 Sebastian Dröge * configure.ac: configure: Use automake 1.11 silent rules instead of shave if available This makes sure that we use something that is still maintained and also brings back libtool 1.5 support. 2010-02-17 10:01:03 +0100 Sebastian Dröge * configure.ac: Don't check for check, gst-check already includes it 2010-02-17 09:48:08 +0100 Sebastian Dröge * Makefile.am: Include lcov.mak to allow unit test coverage report generation 2010-02-14 23:19:25 +0100 Sebastian Dröge * common: Automatic update of common submodule From 96dc793 to 44ecce7 2010-02-12 18:19:13 +0100 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Don't forget to carry over the modifyed stack This can happen when we have several operations stacked one over each other. Fixes #609689 2010-02-12 11:37:24 +0100 Edward Hervey * configure.ac: configure.ac: And back to development we go === release 0.10.14 === 2010-02-11 13:55:52 +0100 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Releasing 0.10.14, "Slicing, Dicing and Chopping" 2010-01-30 15:20:36 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 15d47a6 to 96dc793 2010-01-25 20:35:00 +0100 Edward Hervey * configure.ac: configure.ac: 0.10.13.3 pre-release 2010-01-20 00:55:52 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 14cec89 to 15d47a6 2010-01-12 15:34:50 +0100 Edward Hervey * gnl/gnlobject.c: gnlobject: Specify unit in property descriptions It was documented in gtk-doc but not in the introspectable documentation 2009-12-21 19:13:41 +0100 Mark Nauwelaerts * common: Automatic update of common submodule From 47cb23a to 14cec89 2009-12-01 15:08:54 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 87bf428 to 47cb23a 2009-12-01 12:32:43 +0100 Edward Hervey * configure.ac: 0.10.13.2 pre-release 2009-12-01 14:18:47 +0100 Sebastian Dröge * common: Automatic update of common submodule From da4c75c to 87bf428 2009-11-27 18:56:58 +0100 Edward Hervey * common: Automatic update of common submodule From 53a2485 to da4c75c 2009-11-24 15:20:12 +0100 Edward Hervey * m4/Makefile.am: * m4/codeset.m4: * m4/gettext.m4: * m4/glibc21.m4: * m4/iconv.m4: * m4/intdiv0.m4: * m4/inttypes-pri.m4: * m4/inttypes.m4: * m4/inttypes_h.m4: * m4/isc-posix.m4: * m4/lcmessage.m4: * m4/lib-ld.m4: * m4/lib-link.m4: * m4/lib-prefix.m4: * m4/progtest.m4: * m4/stdint_h.m4: * m4/uintmax_t.m4: * m4/ulonglong.m4: m4: remove unneeded disted m4 They will be installed by autogen.sh 2009-11-19 10:32:07 +0000 Tim-Philipp Müller * common: Automatic update of common submodule From 0702fe1 to 53a2485 2009-11-17 17:38:32 +0000 Jan Schmidt * gnl/Makefile.am: build: Remove duplicate noinst_HEADERS in gnl/Makefile.am 2009-10-16 10:17:56 +0300 Stefan Kost * common: Automatic update of common submodule From 85d1530 to 0702fe1 2009-10-14 11:08:00 +0200 Edward Hervey * gnl/gnlurisource.c: gnlurisource: Fix description 2009-10-14 10:42:21 +0200 Edward Hervey * common: Automatic update of common submodule From a3e3ce4 to 85d1530 2009-10-08 11:01:49 +0100 Jan Schmidt * common: Automatic update of common submodule From 19fa4f3 to a3e3ce4 2009-09-21 18:34:05 +0200 Edward Hervey * gnl/gnlurisource.c: * gnl/gnlurisource.h: gnlurisource: ... it works better with the files. 2009-09-21 11:50:53 +0200 Edward Hervey * docs/libs/gnonlin.args: * docs/libs/gnonlin.signals: docs: Update args/signals 2009-09-21 11:50:44 +0200 Edward Hervey * docs/libs/gnonlin-docs.sgml: * docs/libs/gnonlin-sections.txt: * docs/libs/gnonlin.args: * docs/libs/gnonlin.hierarchy: * docs/libs/gnonlin.interfaces: * docs/libs/inspect/plugin-gnonlin.xml: docs: Add gnlurisource 2009-09-21 11:49:07 +0200 Edward Hervey * configure.ac: * gnl/Makefile.am: * gnl/gnl.c: * gnl/gnl.h: * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: * gnl/gnltypes.h: New GnlURISource plugin, of which GnlFileSource subclasses. Fixes #595570 2009-09-21 11:51:27 +0200 Edward Hervey * configure.ac: configure.ac: And back to development we go... 2009-09-10 08:18:48 +0200 Sebastian Dröge * gnl/gnlfilesource.c: gnlfilesource: If we use uridecodebin get the location from there This prevents assertions that result from getting the property value from the internal filesrc, which is NULL if we use uridecodebin. 2009-09-07 11:51:36 +0200 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: 0.10.13 "Service of Quality" 2009-09-05 10:26:08 +0200 Sebastian Dröge * common: Automatic update of common submodule From 94f95e3 to 19fa4f3 2009-09-02 18:04:53 +0200 Edward Hervey * configure.ac: 0.10.12.3 pre-release 2009-09-01 18:54:19 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: gnlcomposition: Fix QoS handling. QoS handling requires knowledge of the running time which only GnlComposition is aware of. Therefore we remove it from GnlGhostPad and handle it in the composition. Details of the algorithm are contained as comments in gnlcomposition.c 2009-08-29 11:17:57 +0200 Edward Hervey * configure.ac: configure.ac: 0.10.12.2 pre-release 2009-08-24 16:50:24 +0200 Alessandro Decina * gnl/gnlcomposition.c: * gnl/gnlghostpad.c: gnlghostpad: clamp jitter to the transformed timestamp. Fixes "GStreamer-CRITICAL **: gst_event_new_qos: assertion `diff >= 0 || -diff <= timestamp' failed" assertions. 2009-08-24 12:08:45 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Make the pad removal code non-racy. Previously we were checking whether (1) there was a ghostpad and (2) whether the target was the pad being removed. The problem was that the following racyness can happen: 1. A pad gets added and controlled 2. The pad gets removed before having outputted any data (and therefore the ghostpad isn't created). 3. There's no ghostpad... and therefore we don't properly reset ourselves. 2009-08-23 19:50:31 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnloperation.h: gnlcomposition: link operations by pad, inform them of incoming stream priorities. 2009-08-23 19:50:01 +0200 Edward Hervey * gnl/gnloperation.c: GnlOperation: Fix signal signature, add public method to emit signal 2009-08-12 18:31:36 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Drop incoming events when not entirely linked. Fixes #583145 2009-08-12 18:32:19 +0200 Edward Hervey * configure.ac: configure.ac: And back to development we go... === release 0.10.12 === 2009-08-10 18:02:57 +0200 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: releasing 0.10.12, "Lots of people on the clothesline" 2009-08-10 16:28:38 +0200 Edward Hervey * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: gnlfilesource: Use uridecodebin and remove dead code. Using decodebin2 directly fails because it blocks its own pads (therefore making it impossible for us to block them ourselves. 2009-08-06 17:35:21 +0200 Alessandro Decina * gnl/gnlghostpad.c: When ghosting a pad that has negotiated caps, set those caps to the ghost as well. 2009-08-04 17:47:34 +0200 Edward Hervey * configure.ac: configure.ac: 0.10.11.3 pre-release 2009-07-29 20:40:25 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Only remove top-level pad if removed pad is a source pad. "Wow, not all objects in the timeline are sources ?" 2009-07-29 15:58:25 +0200 Edward Hervey * gnl/gnlsource.c: gnlsource: Properly reset internal state when controlled pad has gone. We weren't resetting pendingblock, resulting in some cases where a new pad to control would appear, but would not be used because pendingblock was still TRUE. 2009-07-01 18:53:43 +0200 Edward Hervey * gnl/gnloperation.c: * gnl/gnloperation.h: gnloperation: TEMPORARY COMMIT ! REBASE ME ! Added the signal, the next step is to emit it from the composition which requires replacing the gst_element_link with gst_pad_link... which we should have been using ages ago :) 2009-07-01 16:56:27 +0200 Edward Hervey * gnl/Makefile.am: * gnl/gnlmarshal.list: * gnl/gnloperation.c: gnl: Adding signal marshaller 2009-07-27 20:07:21 +0200 Adam Dingle * gnl/gnlfilesource.c: gnlfilesource: Really use decodebin2. Fixes #589525 2009-07-24 00:43:47 +0300 Stefan Kost * common: Automatic update of common submodule From fedaaee to 94f95e3 2009-07-16 10:20:08 +0200 Edward Hervey * configure.ac: 0.10.11.2 pre-release 2009-07-13 12:24:46 -0400 Olivier Crête * common: Automatic update of common submodule From 5845b63 to fedaaee 2009-06-26 13:44:04 +0100 Jan Schmidt * common: Automatic update of common submodule From f810030 to 5845b63 2009-06-24 19:15:30 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Remove bogus debug statement. 2009-06-24 15:16:42 +0100 Jan Schmidt * common: Automatic update of common submodule From f3bb51b to f810030 2009-06-14 11:35:52 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlcomposition.h: * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: gnlcomposition/gnlfilesource: Don't use 'private' keyword. This avoids build errors with C++ compilers. 2009-06-12 20:00:00 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Don't forget to lower the pad count when removing one. 2009-06-12 18:48:59 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: rewrite from scratch of pipeline unlinking/linking. We currently weren't handling operations in general (and dynamic multi-sinkpad operations in particular) in a safe fashion. This rewrite makes sure that when redoing the pipeline: * no streaming threads are flowing between elements when deactivating * all source pads are blocked (asynchronously) * flush start/stop is sent out if this is the consequence of a seek 2009-06-12 18:46:34 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: When receiving a seek, let update_pipeline send the flush events. 2009-06-12 18:44:46 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: *always* update pipeline when removing objects. The object might currently have a streaming thread flowing out/through of it. We therefore need to properly stop that by forcing a refresh of the internal pipeline before calling the parent class remove_object() 2009-06-12 18:43:51 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Recheck for ghostpad existence in critical codepath. 2009-06-12 18:42:25 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: add G_UNLIKELY and remove useless warning 2009-06-12 18:41:21 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Unset sink ghostpad targets and fix pad-synch for dynamic operations. 2009-06-10 19:42:11 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlobject.h: gnlcomposition: Simplify IN_ACTIVE_SEGMENT macro It also fixes the issue when the object would start before the segment start and end after the segment stop. 2009-06-11 13:51:06 +0100 Jan Schmidt * common: * docs/Makefile.am: * docs/libs/Makefile.am: * docs/upload.mak: docs: Bump common. Remove extra upload.mak. Fix comment. 2009-05-30 17:57:53 +0200 Edward Hervey * gnl/gnlcomposition.c: * tests/check/complex.c: * tests/check/gnloperation.c: GnlComposition: Use expandable GnlObject property. 2009-05-29 19:47:43 +0200 Edward Hervey * gnl/gnlobject.c: * gnl/gnlobject.h: GnlObject: Add an 'expandable' property/flag. This indicates that when the given object is used in a GnlComposition, it should expand to the full extent of the composition's start/duration. 2009-05-29 19:35:52 +0200 Edward Hervey * tests/check/complex.c: checks: Don't check for gstreamer 0.10.4 anymore. We already depend on a much recent version. 2009-05-30 17:59:39 +0200 Edward Hervey * common: * configure.ac: configure.ac: Back to development. 2009-05-26 17:20:52 +0100 Jan Schmidt * common: Automatic update of common submodule From 888e0a2 to c572721 === release 0.10.11 === 2009-05-24 11:08:00 +0200 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: 0.10.11 release "How about green for the bikeshed?" 2009-05-22 10:21:31 +0100 Jan Schmidt * common: Automatic update of common submodule From 6ab11d1 to 888e0a2 2009-05-18 13:07:13 +0200 Edward Hervey * configure.ac: 0.10.10.3 pre-release 2009-05-12 19:25:34 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Don't return FALSE in critical code-path. Since updates are disabled, just return TRUE to make calling code believe it succeeded. 2009-05-10 11:17:16 +0200 Marc-Andre Lureau * autogen.sh: Run libtoolize before aclocal This unbreaks the build in some cases. Fixes bug #582021 2009-05-09 12:58:11 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Also look at element class for pad templates. This happens when an element doesn't come from a factory, like python elements. 2009-05-08 13:50:47 +0200 Edward Hervey * docs/libs/gnonlin.args: * docs/libs/inspect/plugin-gnonlin.xml: gtk-doc: update docs. 2009-05-08 13:49:55 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Document the 'update' property. 2009-04-19 17:10:47 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: Document properties 2009-05-08 12:40:34 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Fix 'update' property description. 2009-05-08 12:39:27 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Use faster GList routines. Use g_list_prepend() when we don't care about the order of the list. Use g_list_insert_sorted() instead of g_list_append() + g_list_sort(). Don't sort a list after removing an item from an already sorted list. 2009-05-08 12:36:35 +0200 Edward Hervey * gnl/gnlcomposition.c: * tests/check/simple.c: gnlcomposition: Update start/duration/stop even with update==False. The reason behind this is that updating the lists and those properties are decently simple/fast to do it at that time. The internal pipeline will still not be updated if update == False. 2009-05-06 16:21:01 +0200 Edward Hervey * gnl/gnlcomposition.c: * tests/check/Makefile.am: * tests/check/common.h: * tests/check/simple.c: gnlcomposition: Implement update-blocking. Fixes #555898 When update is set to False, the internal pipeline is never updated, nor are the stop/duration properties of the composition. This allows for massive modifications with a much lower overhead. 2009-05-06 16:13:16 +0200 Edward Hervey * tests/check/simple.c: tests/simple: Remove check for core > 0.10.4 We depend on a higher version already. 2009-04-30 20:13:21 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Remove duplicated code. We use the same code when an object's start/stop/priority changes. 2009-04-21 22:14:26 +0100 Jan Schmidt * common: Automatic update of common submodule From b3941ea to 6ab11d1 2009-04-19 13:29:41 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad. Implement outgoing_{qos|position} handling. Fixes #578234 2009-04-19 12:37:11 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Add some debugging statements 2009-04-19 12:31:21 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Use our magic ghostpad for sink pads 2009-04-19 12:30:48 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Implement outgoing_seek and incoming_new_segment handling 2009-04-19 12:28:33 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: consider media_start as 0 if it's undefined. This doesn't introduce regressions, but makes our life easier for handling time-shifting with operations. 2009-04-19 10:12:52 +0200 Edward Hervey * gnl/gnlghostpad.c: * gnl/gnlobject.c: * gnl/gnlobject.h: Optimize calculations by using a boolean for (rate == 1.0) cases. 2009-04-19 10:04:01 +0200 Edward Hervey * gnl/gnl.c: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: gnlghostpad: Add a 'gnlghostpad' debugging category 2009-04-19 09:44:16 +0200 Edward Hervey * gnl/gnlghostpad.c: gnlghostpad: Wrap G_*LIKELY with parenthses. Makes indent act funny otherwise 2009-04-19 09:41:59 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: Wrap G_*LIKELY with parenthses. Makes indent act funny otherwise 2009-04-19 09:35:51 +0200 Edward Hervey * gnl/Makefile.am: * gnl/gnl.h: * gnl/gnlghostpad.c: * gnl/gnlghostpad.h: * gnl/gnlobject.c: * gnl/gnlobject.h: gnlghostpad: Move all ghostpad-related methods to a separate file. Helps for maintainability. 2009-04-19 09:34:44 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: Give hints as to likelyhood of branches. 2009-04-19 09:04:15 +0200 Edward Hervey * gnl/gnlobject.c: * gnl/gnlobject.h: gnlobject: Remove 'coverage' items. They were never in the public API. This was leftover from pre-0.10 gnonlin. 2009-04-19 08:59:03 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: No need to emit 'segment-start' message on newsegment anymore. We had that to help GnlComposition. 2009-04-19 08:40:32 +0200 Edward Hervey * gnl/gnlobject.c: gnlobject: Remove the message handling now that we don't use segment messages. 2009-04-17 13:15:24 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Add a stackvalid protected variable to know if we can expose the stack. The problem is that checking if comp->private->current is available isn't good enough, since we might be in the middle of activating the stack. 2009-04-16 15:18:14 +0200 Edward Hervey * gnl/gnlfilesource.c: gnlfilesource: Make decodebin2 the default. 2009-04-04 21:19:56 +0300 Felipe Contreras * common: Automatic update of common submodule From d0ea89e to b3941ea 2009-04-04 14:55:21 +0200 Edward Hervey * common: Automatic update of common submodule From f8b3d91 to d0ea89e 2009-04-03 11:06:51 +0200 Edward Hervey * gnl/gnloperation.c: gnloperation: Check if the return-value-as-argument is valid. 2009-04-03 11:06:36 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlfilesource.c: Remove unneeded variables 2009-04-03 11:05:41 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Take into account the result of the position query. 2009-04-03 11:04:42 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Fix usage of GstIterator. Don't allocate unused variables (GValue, return value) Resync if needed. 2009-04-01 12:44:44 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Fix inline declaration and remove useless variable. 2009-04-01 12:44:25 +0200 Edward Hervey * gnl/gnlcomposition.c: gnlcomposition: Remove dead/useless code. 2009-04-01 12:43:50 +0200 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlfilesource.c: * gnl/gnlobject.c: * gnl/gnloperation.c: * gnl/gnlsource.c: Mark unused arguments using G_GNUC_UNUSED glib macro. 2009-04-01 12:19:21 +0200 Edward Hervey * configure.ac: * gnl/Makefile.am: * gnl/gnlversion.h.in: Remove old old old gnlversion.h.in 2009-03-25 13:09:43 +0100 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlobject.c: Translate incoming QoS events. Fixes #398453. 2009-03-24 17:35:21 +0100 Alessandro Decina * gnl/gnlcomposition.c: gnlcomposition: remove a GST_ERROR_OBJECT i used for debugging. meh. 2009-03-24 17:31:21 +0100 Alessandro Decina * gnl/gnlcomposition.c: gnlcomposition: clear private->ghosteventprobe when the pad goes away. 2009-03-21 16:48:53 +0100 Edward Hervey * tests/check/common.h: * tests/check/complex.c: * tests/check/gnloperation.c: * tests/check/gnlsource.c: * tests/check/simple.c: Fix tests when running on systems with alpha. Fixes #563507 2009-03-21 15:41:15 +0100 Edward Hervey * gnl/gnlcomposition.c: * gnl/gnlobject.c: GnlComposition: Switch to using regular segment seeks. Fixes #575972 2009-03-11 11:57:21 +0100 Edward Hervey * configure.ac: 0.10.10.2 pre-release 2009-03-09 23:14:29 +0000 Jan Schmidt * common: Automatic update of common submodule From 7032163 to f8b3d91 2009-03-08 12:06:54 +0100 Sebastian Dröge * common: Automatic update of common submodule From ffa738d to 7032163 2009-03-08 11:22:30 +0100 Sebastian Dröge * common: Automatic update of common submodule From 3f13e4e to ffa738d 2009-03-07 11:48:02 +0100 Sebastian Dröge * common: Automatic update of common submodule From 3c7456b to 3f13e4e 2009-03-07 10:48:09 +0100 Sebastian Dröge * common: Automatic update of common submodule From 57c83f2 to 3c7456b 2009-03-05 16:52:13 +0100 Alessandro Decina * gnl/gnlcomposition.c: * tests/check/Makefile.am: * tests/check/gnlcomposition.c: GnlComposition: rebuild the stack when an object in the current stack is moved outside the active segment. 2009-03-02 17:46:43 +0100 Edward Hervey * gnl/gnlcomposition.c: GnlComposition: Post a GST_MESSAGE_DURATION when the duration changes. Fixes #569854 2009-02-25 15:09:41 +0000 Jan Schmidt * common: * configure.ac: build: Update shave init statement for changes in common. Bump common. 2009-02-25 11:33:24 +0000 Jan Schmidt * common: Automatic update of common submodule From 9cf8c9b to a6ce5c6 2009-02-24 13:08:26 +0100 Alessandro Decina * gnl/gnlcomposition.c: gnlcomposition: in update_pipeline() set the seek start position if it's different from the currently configured position. 2009-02-23 18:30:16 +0100 Edward Hervey * gnl/gnlsource.c: gnlsource: Unref only pads that (1) exist and (2) are ours. 2009-02-23 13:27:48 +0200 Stefan Kost * docs/libs/gnonlin-docs.sgml: docs: use right entity name for version. 2009-02-22 20:33:17 +0000 Jan Schmidt * configure.ac: * docs/libs/Makefile.am: Use shave for the build 2009-02-22 16:02:16 +0000 Jan Schmidt * common: Automatic update of common submodule From 5d7c9cc to 9cf8c9b 2009-02-21 11:14:35 -0800 David Schleef * common: Automatic update of common submodule From 80c627d to 5d7c9cc 2009-02-13 15:35:51 -0800 David Schleef * .gitignore: Remove common from .gitignore Makes it easier to update common. 2009-02-09 12:04:30 +0100 Edward Hervey * common: Bump revision to use for common submodule. 2009-02-04 16:49:26 +0000 Jan Schmidt * gnl/gnlfilesource.c: gnlfilesource: Avoid crashing on missing elements Even though it should never happen, don't crash if we fail to create a filesrc or decodebin for whatever reason. 2009-02-04 16:46:50 +0000 Jan Schmidt * gnl/gnlsource.c: gnlsource: Move a != NULL check to before the first use. 2009-01-30 22:30:50 +0200 Stefan Kost * docs/libs/gnonlin-docs.sgml: Add releaseinfo with online url. 2009-01-30 18:51:36 +0100 Edward Hervey * gnl/gnlobject.c: Implement handling of GST_QUERY_DURATION for gnlobject and subclasses Partially fixes #569854 2009-01-30 18:44:11 +0100 Edward Hervey * .gitignore: Ignore more files 2009-01-30 17:46:24 +0000 Jan Schmidt * common: Bump common 2009-01-30 09:07:08 +0100 Edward Hervey * autogen.sh: Fix previous commit, wasn't actually setting up a symbolic link 2009-01-30 09:00:21 +0100 Edward Hervey * autogen.sh: * common: Use a symbolic link for the pre-commit client-side hook 2009-01-26 12:00:49 +0100 Alessandro Decina * gnl/gnlcomposition.c: gnl/gnlcomposition.c: Check that GST_MESSAGE_SRC (m) is a GnlObject before casting it. 2009-01-22 13:43:57 +0100 Sebastian Dröge * common: Fix pre-commit hook 2009-01-22 06:18:54 +0100 Edward Hervey * autogen.sh: * common: Install and use pre-commit indentation hook from common 2009-01-21 04:36:21 +0100 Edward Hervey * autogen.sh: autogen.sh : Use git submodule 2009-01-07 12:12:36 +0000 Alessandro Decina gnl/gnlcomposition.c: Release objects lock temporarily when emitting no-more-pads. Original commit message from CVS: Patch by: Alessandro Decina * gnl/gnlcomposition.c: (gnl_composition_ghost_pad_set_target): Release objects lock temporarily when emitting no-more-pads. This should not affect anything considering that the composition is in the process of being built at that point and no actions should be attempting to modify it while that lock is released. Fixes #566796 2008-12-28 15:28:40 +0000 Edward Hervey gnl/gnlcomposition.c: Refine means "do not EXTEND the region" ! Original commit message from CVS: * gnl/gnlcomposition.c: (refine_start_stop_in_region_above_priority): Refine means "do not EXTEND the region" ! This should fix more issues with complex timelines. 2008-12-14 12:24:30 +0000 Edward Hervey gnl/gnlsource.c: Don't forget to call the parent_class send_event handler. Original commit message from CVS: * gnl/gnlsource.c: (gnl_source_send_event): Don't forget to call the parent_class send_event handler. 2008-12-13 16:46:47 +0000 Edward Hervey m4/Makefile.am: Two more .m4 that aren't shipped anymore with gettext >= 0.17. Original commit message from CVS: * m4/Makefile.am: Two more .m4 that aren't shipped anymore with gettext >= 0.17. 2008-12-13 13:03:18 +0000 Edward Hervey m4/Makefile.am: inttypes.m4 hasn't been available since gettext-0.15, and since we now require gettext >= 0.17 ... we... Original commit message from CVS: * m4/Makefile.am: inttypes.m4 hasn't been available since gettext-0.15, and since we now require gettext >= 0.17 ... we can remove it from the list of files to dist. 2008-12-04 20:13:46 +0000 Sebastian Dröge configure.ac: Apparently AC_CONFIG_MACRO_DIR breaks when using more than one macro directory, reverting last change. Original commit message from CVS: * configure.ac: Apparently AC_CONFIG_MACRO_DIR breaks when using more than one macro directory, reverting last change. 2008-12-04 19:51:19 +0000 Sebastian Dröge configure.ac: Set AC_CONFIG_MACRO_DIR to common/m4 to point autoconf to our M4 macros. Original commit message from CVS: * configure.ac: Set AC_CONFIG_MACRO_DIR to common/m4 to point autoconf to our M4 macros. 2008-11-29 13:34:48 +0000 Sebastian Dröge Require gettext 0.17 because older versions don't mix with libtool 2.2. At build time an older gettext version will s... Original commit message from CVS: Patch by: Cygwin Ports maintainer * autogen.sh: * configure.ac: Require gettext 0.17 because older versions don't mix with libtool 2.2. At build time an older gettext version will still work. Fixes bug #556091. 2008-11-05 12:57:08 +0000 Edward Hervey * configure.ac: If you read this commit message, you pay too much attention to details Original commit message from CVS: If you read this commit message, you pay too much attention to details 2008-11-05 12:54:56 +0000 Edward Hervey gnl/Makefile.am: Don't install static libs for plugins. Fixes #550851 for -gnonlin. Original commit message from CVS: * gnl/Makefile.am: Don't install static libs for plugins. Fixes #550851 for -gnonlin. 2008-11-04 12:47:10 +0000 Stefan Kost gnl/Makefile.am: Don't install static libs for plugins. Fixes #550851 for -gnonlin. Original commit message from CVS: * gnl/Makefile.am: Don't install static libs for plugins. Fixes #550851 for -gnonlin. === release 0.10.10 === 2008-11-03 15:50:20 +0000 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Releasing 0.10.10 Original commit message from CVS: Releasing 0.10.10 2008-11-03 15:25:16 +0000 Edward Hervey gnl/gnlobject.c: Small protection for when we get invalid ghostpads. Doesn't change behaviour, but reduces fatal warn... Original commit message from CVS: * gnl/gnlobject.c: (control_internal_pad): Small protection for when we get invalid ghostpads. Doesn't change behaviour, but reduces fatal warnings. 2008-10-26 23:09:33 +0000 Edward Hervey configure.ac: 0.10.9.2 pre-release. Original commit message from CVS: * configure.ac: 0.10.9.2 pre-release. 2008-09-25 15:52:39 +0000 Edward Hervey tests/check/gnloperation.c: Adapt test to 'fixed' behaviour of videomixer fixed by #537361 Original commit message from CVS: * tests/check/gnloperation.c: (GST_START_TEST): Adapt test to 'fixed' behaviour of videomixer fixed by #537361 2008-07-29 17:19:42 +0000 Edward Hervey gnl/gnlcomposition.c: When calculating current position, request it downstream since it has a much better chance of k... Original commit message from CVS: * gnl/gnlcomposition.c: (get_current_position), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (gnl_composition_add_object), (gnl_composition_remove_object): When calculating current position, request it downstream since it has a much better chance of knowing the exact position (because of queues, ..). If that fails, then revert to upstream (in sources) as before. Also, use the current position when stop/start/active/priority child properties change. 2008-07-28 16:40:38 +0000 Edward Hervey gnl/gnlcomposition.c: Fix race-issue when unblocking top-level pad because the update_pipeline method isn't atomic (t... Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_handle_message), (no_more_pads_object_cb), (update_pipeline): Fix race-issue when unblocking top-level pad because the update_pipeline method isn't atomic (the objects lock is released in the middle). Also ignore ERROR/WARNING coming from elements which aren't in the currently configured stack. This is mostly related to a big issue with reverse negotiation in dynamic pipelines. 2008-05-31 15:23:09 +0000 Edward Hervey docs/libs/inspect/plugin-gnonlin.xml: Plugin info as changed (email addresses). Original commit message from CVS: * docs/libs/inspect/plugin-gnonlin.xml: Plugin info as changed (email addresses). 2008-05-31 15:13:51 +0000 Edward Hervey gnl/gnlcomposition.c: Don't use the objects_lock to protect the message handling, instead use a new lock (messages_lo... Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_finalize), (gnl_composition_handle_message), (update_pipeline): Don't use the objects_lock to protect the message handling, instead use a new lock (messages_lock). Avoids a stupid deadlock. 2008-05-30 16:09:44 +0000 Edward Hervey gnl/: Update email adresses of wim and edward to working adresses. Original commit message from CVS: * gnl/gnl.c: * gnl/gnl.h: * gnl/gnlcomposition.c: * gnl/gnlcomposition.h: * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: * gnl/gnlobject.c: * gnl/gnlobject.h: * gnl/gnloperation.c: * gnl/gnlsource.c: * gnl/gnlsource.h: Update email adresses of wim and edward to working adresses. Use less GNL_* casting macros that actually also do typechecking for all parts where it is not needed, and instead use simple casting. 2008-05-27 17:41:45 +0000 Edward Hervey gnl/gnlcomposition.c: Properly accumulate segment_start/segment_done messages so that we only switch segment when all... Original commit message from CVS: * gnl/gnlcomposition.c: (add_message), (flush_messages), (replace_message), (has_message), (dump_messages), (gnl_composition_handle_message), (no_more_pads_object_cb), (update_pipeline): Properly accumulate segment_start/segment_done messages so that we only switch segment when all segment_done have been emitted. Inspired by the logic in gstbin. * gnl/gnloperation.c: (get_unlinked_sink_ghost_pad), (remove_sink_pad), (synchronize_sinks), (gnl_operation_request_new_pad), (gnl_operation_release_pad): Methods to properly remove request pads. * tests/check/common.h: Remove a leak 2008-05-26 17:26:16 +0000 Edward Hervey gnl/: Fix more issues with dynamic-sinkpads operations. Original commit message from CVS: * gnl/gnlcomposition.c: (unblock_child_pads), (get_new_seek_event), (get_current_position), (refine_start_stop_in_region_above_priority), (convert_list_to_tree), (get_stack_list), (get_clean_toplevel_stack), (gnl_composition_change_state), (compare_relink_single_node), (are_same_stacks), (update_pipeline), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (object_pad_removed), (object_pad_added), (gnl_composition_remove_object): * gnl/gnloperation.c: (element_is_valid_filter), (get_nb_static_sinks), (gnl_operation_add_element), (get_unused_static_sink_pad), (get_request_sink_pad), (add_sink_pad), (synchronize_sinks): * gnl/gnloperation.h: Fix more issues with dynamic-sinkpads operations. Gst-indent the code. 2008-05-26 17:21:29 +0000 Edward Hervey gnl/: Round of gst-indent Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_init): * gnl/gnlobject.c: (translate_incoming_seek), (ghostpad_event_function): * gnl/gnlsource.c: (gnl_source_init): Round of gst-indent 2008-05-26 11:31:27 +0000 Edward Hervey tests/check/: Round of gst-indent and more utilities Original commit message from CVS: * tests/check/common.h: * tests/check/complex.c: (GST_START_TEST): * tests/check/gnlsource.c: (GST_START_TEST): * tests/check/simple.c: (GST_START_TEST): Round of gst-indent and more utilities * tests/check/gnloperation.c: (GST_START_TEST), (gnonlin_suite): New test for multiple-input-pad operations. Needs more work 2008-05-26 11:30:03 +0000 Edward Hervey gnl/gnlcomposition.c: Exit properly if stack is empty. Original commit message from CVS: * gnl/gnlcomposition.c: (no_more_pads_object_cb): Exit properly if stack is empty. 2008-05-18 21:47:16 +0000 Jan Schmidt docs/libs/.cvsignore: Ignore a couple more generated files. Original commit message from CVS: * docs/libs/.cvsignore: Ignore a couple more generated files. 2008-05-16 15:20:31 +0000 Jan Schmidt Add some docs. The structure is based on the GStreamer plugin docs build, with an enhancement to allow having GnlObje... Original commit message from CVS: * configure.ac: * docs/Makefile.am: * docs/libs/.cvsignore: * docs/libs/Makefile.am: * docs/libs/gnonlin-docs.sgml: * docs/libs/gnonlin-sections.txt: * docs/libs/gnonlin.args: * docs/libs/gnonlin.hierarchy: * docs/libs/gnonlin.interfaces: * docs/libs/gnonlin.prerequisites: * docs/libs/gnonlin.signals: * docs/libs/gnonlin.types: * docs/libs/inspect-build.stamp: * docs/libs/inspect.stamp: * docs/libs/inspect/plugin-gnonlin.xml: * docs/libs/scanobj-build.stamp: * docs/libs/tmpl/gnl.sgml: * docs/libs/tmpl/gnlcomposition.sgml: * docs/libs/tmpl/gnlobject.sgml: * docs/libs/tmpl/gnloperation.sgml: * docs/libs/tmpl/gnlsource.sgml: * docs/libs/tmpl/gnltimeline.sgml: * gnl/gnlcomposition.c: * gnl/gnlcomposition.h: * gnl/gnlfilesource.c: * gnl/gnlfilesource.h: * gnl/gnlobject.c: * gnl/gnloperation.c: * gnl/gnloperation.h: * gnl/gnlsource.c: * gnl/gnltypes.h: Add some docs. The structure is based on the GStreamer plugin docs build, with an enhancement to allow having GnlObject (a plugin-private base class) documented too. Actual docs contents needs fleshing out. 2008-04-21 11:24:03 +0000 Edward Hervey gnl/gnloperation.c: Fix small bug in debugging. Original commit message from CVS: * gnl/gnloperation.c: (gnl_operation_add_element): Fix small bug in debugging. 2008-04-06 08:59:44 +0000 Damien Lespiau configure.ac: Actually build dlls when cross-compiling with mingw32. Original commit message from CVS: Patch by: Damien Lespiau * configure.ac: Actually build dlls when cross-compiling with mingw32. Fixes bug #526247. 2008-01-10 12:45:31 +0000 Sebastian Dröge autogen.sh: Add -Wno-portability to the automake parameters to stop warnings about GNU make extensions being used. We... Original commit message from CVS: * autogen.sh: Add -Wno-portability to the automake parameters to stop warnings about GNU make extensions being used. We require GNU make in almost every Makefile anyway. * configure.ac: Use AM_PROG_CC_C_O as a compiler that accepts both -c and -o at the same time is required for per target flags. 2007-12-09 04:30:35 +0000 Sebastian Dröge configure.ac: Don't define GST_DISABLE_DEPRECATED for releases. Fixes #498181. Original commit message from CVS: * configure.ac: Don't define GST_DISABLE_DEPRECATED for releases. Fixes #498181. 2007-08-08 13:06:05 +0000 Edward Hervey configure.ac: Back to development cycle. Original commit message from CVS: * configure.ac: Back to development cycle. === release 0.10.9 === 2007-08-08 13:04:39 +0000 Edward Hervey * ChangeLog: * NEWS: * RELEASE: * configure.ac: Releasing 0.10.9 Original commit message from CVS: Releasing 0.10.9 2007-08-01 08:09:45 +0000 Edward Hervey configure.ac: 0.10.8.2 pre-release Original commit message from CVS: * configure.ac: 0.10.8.2 pre-release 2007-07-30 15:08:57 +0000 Edward Hervey tests/check/common.h: Add a bigger function for creating a video source with specific media-start and media-duration. Original commit message from CVS: * tests/check/common.h: Add a bigger function for creating a video source with specific media-start and media-duration. * tests/check/simple.c: (GST_START_TEST): Let's check with non-obvious media-start/duration sources. 2007-07-23 16:45:03 +0000 Edward Hervey gnl/gnlcomposition.c: Added debugging comments on a race issue to fix. Original commit message from CVS: * gnl/gnlcomposition.c: (get_new_seek_event): Added debugging (seek_handling): comments on a race issue to fix. (gnl_composition_event_handler): protect call to get_new_seek_event. (update_pipeline): Update seek events need to be flushing if not in PLAYING, else we will encounter deadlocks. (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed): If composition has changed, we set segment->start to segment_start in the same way we do when a segment has finished. 2007-07-02 14:44:04 +0000 Edward Hervey gnl/gnlobject.c: Don't forward QoS event upstreams until we properly implement the time-shifting. Original commit message from CVS: * gnl/gnlobject.c: (ghostpad_event_function): Don't forward QoS event upstreams until we properly implement the time-shifting. See bug #398453 2007-06-16 16:54:04 +0000 Edward Hervey gnl/gnlcomposition.c: Unblock child source pads when the pads OR the elements are being removed. Original commit message from CVS: * gnl/gnlcomposition.c: (unblock_child_pads), (unblock_childs), (unlock_child_state), (lock_child_state), (gnl_composition_change_state), (object_pad_removed), (gnl_composition_add_object), (gnl_composition_remove_object): Unblock child source pads when the pads OR the elements are being removed. Also unblock those pads when going to READY/NULL. Lock the state of gnlobjects being added, else it screws up state changes of the composition. Properly unref items when using GstIterators. * gnl/gnloperation.c: (gnl_operation_class_init), (gnl_operation_finalize), (get_unused_static_sink_pad), (add_sink_pad): memleak fixes. * tests/check/common.h: * tests/check/simple.c: (GST_START_TEST): memleak fixes. 2007-06-15 16:36:42 +0000 Edward Hervey gnl/gnlsource.c: Let's stop using the main thread for other-thread processing and instead go the proper way and use y... Original commit message from CVS: * gnl/gnlsource.c: (ghost_seek_pad), (pad_blocked_cb): Let's stop using the main thread for other-thread processing and instead go the proper way and use yet-another-thread through g_thread_create(). * gnl/gnlcomposition.c: (gnl_composition_handle_message): Add a FIXME/WARNING related to switching to g_thread_create regarding SEGMENT_DONE message handling. This requires a better look at than I can give right now. 2007-05-22 11:58:46 +0000 Edward Hervey gnl/gnlsource.c: Use 'async-handling' property if available so our state change happens correctly with latest core mo... Original commit message from CVS: * gnl/gnlsource.c: (gnl_source_init): Use 'async-handling' property if available so our state change happens correctly with latest core modifications. 2007-05-22 11:50:17 +0000 Edward Hervey gnl/: Allow adding objects in a composition WHILE playing. Original commit message from CVS: * gnl/gnlcomposition.c: (get_new_seek_event), (get_current_position), (gnl_composition_event_handler), (refine_start_stop_in_region_above_priority), (convert_list_to_tree), (get_stack_list), (get_clean_toplevel_stack), (are_same_stacks), (update_pipeline), (gnl_composition_add_object): * gnl/gnlobject.c: (translate_incoming_seek): * gnl/gnlobject.h: Allow adding objects in a composition WHILE playing. Fixes #437105 2007-05-05 12:04:54 +0000 Edward Hervey configure.ac: Back to development cycle. Original commit message from CVS: * configure.ac: Back to development cycle. === release 0.10.8 === 2007-05-05 12:03:32 +0000 Edward Hervey * NEWS: * RELEASE: * configure.ac: Releasing GNonlin 0.10.8 Original commit message from CVS: Releasing GNonlin 0.10.8 2007-05-05 12:02:53 +0000 Edward Hervey * ChangeLog: I'm too lazy to comment this Original commit message from CVS: *** empty log message *** 2007-05-03 15:17:27 +0000 Edward Hervey configure.ac: 0.10.7.2 pre-release Original commit message from CVS: * configure.ac: 0.10.7.2 pre-release 2007-05-02 17:01:51 +0000 Edward Hervey gnl/gnlobject.h: priority is a guint32 ! Original commit message from CVS: * gnl/gnlobject.h: priority is a guint32 ! * gnl/gnlcomposition.c: (priority_comp), (objects_start_compare), (objects_stop_compare), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed): Fix major stupidity in GCompareFunc, the difference between 2 guint32 is not guaranteed to be correct within a gint :) Fixes #435215 2007-04-12 16:18:45 +0000 Edward Hervey gnl/gnlobject.c: Don't forget to unset the target when removing a ghost pad, else the target will still be linked to ... Original commit message from CVS: * gnl/gnlobject.c: (gnl_object_remove_ghost_pad): Don't forget to unset the target when removing a ghost pad, else the target will still be linked to the no-longer-present-ghostpad internal pad. This is the "+500 power points" Jokosher bugfix. 2007-02-28 21:25:53 +0000 Thomas Vander Stichele configure.ac: adapt to new AG_GST Original commit message from CVS: * configure.ac: adapt to new AG_GST 2007-02-19 15:37:45 +0000 Wim Taymans gnl/: Fix some leaks. Original commit message from CVS: * gnl/gnlcomposition.c: (no_more_pads_object_cb), (compare_relink_single_node), (update_pipeline): * gnl/gnlobject.c: (control_internal_pad), (gnl_object_ghost_pad_full): * gnl/gnlsource.c: (ghost_seek_pad): Fix some leaks. 2007-01-26 17:09:47 +0000 Edward Hervey configure.ac: Back to development cycle. Original commit message from CVS: * configure.ac: Back to development cycle. 2007-01-26 17:08:32 +0000 Edward Hervey * NEWS: * RELEASE: * configure.ac: Releasing GNonLin 0.10.7 "Anything better?" Original commit message from CVS: Releasing GNonLin 0.10.7 "Anything better?" ---------------------------------------------------------------------- 2007-01-26 17:07:48 +0000 Edward Hervey * ChangeLog: I'm too lazy to comment this Original commit message from CVS: *** empty log message *** 2007-01-24 17:08:14 +0000 Edward Hervey configure.ac: Bump requirement to GStreamer core 0.10.9 pre-releasing gnonlin 0.10.6.2 Original commit message from CVS: * configure.ac: Bump requirement to GStreamer core 0.10.9 pre-releasing gnonlin 0.10.6.2 2007-01-22 16:01:56 +0000 Edward Hervey gnl/gnlfilesource.c: Use decodebin2 if the USE_DECODEBIN2 env variable is set. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_init): Use decodebin2 if the USE_DECODEBIN2 env variable is set. 2007-01-10 16:55:26 +0000 Edward Hervey gnl/gnlcomposition.c: Add 'modify' boolean parameter to update_pipeline(). This is needed for cases where the timelin... Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_reset), (segment_done_main_thread), (seek_handling), (handle_seek_event), (gnl_composition_event_handler), (gnl_composition_ghost_pad_set_target), (convert_list_to_tree), (get_stack_list), (get_clean_toplevel_stack), (gnl_composition_change_state), (update_start_stop_duration), (no_more_pads_object_cb), (compare_relink_single_node), (compare_deactivate_single_node), (compare_relink_stack), (update_pipeline), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (gnl_composition_add_object), (gnl_composition_remove_object): Add 'modify' boolean parameter to update_pipeline(). This is needed for cases where the timeline is being updated because of modifications in the timeline. In that case, flush events need to be sent downstream so that the streaming threads are properly released and no longer blocking on downstream elements/pads. Add more debugging statements a bit everywhere. Re-indent all code. 2006-11-27 20:28:20 +0000 Edward Hervey RELEASE: Ouch, forgot to update the release file. Original commit message from CVS: * RELEASE: Ouch, forgot to update the release file. Personal note : REMEMBER IT'S HERE !!! 2006-11-27 20:06:27 +0000 Edward Hervey configure.ac: Back to development cycle. Original commit message from CVS: * configure.ac: Back to development cycle. === release 0.10.6 === 2006-11-27 20:03:45 +0000 Edward Hervey * NEWS: * configure.ac: GNonLin 0.10.6 "Who stole my beard ?" release Original commit message from CVS: GNonLin 0.10.6 "Who stole my beard ?" release ---------------------------------------------------------------------- 2006-11-27 20:02:53 +0000 Edward Hervey * ChangeLog: I'm too lazy to comment this Original commit message from CVS: *** empty log message *** 2006-11-23 20:01:11 +0000 Edward Hervey configure.ac: 0.10.5.2 pre-release Original commit message from CVS: * configure.ac: 0.10.5.2 pre-release 2006-11-03 13:52:07 +0000 Edward Hervey gnl/gnlcomposition.c: Recalculate start/stop/duration of composition when adding a default source. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_add_object): Recalculate start/stop/duration of composition when adding a default source. * tests/check/complex.c: (GST_START_TEST): Modify test so it also test for the case where the composition starts with a default source. 2006-11-03 11:46:52 +0000 Edward Hervey gnl/gnlcomposition.c: If we have a default source, the composition start value is always 0. Original commit message from CVS: * gnl/gnlcomposition.c: (handle_seek_event), (gnl_composition_event_handler), (update_start_stop_duration): If we have a default source, the composition start value is always 0. Don't rely on incoming seek event, create a new one from calculated values. 2006-11-02 11:59:55 +0000 Edward Hervey gnl/gnlcomposition.c: Take into account the priority of default sources when calculating the highest priority. This f... Original commit message from CVS: * gnl/gnlcomposition.c: (convert_list_to_tree): Take into account the priority of default sources when calculating the highest priority. This fixes #362399 * tests/check/gnloperation.c: (GST_START_TEST), (gnonlin_suite): Unit test for fixed behaviour. 2006-11-02 10:14:44 +0000 Edward Hervey gnl/gnlcomposition.c: Better check for the 'next stop' of a stack. It will now look at all objects under the highest ... Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_ghost_pad_set_target), (next_stop_in_region_above_priority), (convert_list_to_tree), (get_stack_list), (get_clean_toplevel_stack), (gnl_composition_remove_object): Better check for the 'next stop' of a stack. It will now look at all objects under the highest priority (and not the lowest as previously). Fixes #356380. Set new pads as active. Add some debug. * gnl/gnlobject.c: (gnl_object_ghost_pad_full): * gnl/gnloperation.c: (gnl_operation_add_element), (add_sink_pad): Activate new pads. * tests/check/gnloperation.c: (GST_START_TEST), (gnonlin_suite): Added new test (test_pyramid_operations) to test issues fixed for #356380. * tests/check/simple.c: (GST_START_TEST): Unref sinkpads before testing some issues. 2006-09-16 16:55:21 +0000 Edward Hervey tests/check/: gst_element_link_filtered() does not take the refcount on the caps. Original commit message from CVS: * tests/check/common.h: * tests/check/complex.c: (GST_START_TEST): gst_element_link_filtered() does not take the refcount on the caps. Therefore we unref the caps after use. 2006-09-16 16:36:10 +0000 Edward Hervey tests/check/common.h: Added convenience function to create a int/float audiotest gnlsource. Original commit message from CVS: * tests/check/common.h: Added convenience function to create a int/float audiotest gnlsource. * tests/check/complex.c: (GST_START_TEST), (gnonlin_suite): Added test_renegotiation to test re-negotiation error seen in #352827 2006-09-16 15:59:31 +0000 Edward Hervey tests/check/.cvsignore: add gnloperation Original commit message from CVS: * tests/check/.cvsignore: add gnloperation 2006-09-16 15:53:09 +0000 Edward Hervey tests/check/Makefile.am: add common.h Original commit message from CVS: * tests/check/Makefile.am: add common.h 2006-09-16 15:40:46 +0000 Edward Hervey tests/check/: Single header file that contains all the common functions and structures for the gnonlin tests. Original commit message from CVS: * tests/check/Makefile.am: * tests/check/common.h: Single header file that contains all the common functions and structures for the gnonlin tests. * tests/check/complex.c: * tests/check/gnlsource.c: (GST_START_TEST): * tests/check/simple.c: Switch to using common.h , makes test files smaller. * tests/check/gnloperation.c: (GST_START_TEST), (gnonlin_suite), (main): Created new test for gnloperations, starting with one that tests adding and removing operations, and checking that the outputted segments are correct. 2006-09-07 08:14:22 +0000 Edward Hervey gnl/gnlfilesource.c: Add pad templates to GnlFileSource so we can use it properly with gst-launch. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_class_init): Add pad templates to GnlFileSource so we can use it properly with gst-launch. 2006-08-31 11:48:47 +0000 Edward Hervey gnl/gnlcomposition.c: Set all properties of default sources in one g_object_set(). Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_add_object): Set all properties of default sources in one g_object_set(). 2006-08-30 15:58:34 +0000 Edward Hervey gnl/gnlcomposition.c: When unlinking source/operations , flush the parent's sink pad, so any stream locks are released. Original commit message from CVS: * gnl/gnlcomposition.c: (compare_deactivate_single_node): When unlinking source/operations , flush the parent's sink pad, so any stream locks are released. * gnl/gnloperation.c: (element_is_valid_filter): Scan pads AND templates to figure out if the element is a valid filter, and properly establish if it has dynamic pads, (get_unused_static_sink_pad): comments, (get_request_sink_pad): New function for getting request sink pads. (add_sink_pad): Add support for request sink pads, (synchronize_sinks): Don't try to add pads if it fails. * gnl/gnlsource.c: (ghost_seek_pad): Emit 'no-more-pads' after unblocking the controlled source pad. Not doing this would cause deadlocks in compositions with operations. 2006-08-25 15:01:13 +0000 Edward Hervey gnl/gnlcomposition.c: Remove the ghostpad if the composition becomes empty. Original commit message from CVS: * gnl/gnlcomposition.c: (update_pipeline): Remove the ghostpad if the composition becomes empty. 2006-08-24 13:00:29 +0000 Edward Hervey gnl/gnlcomposition.c: Only set ghostpad target and/or send seek event once all the objects of the stack are linked. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_reset), (gnl_composition_event_handler), (no_more_pads_object_cb), (compare_relink_single_node), (compare_relink_stack), (update_pipeline): Only set ghostpad target and/or send seek event once all the objects of the stack are linked. Fixes #352389 This is done through the use of comp->private->waitingpads. Simplified update_pipeline accordingly. Added more comments. 2006-08-24 12:59:30 +0000 Edward Hervey gnl/gnlcomposition.c: Only set ghostpad target and/or send seek event once all the objects of the stack are linked. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_reset), (gnl_composition_event_handler), (no_more_pads_object_cb), (compare_relink_single_node), (compare_relink_stack), (update_pipeline): Only set ghostpad target and/or send seek event once all the objects of the stack are linked. This is done through the use of comp->private->waitingpads. Simplified update_pipeline accordingly. Added more comments. 2006-08-23 16:58:55 +0000 Wim Taymans gnl/gnlobject.c: Add extra explanation for priority property, Fixes #352382. Original commit message from CVS: * gnl/gnlobject.c: (gnl_object_class_init), (control_internal_pad), (gnl_object_ghost_pad_no_target): Add extra explanation for priority property, Fixes #352382. 2006-08-23 10:52:42 +0000 Wim Taymans gnl/gnloperation.c: Fix leaks in iterators. Original commit message from CVS: * gnl/gnloperation.c: (gnl_operation_class_init), (element_is_valid_filter), (gnl_operation_add_element), (gnl_operation_set_sinks), (get_unused_static_sink_pad), (add_sink_pad), (gnl_operation_request_new_pad): Fix leaks in iterators. Correctly resync iterators too. 2006-08-18 11:10:22 +0000 Edward Hervey gnl/gnlobject.c: Free internal pad private data when the pad is being finalized, else we leak that structure. Original commit message from CVS: * gnl/gnlobject.c: (internal_pad_finalizing), (control_internal_pad): Free internal pad private data when the pad is being finalized, else we leak that structure. 2006-08-18 10:46:44 +0000 Edward Hervey gnl/gnlcomposition.c: Properly handle ghostpads without targets. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_ghost_pad_set_target): Properly handle ghostpads without targets. (convert_list_to_tree), (compare_relink_single_node), (compare_deactivate_single_node), (update_pipeline): First integration of support for GnlOperations. (gnl_composition_add_object): If composition was empty and initialized, update pipeline. 2006-08-09 16:44:00 +0000 Edward Hervey gnl/gnlobject.c: Remove useless linkfunc/unlinkfunc from GnlPadPrivate structure. Original commit message from CVS: * gnl/gnlobject.c: Remove useless linkfunc/unlinkfunc from GnlPadPrivate structure. (control_internal_pad), Remove call to useless functions. (gnl_object_ghost_pad_no_target), (gnl_object_ghost_pad_set_target): * gnl/gnloperation.c: (gnl_operation_class_init), (element_is_valid_filter), (gnl_operation_add_element), (gnl_operation_set_sinks), (get_unused_static_sink_pad), (add_sink_pad), (gnl_operation_request_new_pad), (gnl_operation_release_pad): Implemented GstElement::request_new_pad virtual method. Improved sink ghost pads synchronisation with the nbsinks property. * gnl/gnloperation.h: Added comments. 2006-07-27 16:00:35 +0000 Edward Hervey gnl/gnl.c: Let's welcome GnlOperation back in the game. Original commit message from CVS: * gnl/gnl.c: Let's welcome GnlOperation back in the game. * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_dispose), (gnl_composition_finalize), (gnl_composition_reset), (segment_done_main_thread), (gnl_composition_handle_message), (get_new_seek_event), (seek_handling), (pad_blocked), (gnl_composition_ghost_pad_set_target), (next_stop_in_region_above_priority), (convert_list_to_tree), (get_stack_list), (get_clean_toplevel_stack), (gnl_composition_change_state), (update_start_stop_duration), (no_more_pads_object_cb), (compare_relink_single_node), (compare_deactivate_single_node), (compare_relink_stack), (unlock_activate_stack), (update_pipeline), (object_pad_added), (gnl_composition_add_object), (gnl_composition_remove_object): * gnl/gnloperation.h: Switch to using a tree of GNode for the current stack. It's faster (since it does more checks in one go) and has more potential for speed improvements when updating the pipeline. * gnl/gnloperation.c: (gnl_operation_class_init), (gnl_operation_reset), (gnl_operation_init), (element_is_valid_filter), (get_src_pad), (gnl_operation_add_element), (gnl_operation_remove_element), (gnl_operation_set_property), (gnl_operation_get_property), (add_sink_pad), (remove_sink_pad), (synchronize_sinks): Added more comments and fixme's. * gnl/gnlobject.c: (internalpad_event_function), (ghostpad_event_function), (gnl_object_ghost_pad_full), (gnl_object_change_state): * gnl/gnlsource.c: (gnl_source_prepare), (element_pad_added_cb), (element_pad_removed_cb), (ghost_seek_pad), (gnl_source_change_state): gst-indent fixups. 2006-07-25 16:29:05 +0000 cascardo at holoscopio dot com gnl/gnlfilesource.c: Typo error. There's no filesource element, it's filesrc. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_init): Typo error. There's no filesource element, it's filesrc. Patch from cascardo at holoscopio dot com 2006-07-20 16:22:54 +0000 Edward Hervey configure.ac: Back to development cycle Original commit message from CVS: * configure.ac: Back to development cycle === release 0.10.5 === 2006-07-20 16:21:12 +0000 Edward Hervey * ChangeLog: * NEWS: 0.10.5 "Beavis and Zidane" release Original commit message from CVS: 0.10.5 "Beavis and Zidane" release 2006-07-20 15:46:41 +0000 Edward Hervey gnl/gnlcomposition.c: Fix typo in debug statement. Original commit message from CVS: * gnl/gnlcomposition.c: Fix typo in debug statement. 2006-07-19 12:31:06 +0000 Edward Hervey Memleak fixes in composition and tests. Original commit message from CVS: * gnl/gnlcomposition.c: (update_pipeline): * tests/check/complex.c: (sinkpad_event_probe), (GST_START_TEST): * tests/check/gnlsource.c: (sinkpad_event_probe), (GST_START_TEST): * tests/check/simple.c: (sinkpad_event_probe), (GST_START_TEST): Memleak fixes in composition and tests. 2006-07-19 11:36:57 +0000 Edward Hervey gnl/gnlsource.c: Try to send pending event before emitting no-more-pads and unblocking controlled src pad. Original commit message from CVS: * gnl/gnlsource.c: (ghost_seek_pad): Try to send pending event before emitting no-more-pads and unblocking controlled src pad. 2006-07-19 10:45:00 +0000 Edward Hervey gnl/gnlcomposition.c: Add default sources properly (sources with prioriy of G_MAXUINT32). Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_class_init), (hash_value_destroy), (gnl_composition_init), (get_stack_list), (update_start_stop_duration), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (gnl_composition_add_object), (gnl_composition_remove_object): Add default sources properly (sources with prioriy of G_MAXUINT32). Update pipeline when: _ object is added/removed within current playing segment _ object within current playing segment whose start/stop/priority/active property changes. * tests/check/complex.c: (GST_START_TEST): * tests/check/gnlsource.c: (GST_START_TEST): Fixed tests to test more behaviours, including new ones. 2006-07-16 15:47:36 +0000 Edward Hervey gnonlin.pc.in: removing file from CVS. Original commit message from CVS: * gnonlin.pc.in: removing file from CVS. * configure.ac: pre-release 0.10.4.3 2006-07-11 16:41:10 +0000 Edward Hervey configure.ac: Require core >= 0.10.8.1 (for ghostpad and blocking fixes). Original commit message from CVS: * configure.ac: Require core >= 0.10.8.1 (for ghostpad and blocking fixes). * gnl/gnlcomposition.c: (hash_value_destroy), (unlock_childs), (gnl_composition_reset), (segment_done_main_thread), (get_new_seek_event), (handle_seek_event), (gnl_composition_ghost_pad_set_target), (gnl_composition_change_state), (no_more_pads_object_cb), (update_pipeline), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (object_pad_added), (gnl_composition_add_object), (gnl_composition_remove_object): All source pads created are now blocked by default. Don't lock the state of sources in READY. Let them all reach paused. Remove cruft now fixed in GStreamer core. * gnl/gnlobject.c: (internalpad_event_function), (ghostpad_event_function), (control_internal_pad), (ghostpad_link_function), (gnl_object_ghost_pad_full), (gnl_object_ghost_pad_no_target), (gnl_object_ghost_pad_set_target): * gnl/gnlobject.h: Remove all the cruft which is now fixed in GStreamer core. * gnl/gnlsource.c: (gnl_source_prepare), (element_pad_added_cb), (element_pad_removed_cb), (ghost_seek_pad), (pad_blocked_cb), (gnl_source_change_state): Remove all the pad blocking on events with ghostpad cruft which is now fixed in GStreamer core. 2006-07-03 09:57:02 +0000 Edward Hervey Make gnonlin build properly on CygWin. Original commit message from CVS: * Makefile.am: * configure.ac: Make gnonlin build properly on CygWin. Fixes #341492 2006-06-18 11:55:59 +0000 Edward Hervey docs/random/design: Precision about gnloperation sinks property. Original commit message from CVS: * docs/random/design: Precision about gnloperation sinks property. * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_finalize), (lock_child_state), (gnl_composition_reset), (segment_done_main_thread), (gnl_composition_handle_message), (seek_handling), (handle_seek_event), (pad_blocked), (gnl_composition_ghost_pad_set_target), (update_start_stop_duration), (compare_relink_stack), (update_pipeline), (object_start_changed), (object_stop_changed): Non-racy seeking handling. All seeks are put in the main thread. A lock-protected value takes care of discarding previous seeks. This avoids freeze situations when scrubbing/seeking a lot. Use of pad blocks for non-used sources. * gnl/gnlobject.c: (internalpad_event_function), (gnl_pad_set_blocked_async), (gnl_pad_add_event_probe), (gnl_pad_remove_event_probe): * gnl/gnlobject.h: Emit 'segment-start' message when we receive a newsegment event. This allows the composition to know when a seek is really starting and therefore stop all pending seeks. Added functions (hacks) for recursive ghostpad handling. Should go away once the issue is fixed in core (See #341029). * gnl/gnlsource.c: (element_pad_added_cb), (element_pad_removed_cb), (ghost_seek_pad), (gnl_source_change_state): Use gnl_pad_* recursive ghostpads hack functions. * tests/check/simple.c: (GST_START_TEST): Fix comment 2006-06-12 13:15:39 +0000 Edward Hervey configure.ac: pre-release 0.10.4.2 Original commit message from CVS: * configure.ac: pre-release 0.10.4.2 2006-06-12 12:43:43 +0000 Edward Hervey tests/check/complex.c: Added unit-test for the previous fixes to gnlcomposition with default sources. Original commit message from CVS: * tests/check/complex.c: (GST_START_TEST), (gnonlin_suite): Added unit-test for the previous fixes to gnlcomposition with default sources. 2006-06-12 11:33:05 +0000 Edward Hervey gnl/gnlcomposition.c: Fixed wrong get_stack_list(), stop at the correct place. Original commit message from CVS: * gnl/gnlcomposition.c: (next_stop_in_region_above_priority), (get_stack_list): Fixed wrong get_stack_list(), stop at the correct place. Stop iterating as soon as possible in next_stop_in_region..(). 2006-06-11 11:22:58 +0000 Edward Hervey docs/random/design: Fleshed up design doc a bit more. Original commit message from CVS: * docs/random/design: Fleshed up design doc a bit more. 2006-06-10 23:27:10 +0000 Thomas Vander Stichele * tests/check/.gitignore: ignore more and fix the build Original commit message from CVS: ignore more and fix the build 2006-06-08 15:44:59 +0000 Edward Hervey gnl/gnlcomposition.c: Composition now looks for anything hiding the configured segment, in order to properly adjust t... Original commit message from CVS: * gnl/gnlcomposition.c: (next_stop_in_region_above_priority), (get_clean_toplevel_stack): Composition now looks for anything hiding the configured segment, in order to properly adjust the end of the configured segment. This should make default sources work. * gnl/gnloperation.c: (gnl_operation_class_init), (gnl_operation_init), (element_is_valid_filter), (gnl_operation_add_element), (gnl_operation_remove_element), (gnl_operation_set_sinks), (gnl_operation_set_property), (gnl_operation_get_property), (add_sink_pad), (remove_sink_pad), (synchronize_sinks), (gnl_operation_prepare): * gnl/gnloperation.h: First big update. Still some more work to do though. * tests/check/complex.c: (gnonlin_suite): Re-activated testsuite to check for 'hiding' situations. 2006-06-06 16:29:55 +0000 Edward Hervey docs/random/design: Updated to real 0.10 behaviour + added details. Original commit message from CVS: * docs/random/design: Updated to real 0.10 behaviour + added details. 2006-05-31 09:43:48 +0000 Edward Hervey gnl/gnlsource.c: initial seek event is in object time realm. Original commit message from CVS: * gnl/gnlsource.c: (gnl_source_prepare): initial seek event is in object time realm. 2006-05-23 10:02:05 +0000 Edward Hervey gnl/gnlcomposition.c: Protect critical part with comp_objects_lock. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_dispose), (gnl_composition_ghost_pad_set_target), (update_pipeline), (gnl_composition_event_handler): Protect critical part with comp_objects_lock. Don't unref pad given to gnl_composition_ghost_pad_set_target(). * tests/check/complex.c: (GST_START_TEST): * tests/check/gnlsource.c: (GST_START_TEST): * tests/check/simple.c: (GST_START_TEST): memleak fixes in tests. 2006-05-22 16:18:43 +0000 Edward Hervey gnl/gnlcomposition.c: Change the composition's ghostpad target at the same time as we send the seek event. This avoid... Original commit message from CVS: * gnl/gnlcomposition.c: (no_more_pads_object_cb), (update_pipeline): Change the composition's ghostpad target at the same time as we send the seek event. This avoids the case where we check for the pad availability too early (not present) and then never switch the ghostpad target. * gnl/gnlsource.c: (gnl_source_change_state): if we don't have a controlled element, don't even care about calling parent's _change_state() 2006-05-21 09:21:18 +0000 Edward Hervey gnl/gnlsource.c: Handle the case where more than one pads of the controlled element is a valid src pad (intersects wi... Original commit message from CVS: * gnl/gnlsource.c: (element_pad_added_cb), (ghost_seek_pad), (gnl_source_change_state): Handle the case where more than one pads of the controlled element is a valid src pad (intersects with the caps). 2006-05-11 12:07:31 +0000 Edward Hervey configure.ac: Back to development cycle. Original commit message from CVS: * configure.ac: Back to development cycle. === release 0.10.4 === 2006-05-11 12:06:15 +0000 Edward Hervey * ChangeLog: * NEWS: * configure.ac: Releasing 0.10.4 "Solomillo de Ternera" Original commit message from CVS: Releasing 0.10.4 "Solomillo de Ternera" 2006-05-10 18:02:58 +0000 Edward Hervey gnl/gnlcomposition.c: Fix for correctly playing compositions with gaps between objects. Original commit message from CVS: * gnl/gnlcomposition.c: (get_clean_toplevel_stack), (update_pipeline): Fix for correctly playing compositions with gaps between objects. * tests/check/Makefile.am: * tests/check/complex.c: (gst_element_factory_make_or_warn), (composition_pad_added_cb), (compare_segments), (sinkpad_event_probe), (sinkpad_buffer_probe), (videotest_gnl_src), (videotest_in_bin_gnl_src), (segment_new), (GST_START_TEST), (gnonlin_suite), (main): Added new unittests for compositions with gaps between objects. * tests/check/simple.c: (gnonlin_suite): Moved de-activated test_one_above_another to complex 2006-05-10 14:48:50 +0000 Edward Hervey * configure.ac: Reverting cruft that went in previous commit to configure.ac Original commit message from CVS: Reverting cruft that went in previous commit to configure.ac 2006-05-10 14:44:10 +0000 Edward Hervey configure.ac: 0.10.3.2 pre-release Original commit message from CVS: * configure.ac: 0.10.3.2 pre-release 2006-05-08 16:47:47 +0000 Edward Hervey gnl/gnlsource.c: Don't call ghost_seek_pad() in the event probe and only discard the events. The buffer probe will ev... Original commit message from CVS: * gnl/gnlsource.c: (pad_event_probe): Don't call ghost_seek_pad() in the event probe and only discard the events. The buffer probe will eventually take care of that. This guarantees that upstream elements are initialized and ready to handle seek events. 2006-05-08 14:04:34 +0000 Edward Hervey gnl/: Indent properly using gstreamer/tools/gst-indent Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_finalize), (gnl_composition_handle_message), (gnl_composition_ghost_pad_set_target), (gnl_composition_change_state), (no_more_pads_object_cb), (update_pipeline): * gnl/gnlfilesource.c: (gnl_filesource_init): * gnl/gnlobject.c: (translate_incoming_seek), (gnl_object_ghost_pad_full), (gnl_object_remove_ghost_pad), (gnl_object_change_state): * gnl/gnlsource.c: (gnl_source_class_init), (gnl_source_prepare), (element_pad_added_cb), (element_pad_removed_cb), (ghost_seek_pad), (pad_event_probe), (has_dynamic_srcpads), (gnl_source_control_element_func), (gnl_source_add_element), (gnl_source_remove_element), (gnl_source_change_state): * gnl/gnlsource.h: Indent properly using gstreamer/tools/gst-indent 2006-05-08 14:00:10 +0000 Edward Hervey gnl/gnlcomposition.c: Convert non-fatal GST_WARNING to GST_DEBUG. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_ghost_pad_set_target), (get_src_pad), (update_pipeline): Convert non-fatal GST_WARNING to GST_DEBUG. 2006-05-08 12:02:29 +0000 Edward Hervey autogen.sh: Update dependency checking Original commit message from CVS: * autogen.sh: Update dependency checking 2006-05-08 11:29:53 +0000 Edward Hervey gnl/gnlfilesource.c: No more redundant bin containing filesrc and decodebin. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_class_init), (gnl_filesource_init): No more redundant bin containing filesrc and decodebin. We use the GnlSource controls_one and control_element() class properties to inform GnlSource it should control decodebin but still accept filesrc. * gnl/gnlsource.c: (gnl_source_class_init), (gnl_source_control_element_func), (gnl_source_add_element), (gnl_source_remove_element): * gnl/gnlsource.h: Added two class properties: _ controls_one which tells whether the class controls only one object _ control_element() virtual_method which should be called by subclasses that have several elements to specify which element should be checked for pad-added/pad-removed. 2006-05-05 16:23:06 +0000 Edward Hervey gnl/gnlfilesource.*: Simplify GnlFileSource to be a subclass of GnlSource. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_class_init), (decodebin_pad_added_cb), (find_ghost_pad), (decodebin_pad_removed_cb), (gnl_filesource_init): * gnl/gnlfilesource.h: Simplify GnlFileSource to be a subclass of GnlSource. * gnl/gnlsource.c: (element_pad_added_cb), (gnl_source_change_state): Add event probe for dynamic pads too. Don't error out in state_change. 2006-05-05 12:34:32 +0000 Edward Hervey gnl/: Remove gnlmarshal cruft that dates from... well.... Original commit message from CVS: * gnl/Makefile.am: * gnl/gnlfilesource.c: * gnl/gnlmarshal.list: * gnl/gnlobject.c: * gnl/gnloperation.c: * gnl/gnlsource.c: Remove gnlmarshal cruft that dates from... well.... Ok, fine, that was here since 0.8 era and not used since :) 2006-05-05 12:10:23 +0000 Edward Hervey gnl/gnlcomposition.c: If srcpad of top gnlobject is not available, store pending child seek so we can send it when th... Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_dispose), (lock_child_state), (gnl_composition_reset), (get_new_seek_event), (gnl_composition_change_state), (no_more_pads_object_cb), (update_pipeline): If srcpad of top gnlobject is not available, store pending child seek so we can send it when the pad appears. Lock all childs in READY. They will be automatically (un)locked in update_pipeline() Add GST_SEEK_FLAG_ACCURATE to the seek event. * gnl/gnlobject.c: (translate_incoming_seek), (gnl_object_change_state): Add GST_SEEK_FLAG_ACCURATE to incoming seek events if it wasn't already there. GnlObject::update() virtual method is called in READY->PAUSED before calling the parent _change_state() * gnl/gnlsource.c: (gnl_source_class_init), (gnl_source_prepare), (element_pad_removed_cb), (ghost_seek_pad), (pad_blocked_cb), (pad_event_probe), (gnl_source_add_element), (gnl_source_change_state): We need to block on events. Added an event_probe for that. Implement GnlObject::prepare() so we can create a seek event if the GnlSource is used stand-alone (not in composition). * tests/check/Makefile.am: * tests/check/gnlsource.c: (gst_element_factory_make_or_warn), (gnlsource_pad_added_cb), (compare_segments), (sinkpad_event_probe), (sinkpad_buffer_probe), (videotest_gnl_src), (videotest_in_bin_gnl_src), (segment_new), (GST_START_TEST), (gnonlin_suite), (main): New unit test for testing standalone GnlSource behaviour * tests/check/simple.c: (composition_pad_added_cb), (sinkpad_event_probe), (sinkpad_buffer_probe), (videotest_in_bin_gnl_src), (GST_START_TEST), (gnonlin_suite): I have to admit the tests were... really borked. Now they're much more sensitive. 2006-04-25 15:44:10 +0000 Edward Hervey configure.ac: Back to development cycle Original commit message from CVS: * configure.ac: Back to development cycle === release 0.10.3 === 2006-04-25 15:42:21 +0000 Edward Hervey * NEWS: * configure.ac: Releasing 0.10.3 "Birthday Cake" Original commit message from CVS: Releasing 0.10.3 "Birthday Cake" 2006-04-25 15:41:28 +0000 Edward Hervey * ChangeLog: I'm too lazy to comment this Original commit message from CVS: *** empty log message *** 2006-04-25 11:28:43 +0000 Edward Hervey gnl/gnlfilesource.c: First event sent in _prepare() should have the flush flag. Original commit message from CVS: * gnl/gnlfilesource.c: (gnl_filesource_prepare), (gnl_filesource_send_event): First event sent in _prepare() should have the flush flag. Smells like a second paperbag release... 2006-04-21 16:35:27 +0000 Edward Hervey configure.ac: Back to development cycle Original commit message from CVS: * configure.ac: Back to development cycle === release 0.10.2 === 2006-04-21 16:33:33 +0000 Edward Hervey * ChangeLog: * NEWS: * common: * configure.ac: Releasing 0.10.2 "... Y un cafe solo" Original commit message from CVS: Releasing 0.10.2 "... Y un cafe solo" 2006-04-10 12:05:52 +0000 Edward Hervey gnl/gnlobject.c: 'really' wrong state_change handling Original commit message from CVS: * gnl/gnlobject.c: (gnl_object_change_state): 'really' wrong state_change handling 2006-04-07 09:18:02 +0000 Edward Hervey COPYING: Convert all remaining files to be coherent with the change to LGPL licensing. Original commit message from CVS: * COPYING: Convert all remaining files to be coherent with the change to LGPL licensing. 2006-04-07 09:14:17 +0000 Edward Hervey configure.ac: Back to development (for good) Original commit message from CVS: * configure.ac: Back to development (for good) === release 0.10.1 === 2006-04-06 17:03:33 +0000 Thomas Vander Stichele * ChangeLog: * NEWS: * RELEASE: * configure.ac: releasing 0.10.1 Original commit message from CVS: releasing 0.10.1 2006-04-06 16:53:51 +0000 Thomas Vander Stichele * po/Makefile.in.in: * po/Rules-quot: * po/boldquot.sed: * po/en@boldquot.header: * po/en@quot.header: * po/insert-header.sin: * po/quot.sed: * po/remove-potcdate.sin: remove po dir Original commit message from CVS: remove po dir 2006-04-06 15:47:50 +0000 Edward Hervey configure.ac: back to development Original commit message from CVS: * configure.ac: back to development 2006-04-06 15:45:22 +0000 Edward Hervey * configure.ac: idiotic cicl Original commit message from CVS: idiotic cicl 2006-04-06 15:42:00 +0000 Edward Hervey * ChangeLog: * common: I'm too lazy to comment this Original commit message from CVS: *** empty log message *** 2006-04-05 17:46:30 +0000 Edward Hervey configure.ac: 0.10.0.11 pre-release Original commit message from CVS: * configure.ac: 0.10.0.11 pre-release 2006-04-05 17:45:37 +0000 Edward Hervey gnl/gnlcomposition.c: more memleak fixes. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_finalize), (gnl_composition_handle_message), (gnl_composition_change_state), (gnl_composition_add_object), (gnl_composition_remove_object): more memleak fixes. * gnl/gnlsource.c: (gnl_source_dispose), (gnl_source_finalize): more debug. * tests/check/simple.c: (GST_START_TEST), (gnonlin_suite): Add more debug and refcount checks. 2006-04-05 13:47:07 +0000 Edward Hervey * tests/check/Makefile.am: non debug, makes thomasvs blood pressure go up too quickly... and makes the buildbot go crazy too. Original commit message from CVS: non debug, makes thomasvs blood pressure go up too quickly... and makes the buildbot go crazy too. 2006-04-05 13:32:00 +0000 Edward Hervey * tests/check/Makefile.am: debug Original commit message from CVS: debug 2006-04-05 13:10:17 +0000 Edward Hervey tests/check/Makefile.am: Increasing timeout for tests since we're asking a lot Original commit message from CVS: * tests/check/Makefile.am: Increasing timeout for tests since we're asking a lot 2006-04-05 13:03:53 +0000 Edward Hervey gnl/gnlcomposition.c: Only force state change in (update_pipeline) when not being called from state_change(). Original commit message from CVS: * gnl/gnlcomposition.c: (ready_and_lock_child_state), (gnl_composition_handle_message), (handle_seek_event), (gnl_composition_change_state), (update_pipeline), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (gnl_composition_add_object), (gnl_composition_remove_object): Only force state change in (update_pipeline) when not being called from state_change(). Properly emit EOS/SEGMENT_DONE at the end, but don't emit SEGMENT_DONE for intermediate steps. * gnl/gnlfilesource.c: cleanup * gnl/gnlobject.c: (gnl_object_ghost_pad_full), (gnl_object_remove_ghost_pad): more debug statements * gnl/gnlsource.c: (gnl_source_class_init), (gnl_source_dispose), (gnl_source_finalize), (element_pad_added_cb), (element_pad_removed_cb), (ghost_seek_pad), (pad_blocked_cb), (has_dynamic_srcpads), (gnl_source_add_element), (gnl_source_remove_element), (gnl_source_send_event), (gnl_source_change_state): Make GnlSource more generic. It now works with dynamic and non-dynamic source pads. Eventually we'll be able to subclass GnlFileSource from GnlSource and remove most of the duplicated code. * tests/check/simple.c: (GST_START_TEST): Removed wrong newsegment checks Added lots more assertions 2006-04-05 09:10:51 +0000 Edward Hervey tests/check/Makefile.am: Remove buildbot debug craziness since I now know what the problem is. Original commit message from CVS: * tests/check/Makefile.am: Remove buildbot debug craziness since I now know what the problem is. 2006-04-05 09:06:42 +0000 Christian Schaller * common: * gnonlin.spec.in: removing uneeded stuff from the spec file Original commit message from CVS: removing uneeded stuff from the spec file 2006-04-04 15:25:27 +0000 Edward Hervey tests/check/simple.c: Let's do a simple link. Original commit message from CVS: * tests/check/simple.c: (composition_pad_added_cb): Let's do a simple link. 2006-04-04 14:54:42 +0000 Edward Hervey tests/check/Makefile.am: GST_DEBUG=*:4 Original commit message from CVS: * tests/check/Makefile.am: GST_DEBUG=*:4 2006-04-04 14:49:09 +0000 Edward Hervey tests/check/Makefile.am: Higher debug level :( Original commit message from CVS: * tests/check/Makefile.am: Higher debug level :( 2006-04-04 14:05:18 +0000 Edward Hervey gnl/gnlcomposition.c: If new target and previous target are the same, don't retarget. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_ghost_pad_set_target): If new target and previous target are the same, don't retarget. 2006-04-04 12:37:52 +0000 Edward Hervey tests/check/Makefile.am: Forcing GST_DEBUG to figure out what's going wrong with the buildbots. Original commit message from CVS: * tests/check/Makefile.am: Forcing GST_DEBUG to figure out what's going wrong with the buildbots. 2006-04-04 12:25:20 +0000 Edward Hervey gnl/gnlobject.c: Bug in converting media time over media_stop Original commit message from CVS: * gnl/gnlobject.c: (gnl_media_to_object_time): Bug in converting media time over media_stop 2006-04-04 11:16:41 +0000 Edward Hervey tests/check/simple.c: Added new test to check start/duration/stop correctness for sources and compositions. Original commit message from CVS: * tests/check/simple.c: (GST_START_TEST), (gnonlin_suite): Added new test to check start/duration/stop correctness for sources and compositions. 2006-04-04 10:54:10 +0000 Edward Hervey gnl/gnlobject.c: Free the GnlPadPrivate once we have removed the pad. Original commit message from CVS: * gnl/gnlobject.c: (gnl_object_remove_ghost_pad): Free the GnlPadPrivate once we have removed the pad. 2006-04-04 09:29:58 +0000 Edward Hervey gnl/gnlcomposition.c: update_pipeline was returning the wrong information. Original commit message from CVS: * gnl/gnlcomposition.c: (update_pipeline): update_pipeline was returning the wrong information. It now returns FALSE only if there was an error updating the pipeline. * tests/check/simple.c: (GST_START_TEST): The test fails if the state_change() DOES return FAILURE. 2006-04-03 13:32:23 +0000 Edward Hervey configure.ac: New pre-release (0.10.0.10). Original commit message from CVS: * configure.ac: New pre-release (0.10.0.10). Required version of gstreamer core/-base is 0.10.4. * tests/check/simple.c: (gnonlin_suite): Only run the tests with core > 0.10.4.0 2006-04-03 11:25:40 +0000 Edward Hervey tests/check/simple.c: When the pipeline is set to NULL, on slow/loaded systems the pipeline might not have reached th... Original commit message from CVS: * tests/check/simple.c: (GST_START_TEST): When the pipeline is set to NULL, on slow/loaded systems the pipeline might not have reached that state yet when checking for the refcount of the pipeline/bus. We now check that refcount is between 1 and 2 now. 2006-04-03 09:51:48 +0000 Edward Hervey gnl/: Indent the source code using gstreamer/tools/gst-indent Original commit message from CVS: * gnl/gnl.c: (plugin_init): * gnl/gnlcomposition.c: (gnl_composition_class_init), (hash_value_destroy), (gnl_composition_init), (gnl_composition_dispose), (gnl_composition_finalize), (unlock_child_state), (ready_and_lock_child_state), (gnl_composition_reset), (gnl_composition_handle_message), (priority_comp), (have_to_update_pipeline), (get_new_seek_event), (handle_seek_event), (gnl_composition_event_handler), (gnl_composition_ghost_pad_set_target), (get_stack_list), (get_clean_toplevel_stack), (get_src_pad), (gnl_composition_prepare), (gnl_composition_change_state), (objects_start_compare), (objects_stop_compare), (update_start_stop_duration), (no_more_pads_object_cb), (compare_relink_stack), (update_pipeline), (object_start_changed), (object_stop_changed), (object_priority_changed), (object_active_changed), (object_pad_removed), (gnl_composition_add_object), (gnl_composition_remove_object): * gnl/gnlcomposition.h: * gnl/gnlfilesource.c: (gnl_filesource_class_init), (compare_src_pad), (get_valid_src_pad), (ghost_seek_pad), (pad_blocked_cb), (decodebin_new_pad_cb), (decodebin_pad_removed_cb), (gnl_filesource_init), (gnl_filesource_dispose), (gnl_filesource_finalize), (gnl_filesource_prepare), (gnl_filesource_send_event), (gnl_filesource_set_property), (gnl_filesource_get_property): * gnl/gnlfilesource.h: * gnl/gnlobject.c: (gnl_object_dispose), (ghostpad_event_function), (control_internal_pad), (ghostpad_link_function), (ghostpad_unlink_function), (gnl_object_change_state): * gnl/gnlobject.h: * gnl/gnloperation.c: (gnl_operation_class_init), (gnl_operation_init): * gnl/gnloperation.h: * gnl/gnlsource.c: (gnl_source_class_init), (gnl_source_init), (gnl_source_dispose), (gnl_source_finalize), (compare_src_pad), (get_valid_src_pad), (no_more_pads_in_child), (gnl_source_add_element), (gnl_source_remove_element): * gnl/gnlsource.h: Indent the source code using gstreamer/tools/gst-indent 2006-04-03 08:49:17 +0000 Thomas Vander Stichele clean up like the other modules Original commit message from CVS: * configure.ac: * gnl/Makefile.am: * gnl/gnl.c: clean up like the other modules 2006-03-31 17:16:43 +0000 Edward Hervey tests/check/simple.c: more checking. Original commit message from CVS: * tests/check/simple.c: (GST_START_TEST): more checking. 2006-03-31 17:01:48 +0000 Edward Hervey configure.ac: Add debug, valgrind and gcov. Remove c++/malloc cruft Original commit message from CVS: * configure.ac: Add debug, valgrind and gcov. Remove c++/malloc cruft * gnl/gnlcomposition.c: (hash_value_destroy), (gnl_composition_dispose), (gnl_composition_finalize), (unlock_child_state), (ready_and_lock_child_state), (gnl_composition_reset), (gnl_composition_handle_message), (gnl_composition_change_state), (update_pipeline), (object_pad_removed), (gnl_composition_add_object): Handle pads being removed on controlled objects, and remove the composition ghostpad if the object is the top-level one. Memleak fixes. Properly lock/unlock the state of the childs between READY and PAUSED. Properly return success or not in update_pipeline() * gnl/gnlfilesource.c: (decodebin_pad_removed_cb): Use gnl_object_remove_ghost_pad () * gnl/gnlobject.c: (gnl_object_class_init), (gnl_object_dispose), (translate_incoming_seek), (translate_outgoing_new_segment), (internalpad_event_function), (control_internal_pad), (ghostpad_link_function), (ghostpad_unlink_function), (gnl_object_remove_ghost_pad), (gnl_object_change_state): * gnl/gnlobject.h: Add gnl_object_remove_ghost_pad() to properly remove the GnlPadPrivate data from the ghostpads used in GnlObjects. Memleak fixes. * gnl/gnlsource.c: (gnl_source_remove_element): * tests/check/Makefile.am: * tests/check/simple.c: (composition_pad_added_cb), (GST_START_TEST): 2006-03-30 11:07:04 +0000 Thomas Vander Stichele use pluginsdir exposed by core and base .pc files to find elements like videotestsrc and fakesink Original commit message from CVS: * Makefile.am: * configure.ac: * tests/check/Makefile.am: use pluginsdir exposed by core and base .pc files to find elements like videotestsrc and fakesink 2006-03-30 10:16:26 +0000 Edward Hervey autogen.sh: Fix up autogen.sh for common/ since we're now in gstreamer cvs. Original commit message from CVS: * autogen.sh: Fix up autogen.sh for common/ since we're now in gstreamer cvs. 2006-03-30 10:10:23 +0000 Edward Hervey Makefile.am: test/directory is gone Original commit message from CVS: * Makefile.am: test/directory is gone 2006-03-30 09:55:39 +0000 Thomas Vander Stichele * .gitignore: * tests/Makefile.am: * tests/check/.gitignore: ignore more Original commit message from CVS: ignore more 2006-03-30 09:54:31 +0000 Thomas Vander Stichele * test/.gitignore: * test/Makefile.am: * test/composition.c: * test/composition2.c: * test/dynamic1.c: * test/dynamic2.c: * test/group.c: * test/layertest.c: * test/layertest2.c: * test/layertest3.c: * test/layertest4.c: * test/pipelines.c: * test/pipelines.h: * test/simple.c: * test/simple1.c: * test/simple2.c: * test/simple3.c: * test/simple4.c: * test/simple5.c: * test/sourcetest1.c: * test/sourcetest2.c: * test/sourcetest3.c: * test/sourcetest4.c: * test/vlayertest.c: test was moved Original commit message from CVS: test was moved 2006-03-30 09:53:56 +0000 Thomas Vander Stichele make similar to the other gstreamer modules remove gettext stuff Original commit message from CVS: * Makefile.am: * autogen.sh: * configure.ac: * po/Makefile.in.in: * tests/check/Makefile.am: * tests/check/simple.c: (composition_pad_added_cb), (compare_segments), (sinkpad_event_probe), (videotest_gnl_src), (segment_new), (GST_START_TEST), (gnonlin_suite), (main): make similar to the other gstreamer modules remove gettext stuff 2006-03-29 17:21:00 +0000 Edward Hervey configure.ac: pre-release Original commit message from CVS: * configure.ac: pre-release * gnl/gnl.c: Remove gnloperation for (pre-)release * test/Makefile.am: Fix for GST_PLUGIN_PATH since the tests use videotestsrc. * test/simple.c: (gst_element_factory_make_or_warn), (videotest_gnl_src), (GST_START_TEST), (gnonlin_suite): Makes checks for element_factory_make() Make sure we are looking for the proper values. Comment out known broken test for (pre-)release 2006-03-29 10:06:33 +0000 Edward Hervey Makefile.am: whoops, trailing back-slash Original commit message from CVS: * Makefile.am: whoops, trailing back-slash 2006-03-29 10:03:08 +0000 Edward Hervey Removing pkgconfig file generation. It's no longer needed since GNonLin are GStreamer plugins and check should be don... Original commit message from CVS: * Makefile.am: * configure.ac: * pkgconfig/Makefile.am: * pkgconfig/gnonlin.pc.in: Removing pkgconfig file generation. It's no longer needed since GNonLin are GStreamer plugins and check should be done using the existing tools, and/or at runtime. 2006-03-29 09:57:17 +0000 Edward Hervey Added testsuite system. Original commit message from CVS: * configure.ac: * test/Makefile.am: * test/simple.c: (composition_pad_added_cb), (compare_segments), (sinkpad_event_probe), (videotest_gnl_src), (segment_new), (GST_START_TEST), (gnonlin_suite), (main): Added testsuite system. Added first batch of test to test: _ proper start/stop/duration settings/update on sources and compositions _ proper data flow in various source combination 2006-03-28 16:38:53 +0000 Edward Hervey gnl/gnlcomposition.c: Add pad template. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_class_init), (gnl_composition_reset), (gnl_composition_event_handler), (gnl_composition_ghost_pad_set_target), (get_clean_toplevel_stack), (no_more_pads_object_cb), (update_pipeline): Add pad template. More debug. * gnl/gnlfilesource.c: (decodebin_pad_removed_cb), (gnl_filesource_init): Add pad template Reset the ghostpad if the decodebin pad gets removed. * gnl/gnlobject.c: (gnl_object_class_init), (gnl_object_cleanup_func), (gnl_object_cleanup), (gnl_object_change_state): Call GnlObject::prepare AFTER calling parent change_state, otherwise pads won't have been re-activated, flushing flash won't have been removed and seek on the pads will definitely not work. Essential for re-using gnlobjects. * gnl/gnlobject.h: * gnl/gnlsource.c: (gnl_source_class_init): Add pad template 2006-03-24 16:29:21 +0000 Wim Taymans gnl/gnlobject.*: Re-indent, move uncommon code out of the main flow. Original commit message from CVS: * gnl/gnlobject.c: (gnl_object_class_init), (gnl_object_init), (gnl_object_to_media_time), (gnl_media_to_object_time), (gnl_object_covers_func), (gnl_object_covers), (gnl_object_prepare_func), (gnl_object_prepare), (gnl_object_release_pad), (translate_incoming_seek), (translate_outgoing_seek), (translate_outgoing_new_segment), (internalpad_event_function), (translate_incoming_position_query), (internalpad_query_function), (internalpad_unlink_function), (flush_hack_check), (ghostpad_event_function), (ghostpad_query_function), (control_internal_pad), (ghostpad_link_function), (ghostpad_unlink_function), (gnl_object_ghost_pad_full), (gnl_object_ghost_pad), (gnl_object_ghost_pad_no_target), (gnl_object_ghost_pad_set_target), (translate_message_segment_start), (translate_message_segment_done), (gnl_object_handle_message), (gnl_object_set_caps), (update_values), (gnl_object_set_property), (gnl_object_get_property), (gnl_object_change_state): * gnl/gnlobject.h: Re-indent, move uncommon code out of the main flow. * gnl/gnlsource.c: (gnl_source_init), (gnl_source_dispose), (gnl_source_finalize), (no_more_pads_in_child), (gnl_source_add_element), (gnl_source_remove_element): * gnl/gnlsource.h: cleanups. 2006-02-20 18:24:36 +0000 Edward Hervey gnl/gnlcomposition.c: Fix proper pipeline update when going from one segment to another (as opposed to seek changes). Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_handle_message), (get_new_seek_event), (get_stack_list), (get_clean_toplevel_stack), (update_pipeline): Fix proper pipeline update when going from one segment to another (as opposed to seek changes). More efficient _update_pipeline() Lots of debug cleanups. * gnl/gnlobject.c: (translate_incoming_position_query), (ghostpad_query_function), (translate_message_segment_done): Implement position query time-shifting. 2006-01-27 09:48:08 +0000 Edward Hervey configure.ac: Added requirement for explicit --prefix setting, so that GNonLin doesn't get installed in a path that i... Original commit message from CVS: * configure.ac: Added requirement for explicit --prefix setting, so that GNonLin doesn't get installed in a path that isn't checked by the GStreamer registry. 2006-01-18 10:38:51 +0000 Edward Hervey 0.10.0.5 release "Doner con patatas" Original commit message from CVS: 0.10.0.5 release "Doner con patatas" * RELEASE: * configure.ac: 2006-01-09 14:17:04 +0000 Christian Schaller * gnonlin.spec.in: update for latest changes Original commit message from CVS: update for latest changes 2006-01-07 18:45:32 +0000 Edward Hervey configure.ac: Remove useless checks Original commit message from CVS: * configure.ac: Remove useless checks * gnl/Makefile.am: * gnl/gnl.c: * gnl/gnl.h: * gnl/gnltimeline.c: * gnl/gnltimeline.h: Remove gnltimeline.[ch] 2006-01-07 18:44:31 +0000 Edward Hervey gnl/gnlcomposition.c: I got mix up with >= vs > and <= vs < Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_handle_message): I got mix up with >= vs > and <= vs < * gnl/gnlobject.c: (translate_message_segment_done): Handle the case where media_stop is GST_CLOCK_TIME_NONE 2005-12-23 15:31:56 +0000 Edward Hervey Add gnonlin.pc for programs depending on gnonlin. Original commit message from CVS: * Makefile.am: * configure.ac: * pkgconfig/Makefile.am: * pkgconfig/gnonlin.pc.in: Add gnonlin.pc for programs depending on gnonlin. 2005-12-21 18:47:19 +0000 Edward Hervey configure.ac: Back to devel cycle Original commit message from CVS: * configure.ac: Back to devel cycle 2005-12-21 18:37:28 +0000 Edward Hervey Releasing 0.10.0.2 beta release Original commit message from CVS: Releasing 0.10.0.2 beta release * RELEASE: * configure.ac: 2005-12-21 17:54:31 +0000 Edward Hervey configure.ac: Ignore po/ for the time being Original commit message from CVS: * configure.ac: Ignore po/ for the time being * gnl/Makefile.am: Add gnltypes.h to the list of headers Simplify the whole process * gnl/gnl.c: * gnl/gnl.h: * gnl/gnlcomposition.h: * gnl/gnlfilesource.h: * gnl/gnlobject.h: * gnl/gnloperation.h: * gnl/gnlsource.h: Use local include headers ( ==> "*.h") 2005-12-20 18:35:46 +0000 Edward Hervey gnl/gnlfilesource.c: Cleanup Original commit message from CVS: * gnl/gnlfilesource.c: Cleanup * gnl/gnlobject.c: (translate_message_segment_done): Handle cases where the GST_SEGMENT_DONE has a format different from GST_FORMAT_TIME. In those situations, bump the segment-done value to object->media_start. 2005-12-20 15:46:56 +0000 Edward Hervey gnl/gnlcomposition.c: Update to use GstBin::handle_message virtual method. Original commit message from CVS: * gnl/gnlcomposition.c: (gnl_composition_class_init), (gnl_composition_init), (gnl_composition_handle_message), (objects_stop_compare): Update to use GstBin::handle_message virtual method. Fix **stupid** bug in stop_compare GCompareFunc. * gnl/gnlobject.c: (gnl_object_class_init), (gnl_object_init), (gnl_object_handle_message): * gnl/gnlobject.h: Update to use GstBin::handle_message virtual method 2005-12-14 14:12:43 +0000 Edward Hervey gnl/gnlcomposition.c: Returning the difference of 2 unsigned int, expecting the result to be negative is... somewhat ... Original commit message from CVS: * gnl/gnlcomposition.c: (objects_start_compare), (objects_stop_compare): Returning the difference of 2 unsigned int, expecting the result to be negative is... somewhat idiotic :) * gnl/gnlcomposition.c: (gnl_composition_class_init), (gnl_composition_add_object): * gnl/gnlfilesource.c: (gnl_filesource_class_init): * gnl/gnlobject.c: (gnl_object_class_init), (gnl_object_covers_func), (translate_message_segment_start), (translate_message_segment_done), (update_values): * gnl/gnloperation.c: (gnl_operation_class_init): * gnl/gnlsource.c: (gnl_source_class_init): GNonLin debugging category is now BOLD BLUE. Updated debugging statements with GST_TIME_FORMAT. 2005-12-12 11:00:30 +0000 Edward Hervey gnl/gnlcomposition.c: Whoops, don't unref too much. Original commit message from CVS: * gnl/gnlcomposition.c: (compare_relink_stack): Whoops, don't unref too much. 2005-12-10 19:35:40 +0000 Edward Hervey gnl/gnlcomposition.c: Implement better event handling and fix memory leaks. Original commit message from CVS: reviewed by: * gnl/gnlcomposition.c: (gnl_composition_init), (gnl_composition_reset), (gnl_composition_sync_handler), (have_to_update_pipeline), (get_new_seek_event), (handle_seek_event), (gnl_composition_event_handler), (gnl_composition_ghost_pad_set_target), (objects_start_compare), (objects_stop_compare), (no_more_pads_object_cb), (compare_relink_stack), (update_pipeline), (gnl_composition_add_object): Implement better event handling and fix memory leaks. * gnl/gnlfilesource.c: (gnl_filesource_prepare): * gnl/gnlobject.c: (gnl_object_to_media_time), (gnl_media_to_object_time), (translate_incoming_seek), (translate_outgoing_seek), (translate_outgoing_new_segment), (internalpad_event_function), (internalpad_query_function), (internalpad_unlink_function), (ghostpad_event_function), (control_internal_pad), (ghostpad_unlink_function), (gnl_object_ghost_pad_set_target): Implement better event handling 2005-12-08 14:46:22 +0000 Edward Hervey gnl/gnlfilesource.c: Do this hack from another thread. Original commit message from CVS: * gnl/gnlfilesource.c: (pad_blocked_cb): Do this hack from another thread.