=== release 1.5.91 === 2015-09-18 Sebastian Dröge * configure.ac: releasing 1.5.91 2015-09-18 11:49:03 +0200 Sebastian Dröge * po/zh_CN.po: po: Update translations 2015-09-15 10:56:40 +0900 Vineeth TM * libs/gst/check/gstcheck.c: * plugins/elements/gstdownloadbuffer.c: * tests/benchmarks/gstbufferstress.c: * tests/benchmarks/gstclockstress.c: * tests/benchmarks/gstpollstress.c: downloadbuffer, benchmarks: fix error leaks in failure code paths https://bugzilla.gnome.org/show_bug.cgi?id=755019 2015-09-15 10:52:55 +0900 Vineeth TM * libs/gst/check/gstcheck.c: check: Fix indentation https://bugzilla.gnome.org/show_bug.cgi?id=755019 2015-09-15 18:05:11 +0100 Tim-Philipp Müller * gst/gstbufferpool.c: bufferpool: sprinkle some allow-none and out annotations for g-i 2015-09-14 11:01:11 +0900 Vineeth TM * gst/gstbin.c: bin: fix typo in log message when threadpool alloc fails https://bugzilla.gnome.org/show_bug.cgi?id=754975 2015-09-11 17:58:48 +0300 Igor Rondarev * configure.ac: * gst/Makefile.am: configure: Check for socketpair() in -lsocket too On e.g. QNX it is in an external library, not libc. https://bugzilla.gnome.org/show_bug.cgi?id=754875 2015-09-09 13:10:04 +0530 Arun Raghavan * docs/gst/gstreamer-sections.txt: Revert "docs: Make sure gst_debug_bin_to_dot_data() is documented" This reverts commit 0dffeb03018d12be522c2d97aaaf8102153bd7c0. The commit is erroneous and documents the function twice. 2015-07-23 12:18:51 +0530 Arun Raghavan * docs/gst/gstreamer-sections.txt: docs: Make sure gst_debug_bin_to_dot_data() is documented Thanks to Nirbheek Chauhan for pointing this out. 2015-08-05 10:07:50 +0200 Stian Selnes * libs/gst/check/gstharness.c: * tests/check/libs/gstharness.c: harness: Fix race for gst_harness_element_ref In order for gst_harness_new_full to be MT-safe the increase and decrease of HARNESS_REF must be MT-safe. This allows for creating multiple harnesses from different threads wrapping the same element. https://bugzilla.gnome.org/show_bug.cgi?id=754661 2015-08-05 09:59:39 +0200 Stian Selnes * libs/gst/check/gstharness.c: harness: Allow-none for custom stress init func It should be allowed to not have a function to initialize the user data since it's often not necessary; it may already be initialized. https://bugzilla.gnome.org/show_bug.cgi?id=754661 2015-09-06 09:58:09 +0100 Tim-Philipp Müller * docs/plugins/gstreamer-plugins.signals: docs: remove signal that no longer exists from docs 2015-09-05 11:20:49 +0100 Tim-Philipp Müller * po/af.po: * po/az.po: * po/be.po: * po/bg.po: * po/ca.po: * po/cs.po: * po/da.po: * po/de.po: * po/el.po: * po/en_GB.po: * po/eo.po: * po/es.po: * po/eu.po: * po/fi.po: * po/fr.po: * po/gl.po: * po/hr.po: * po/hu.po: * po/id.po: * po/it.po: * po/ja.po: * po/lt.po: * po/nb.po: * po/nl.po: * po/pl.po: * po/pt_BR.po: * po/ro.po: * po/ru.po: * po/rw.po: * po/sk.po: * po/sl.po: * po/sq.po: * po/sr.po: * po/sv.po: * po/tr.po: * po/uk.po: * po/vi.po: * po/zh_CN.po: * po/zh_TW.po: po: update for translated string changes 2015-09-05 11:18:27 +0100 Tim-Philipp Müller * tools/gst-launch.c: tools: gst-launch: fix --exclude command line option This has not worked (as in: crashed) since 2005, so perhaps it should just be removed instead. 2015-08-31 12:07:10 +0100 Tim-Philipp Müller * plugins/elements/gstqueue2.c: Revert "queue2: Process SEEKING query" This caused problems with oggdemux when queue2 was operating in queue mode and the souphttpsrc upstream is not seekable because the server doesn't support range requests. It would then still claim seekability and then things go wrong from there. This reverts commit 7b0b93dafe4ac547552cdb66ade5d8aa0405e7b4. https://bugzilla.gnome.org/show_bug.cgi?id=753887 2015-08-29 20:14:44 +0200 Havard Graff * libs/gst/check/gstharness.c: harness: misc bugfixes 1. Get a list of pad templates from the element class, not the factory. This allows us to interact with test-elements that does not have a factory. 2. Use the pad_template_caps in caps-queries when caps is not set explicitly on the pad. Not doing so is simply wrong, and prohibits interactions with special templates used for testing. https://bugzilla.gnome.org/show_bug.cgi?id=754193 2015-08-26 09:29:05 -0300 Thiago Santos * tests/check/gst/gstevent.c: tests: event: fix build failure gst/gstevent.c:250:5: error: ‘for’ loop initial declarations are only allowed in C99 or C11 mode 2015-08-24 21:04:37 +0300 Vivia Nikolaidou * gst/gstbin.c: * tests/check/gst/gstbin.c: bin: Make sure we don't add/remove a bin to/from itself Doing so would deadlock from trying to acquire the object lock twice https://bugzilla.gnome.org/show_bug.cgi?id=754036 2015-08-21 14:28:48 -0700 Nicolas Dufresne * libs/gst/base/gstbasetransform.c: basetransform: Reconfigure before propose_allocation There exist cases where a reconfigure event was propagated from downstream, but caps didn't change. In this case, we would reconfigure only when the next buffer arrives. The problem is that due to the allocation query being cached, the return query parameters endup outdated. In this patch we refactor the reconfigurating code into a function, and along with reconfiguring when a new buffer comes in, we also reconfigure when a query allocation arrives. https://bugzilla.gnome.org/show_bug.cgi?id=753850 2015-08-07 15:39:59 -0400 Nicolas Dufresne * tests/check/libs/basesrc.c: basesrc-test: Fix race testing segment update As this test is using a short sleep (GST_USECOND, which is 10ms in microsecond), sometimes that EOS event is received before the loop in basesrc have run _do_seek() and pushed the update segment. To solve this issue, we wait for the initial segment (and flush it) then we wait for the second segment before sending EOS. https://bugzilla.gnome.org/show_bug.cgi?id=753365 2015-08-19 11:46:07 +0200 Thibault Saunier * scripts/gst-uninstalled: bin: Add NLE to GST_PLUGIN_PATH