Original commit message from CVS:
* gst/id3demux/gstid3demux.c: (gst_id3demux_add_srcpad),
(gst_id3demux_sink_activate):
Don't leak caps: make gst_id3demux_add_srcpad() not take ownership of the
caps passed to it (previouslly one code path assumes it takes ownership
while another one assumes it doesn't).
* configure.ac:
* tests/files/Makefile.am:
* tests/files/id3-407349-1.tag:
* tests/files/id3-407349-2.tag:
Add directory where data for unit tests can be stored.
* tests/Makefile.am:
* tests/check/Makefile.am:
* tests/check/elements/.cvsignore:
* tests/check/elements/id3demux.c: (pad_added_cb), (error_cb),
(read_tags_from_file), (run_check_for_file),
(check_date_1977_06_23), (GST_START_TEST), (id3demux_suite):
Add unit test for id3demux, and in particular for bug #407349. Only
testing pull-mode for now; push mode doesn't work yet because the test
files are smaller than ID3_TYPE_FIND_MIN_SIZE.
Original commit message from CVS:
* gst/id3demux/id3tags.c: (id3demux_id3v2_frames_to_tag_list):
* gst/id3demux/id3tags.h:
* gst/id3demux/id3v2frames.c: (id3demux_id3v2_parse_frame),
(parse_obsolete_tdat_frame):
Do not convert obsolete TDA/TDAT frames to TDRC frames, otherwise
the four-digit number will be interpreted as a year, whereas it is
month and day in DDMM format. Instead, parse TDAT frames and fix up
the date in the GST_TAG_DATE tag later if we also extracted a year.
Fixes#407349.
Original commit message from CVS:
* ext/gconf/gstswitchsink.c: (gst_switch_sink_dispose),
(gst_switch_commit_new_kid):
Fix up the dispose logic so it doesn't leak, and fix setting of
the child state so that we don't set a child to our current state
just as we are changing it to something else.
Original commit message from CVS:
* gst/spectrum/gstspectrum.c: (gst_spectrum_class_init),
(gst_spectrum_init), (gst_spectrum_set_property),
(gst_spectrum_transform_ip):
Fix and cleanup default property values.
Add FIXMEs for stuff that looks rather wrong.
Original commit message from CVS:
* ext/gconf/gstswitchsink.c: (gst_switch_sink_dispose),
(gst_switch_commit_new_kid):
Fix up the reference counting of the child elements.
Original commit message from CVS:
* ext/gconf/gstswitchsink.c: (gst_switch_sink_reset),
(gst_switch_commit_new_kid), (gst_switch_sink_set_child):
Install fakesink in NULL by fixing some broken logic. This obviates
the need to manually set _IS_SINK.
Add some comments and remove a little cruft while I'm at it.
Original commit message from CVS:
* ext/gconf/gstswitchsink.c: (gst_switch_sink_reset):
Mark us as a sink when we have no fakesink in NULL. Fixes#414887.
Original commit message from CVS:
* gst/spectrum/demo-audiotest.c: (message_handler):
* gst/spectrum/demo-osssrc.c: (message_handler):
Remove two obsolete and confusing comments.
Original commit message from CVS:
* tests/check/Makefile.am:
Gah! Also disable gconfvideosink from the tests, otherwise
it will instantiate autovideosink, and dfbvideosink and
leak on the buildbots.
Original commit message from CVS:
* ext/cdio/gstcdiocddasrc.c: (gst_cdio_cdda_src_open),
(gst_cdio_cdda_src_finalize):
Make sure we always destroy our libcdio handle.
Original commit message from CVS:
* tests/check/Makefile.am:
Disable autovideosink so the buildbots don't barf over memory
leaked in the directfb sink.
Original commit message from CVS:
* gst/multipart/multipartdemux.c: (gst_multipart_demux_init),
(gst_multipart_find_pad_by_mime):
Use gst_pad_new_from_static_template instead of
static_pad_template_get+pad_new.
Original commit message from CVS:
* ext/dv/gstdvdec.c: (gst_dvdec_init):
Use gst_pad_new_from_static_template instead of
static_pad_template_get+pad_new.
Original commit message from CVS:
Patch by: Loïc Minier <lool+gnome at via ecp fr>
* ext/libcaca/Makefile.am:
* gst/debug/Makefile.am:
Don't mix tabs and spaces (#414168).
Original commit message from CVS:
* ext/cdio/gstcdiocddasrc.c: (gst_cdio_cdda_src_probe_devices),
(gst_cdio_cdda_src_read_sector), (gst_cdio_cdda_src_open),
(gst_cdio_cdda_src_finalize):
Small code cleanups.
Don't use pad_alloc as the base class cannot deal with the error codes.
Original commit message from CVS:
Patch by: René Stadler <mail@renestadler.de>
* gst/wavparse/gstwavparse.c: (uint64_ceiling_scale_int),
(gst_wavparse_perform_seek), (gst_wavparse_stream_headers),
(gst_wavparse_stream_data):
Handle rounding better to not drop last sample frame. Fixes#356692
Original commit message from CVS:
* tests/check/Makefile.am:
Disable cacasink from the states check too - it also calls exit(1)
on us when it can't find a terminal to talk to.
Original commit message from CVS:
* ext/hal/gsthalaudiosink.c: (do_toggle_element):
* ext/hal/gsthalaudiosrc.c: (do_toggle_element):
Having NULL as UDI previously selected the default sink/src. Change
this back but mention it in the debug output.
* ext/hal/hal.c: (gst_hal_get_alsa_element),
(gst_hal_get_oss_element), (gst_hal_get_string),
(gst_hal_render_bin_from_udi), (gst_hal_get_audio_sink),
(gst_hal_get_audio_src):
* ext/hal/hal.h:
Refactor a bit, check all error conditions, greatly improve debugging
and fix some possible memory leaks. Also implement OSS support
and allow specifying an UDI that points to a real device. For this the
child device which supports ALSA (preferred) or OSS is used.
As a side effect this makes it impossible now to get a alsasink in
halaudiosrc and a alsasrc in halaudiosink.
Original commit message from CVS:
* gst/rtsp/gstrtspsrc.c: (find_stream_by_channel),
(find_stream_by_udpsrc), (gst_rtspsrc_handle_message):
Errors from the udp sources are not fatal unless all of them are in
error.
Original commit message from CVS:
* tests/check/Makefile.am:
Disable aasink in the states test. I suspect this is the element that
is calling exit(1) when it can't proceed.
Original commit message from CVS:
* tests/check/Makefile.am:
Draw plugins in from the build tree sys/ dir, rather than picking
up the already installed versions.
Original commit message from CVS:
2007-03-01 Zaheer Abbas Merali <zaheerabbas at merali dot org>
* sys/ximage/gstximagesrc.c: (gst_ximage_src_open_display):
Error out correctly when getting xcontext fails.
Original commit message from CVS:
* gst/rtsp/gstrtpdec.c: (gst_rtpdec_change_state):
Make state change to PAUSED NO_PREROLL because that's what it will be in
the future and rtspsrc relies on it.
* gst/rtsp/gstrtspsrc.c: (gst_rtspsrc_stream_configure_transport),
(gst_rtspsrc_change_state):
Don't error out when we don't get an error from the state change
function.
Original commit message from CVS:
* ext/hal/gsthalaudiosink.c: (do_toggle_element):
* ext/hal/gsthalaudiosrc.c: (do_toggle_element):
Check if the device UDI is set before trying to query HAL
about it and give a useful error message if it wasn't set.
* ext/hal/hal.c: (gst_hal_get_string):
Don't query HAL for NULL UDIs. Passing NULL as UDI to HAL
gives an assertion failure in D-Bus when running with
DBUS_FATAL_WARNINGS=1.
