Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (h264_video_type_find):
Don't do lots of 4-byte peeks, but use the 'new' data scan helper
for this instead; don't check if we've found enough markers after
each and every step, it's enough to do that only if we've actually
found a new marker.
Embed a G_UNLIKELY into the IS_MPEG_HEADER macro.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c:
(DATA_SCAN_CTX_CHUNK_SIZE), (DataScanCtx), (data_scan_ctx_advance),
(data_scan_ctx_ensure_data), (GST_MPEGVID_TYPEFIND_TRY_SYNC),
(mpeg_video_stream_type_find):
Move scan helper thingy to the beginning of the file so we can use
it in other typefind functions. Rename it to something more
generic. Also improve handling of things towards the end of the
typefind data: peek as much as we can if we know the size of the
data, rather than just min_size.
Original commit message from CVS:
Based on a patch by:
Björn Benderius <bjoern dot benderius at axis dot com>
* gst/ffmpegcolorspace/avcodec.h:
* gst/ffmpegcolorspace/gstffmpegcodecmap.c:
(gst_ffmpeg_pixfmt_to_caps), (gst_ffmpeg_caps_to_pixfmt),
(gst_ffmpegcsp_avpicture_fill):
* gst/ffmpegcolorspace/imgconvert.c: (nv12_to_nv21):
* gst/ffmpegcolorspace/imgconvert_template.h:
Add conversions from/to NV12 and NV21 and conversions between those
two formats. Fixes bug #532166.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (h264_video_type_find):
Abort the h264 typefinding as soon as _peek() doesn't return anything,
which happens for example with files smaller than 128kb.
Original commit message from CVS:
Patch by: Sjoerd Simons <sjoerd at luon dot net>
* gst/audioresample/gstaudioresample.c: (gst_audioresample_init):
Let audioresample use the buffer allocation of basetransform instead
of it's own stuff.
* tests/check/elements/audioresample.c: (alloc_only_48000),
(GST_START_TEST), (audioresample_suite):
Add unit test for the recent basetransform bugfix, where upstream
changes caps to something that can't be passed through anymore.
Original commit message from CVS:
* gst/volume/gstvolume.c: (volume_transform_ip):
Return NOT_NEGOTIATED if we didn't set a process function yet for some
reason instead of crashing later. Might fix bug #509125.
Original commit message from CVS:
Based on a patch by: Tim-Philipp Müller <tim.muller at collabora co uk>
* gst/audioconvert/audioconvert.c: (audio_convert_prepare_context):
* gst/audioconvert/audioconvert.h:
* gst/audioconvert/gstaudioconvert.c:
(gst_audio_convert_parse_caps),
(structure_has_fixed_channel_positions),
(gst_audio_convert_transform_caps):
* gst/audioconvert/gstchannelmix.c: (gst_channel_mix_fill_matrix):
Add support for more than 8 channels and NONE channel layouts. For
more than 8 channels no channel conversion is supported yet, only
format conversions are supported. Fixes bug #398033.
* tests/check/elements/audioconvert.c: (verify_convert),
(GST_START_TEST), (audioconvert_suite):
Add some unit tests by Tim for checking the NONE channel layouts
and more than 8 channels and add some more unit tests for channel
conversions.
Original commit message from CVS:
* gst/playback/gstdecodebin2.c: (connect_pad):
When autoplugging fails, set the element back to NULL before
unreffing it.
Original commit message from CVS:
* gst/subparse/samiparse.c: (handle_start_sync),
(end_sami_element), (characters_sami):
Remove trailing, leading and double whitespaces.
Correctly timestamp buffers and output the last buffer too.
* tests/check/elements/subparse.c: (GST_START_TEST),
(subparse_suite):
Add a simple unit test for SAMI parsing.
Original commit message from CVS:
Patch by: Young-Ho Cha <ganadist at chollian dot net>
* gst/subparse/samiparse.c: (handle_start_sync),
(start_sami_element), (end_sami_element), (characters_sami),
(sami_context_reset):
Only output characters inside the "sync" elements. There could be
other elements like "style" that have some content but should
not be printed. Fixes bug #467911.
Original commit message from CVS:
* gst/playback/gstplaybasebin.c: (set_audio_mute),
(set_active_source):
* gst/playback/gstplaybasebin.h:
* gst/playback/gstplaybin.c: (gst_play_bin_class_init),
(playbin_set_audio_mute):
Allow setting -1 as current-audio to mute the current audio stream,
similar to what is done for subtitles. Fixes bug #342294.
Original commit message from CVS:
* gst/subparse/gstsubparse.c: (handle_buffer),
(gst_sub_parse_sink_event):
* gst/subparse/tmplayerparse.c: (tmplayer_process_buffer),
(tmplayer_parse_line):
Fix parsing of tmplayer subtitle variant where every single line contains
text and there isn't an empty line after each line to determine the
duration (#530962). Improve EOS handling for tmplayer subtitles a bit by
making sure that we push out the last line of text without a duration if
there's still text left in the buffer at the end.
Original commit message from CVS:
* gst/subparse/gstsubparse.c: (feed_textbuf):
Fix detection of discontinuities based on the buffer offset (doesn't work
so well if no buffer offset is set) and also check for the DISCONT buffer
flag. This keeps the parser state from being reset after each buffer in
the unit test.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (mpeg_video_stream_type_find):
Further fine-tuning: don't absolutely require sequence or GOP headers
(as introduced in the previous commit), but adjust the typefind
probabilities returned accordingly if we don't see them. Also make sure
picture header and first slice are somewhat close to each other (which
is not perfect but still better than requiring a fixed offset or having
no limit at all).
Original commit message from CVS:
* gst/playback/test4.c:
* gst/playback/test5.c:
* gst/playback/test6.c:
* gst/playback/test7.c:
Also include config.h when relying on defines from it. Fixes the
build. Its been a please to serve :)
Original commit message from CVS:
* gst/videotestsrc/videotestsrc.c (paint_setup_NV12),
(paint_setup_NV21), (paint_hline_NV12_NV21):
Add support for NV12 and NV21 in videotestsrc
Original commit message from CVS:
* gst/videoscale/gstvideoscale.c:
* gst/videoscale/vs_4tap.c: (vs_image_scale_4tap_Y):
* gst/videoscale/vs_image.c: (vs_image_scale_nearest_RGBA),
(vs_image_scale_linear_RGBA), (vs_image_scale_nearest_RGB),
(vs_image_scale_linear_RGB), (vs_image_scale_nearest_YUYV),
(vs_image_scale_linear_YUYV), (vs_image_scale_nearest_UYVY),
(vs_image_scale_linear_UYVY), (vs_image_scale_nearest_Y),
(vs_image_scale_linear_Y), (vs_image_scale_nearest_RGB565),
(vs_image_scale_linear_RGB565), (vs_image_scale_nearest_RGB555),
(vs_image_scale_linear_RGB555):
Support 1x1 images as input and output as for example the BBC HQ new
streams have 1x1 GIFs in the playlists for some reason.
Original commit message from CVS:
* gst/playback/gstdecodebin.c: (free_pad_probe_for_element),
(try_to_link_1):
If we can't activate one of the decoders we plugged in (such as,
say, musepackdec) for some reason (it might not support push mode,
for example), remove any pad probes that close_pad_link() might
have set up. This makes sure we later don't try to remove a probe
for a pad that doesn't exist any longer, and avoids nast warnings
and probably other things too.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c:
(mpeg_video_stream_ctx_ensure_data), (mpeg_video_stream_type_find),
(plugin_init):
Rework mpeg video stream typefinding a bit more: make sure sequence,
GOP, picture and slice headers appear in the order they should and
that we've in fact at least had one of each; fix picture header
detection; decouple picture and slice header check - don't assume
they're at a fixed offset, there may be extra data in between. Also,
announce varying degrees of probability depending on what we found
exactly (multiple pictures, at least one picture, just sequence and
GOP headers). Finally, in _ensure_data(), take into account that we
might be typefinding smaller amounts of data, such as the first
buffer of a stream, so fall back to the minimum size needed as long
as that's available, instead of erroring out if there's less than
2kB of data. Fixes#526173. Conveniently also doesn't recognise the
fuzzed file from #399342 as valid.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (MpegVideoStreamCtx),
(mpeg_video_stream_ctx_advance), (mpeg_video_stream_ctx_ensure_data),
(mpeg_video_stream_type_find):
Refactor a bit: use context structure to track parsing offset and size of
available data and make the code a bit clearer. Fixes bad memory access
in #356937.
Original commit message from CVS:
* gst/playback/test4.c:
* gst/playback/test5.c:
* gst/playback/test6.c:
* gst/tcp/gstmultifdsink.c:
Include stdlib.h and unistd.h only if the appropriate HAVE_*_H macro
is defined.
Original commit message from CVS:
* gst/typefind/gsttypefindfunctions.c: (musepack_type_find),
(plugin_init):
Add "mpp" and "mp+" as possible extensions for MusePack files.
Add typefinding for MusePack StreamVersion 8 files and include the
stream version in the caps.
Original commit message from CVS:
* docs/design/draft-keyframe-force.txt:
Fix typo.
* gst/playback/gstqueue2.c: (update_buffering),
(gst_queue_handle_src_query):
Set buffering mode in the messages.
Set buffering percent in the query.
* tests/examples/seek/seek.c: (update_fill), (msg_state_changed),
(do_stream_buffering), (do_download_buffering), (msg_buffering):
Do some more fancy things based on the buffering method in use.
Original commit message from CVS:
* gst/playback/gstqueue2.c: (update_buffering),
(gst_queue_close_temp_location_file), (gst_queue_handle_src_query),
(gst_queue_src_checkgetrange_function):
Include extra buffering stats in the buffering message.
Implement BUFFERING query.
* gst/playback/gsturidecodebin.c: (do_async_start),
(do_async_done), (type_found), (setup_streaming), (setup_source),
(gst_uri_decode_bin_change_state):
Only add decodebin2 when the type is found in streaming mode.
Make uridecodebin async to PAUSED even when we don't have decodebin2
added yet.
Original commit message from CVS:
* gst-libs/gst/audio/gstbaseaudiosink.c:
(gst_base_audio_sink_class_init):
* gst-libs/gst/audio/gstbaseaudiosrc.c:
(gst_base_audio_src_class_init):
* gst/playback/gstplayback.c: (plugin_init):
* gst/volume/gstvolume.c: (plugin_init):
Work around missing bits of thread-safety on older GLibs some
more to avoid assertions when starting up multiple playbin
objects concurrently (see #512382).
Original commit message from CVS:
* gst/playback/gstdecodebin2.c: (gst_decode_bin_class_init),
(gst_decode_bin_init), (gst_decode_bin_dispose),
(gst_decode_bin_set_sink_caps), (gst_decode_bin_get_sink_caps),
(gst_decode_bin_set_property), (gst_decode_bin_get_property),
(analyze_new_pad), (connect_pad), (expose_pad),
(gst_decode_group_new), (gst_decode_group_control_demuxer_pad),
(gst_decode_group_expose), (gst_decode_group_free),
(do_async_start), (do_async_done), (gst_decode_bin_change_state):
Remove fakesink hack, we can now implement this more elegantly.
Added property to bypass typefinding.
Removed underrun callback and demuxer pad probe, we now use the srcpad
probe to expose groups.
API::sink-caps property
* gst/playback/gstplaybin2.c: (no_more_pads_cb):
Guard against multiple emissions of the no_more_pads signal, which
happens when we are dealing with chained oggs.
* gst/playback/gsturidecodebin.c: (remove_decoders),
(make_decoder), (type_found), (setup_streaming), (source_new_pad),
(setup_source):
For streams, use our own typefind element and plug our queue after it.
We will need this to determine the type of buffering to use for the
queue soon.
Original commit message from CVS:
* gst/playback/gstqueue2.c: (update_out_rates),
(gst_queue_open_temp_location_file),
(gst_queue_close_temp_location_file), (gst_queue_handle_src_event),
(gst_queue_handle_src_query), (gst_queue_set_property):
Update the estimated input data when we push out a buffer.
Add some debug info about the temp file.
Only forward src events when we are not using a temp file.
Don't block the duration query, we need to find something better.
Don't leak the temp filename.
Original commit message from CVS:
* gst/playback/gstqueue2.c: (gst_queue_is_filled):
The queue is never filled when there are no buffers in the queue at all.
Fixes#523993.
Original commit message from CVS:
* gst/playback/gstplaybin2.c: (gst_play_bin_class_init),
(init_group), (free_group), (gst_play_bin_init),
(gst_play_bin_finalize), (gst_play_bin_set_uri),
(gst_play_bin_set_suburi), (gst_play_bin_get_video_tags),
(gst_play_bin_get_audio_tags), (gst_play_bin_get_text_tags),
(gst_play_bin_set_current_video_stream),
(gst_play_bin_set_current_audio_stream),
(gst_play_bin_set_current_text_stream),
(gst_play_bin_set_encoding), (gst_play_bin_set_property),
(gst_play_bin_get_property), (pad_added_cb), (pad_removed_cb),
(no_more_pads_cb), (perform_eos), (autoplug_select_cb),
(activate_group), (deactivate_group), (setup_next_source),
(save_current_group), (gst_play_bin_change_state):
Update some docs.
Add new locks and conds to protect pipeline creation and group
switching.
Implement the sub-uri property.
Keep track of pending uridecodebin creation and configure the output
pipeline after all streams are configured.
Propagate subtitle encoding to the uridecodebins.
Implement getting the video/audio/visualisation elements.
Use input-selector for stream switching.
If we are asked to do visualisation, prefer to autoplug raw sinks
instead of sinks that accept encoded data.
Original commit message from CVS:
* gst/playback/gstplaysink.c: (gst_play_sink_class_init),
(gst_play_sink_init), (gst_play_sink_dispose),
(gst_play_sink_set_video_sink), (gst_play_sink_get_video_sink),
(gst_play_sink_set_audio_sink), (gst_play_sink_get_audio_sink),
(gst_play_sink_vis_unblocked), (gst_play_sink_vis_blocked),
(gst_play_sink_set_vis_plugin), (gst_play_sink_get_vis_plugin),
(gst_play_sink_set_volume), (gst_play_sink_get_volume),
(gst_play_sink_set_mute), (gen_video_chain), (gen_text_chain),
(gen_audio_chain), (gen_vis_chain), (gst_play_sink_reconfigure),
(gst_play_sink_set_font_desc), (gst_play_sink_get_font_desc),
(gst_play_sink_send_event_to_sink), (gst_play_sink_change_state):
* gst/playback/gstplaysink.h:
Add methods to get audio/video/vis elements.
Add methods to set the font description for the overlay.
Remove properties, we're using this element with its methods only.
Add support for subtitles.
Rearrange the locking a bit to not use the object lock for protecting
the pipeline construction.
Try to use the volume and mute property on the sink when its available.
Implement the mute option with volume when the sink does not have a mute
property.
Only add volume element when the sink has no volume property.
Only do visualisations with raw audio pads.
Original commit message from CVS:
* gst/playback/gsturidecodebin.c:
(gst_uri_decode_bin_autoplug_factories),
(gst_uri_decode_bin_class_init), (gst_uri_decode_bin_init),
(gst_uri_decode_bin_finalize), (gst_uri_decode_bin_set_encoding),
(gst_uri_decode_bin_set_property),
(gst_uri_decode_bin_get_property), (no_more_pads_full),
(new_decoded_pad_cb), (gen_source_element), (remove_decoders),
(proxy_autoplug_factories_signal), (make_decoder),
(source_new_pad), (setup_source):
Add a readonly source property and notify.
Add new lock for protecting the construction of the pipeline.
Keep track of the decodebins we plugged.
Correctly proxy the autoplug signal so that it actually continues.
Proxy subtitle-encoding to the decodebins.
Original commit message from CVS:
* gst/playback/gstdecodebin2.c: (gst_decode_bin_dispose),
(gst_decode_bin_set_caps), (gst_decode_bin_get_caps),
(gst_decode_bin_set_subs_encoding),
(gst_decode_bin_get_subs_encoding),
(gst_decode_bin_autoplug_factories), (connect_pad), (are_raw_caps),
(deactivate_free_recursive):
Protect caps property with the object lock.
Protect encoding property with the object lock.
Keep list of elements we added that have the subtitle-encoding property.
Distribute the subtitle-encoding to all of the elements when it
changes.
Original commit message from CVS:
* gst/volume/gstvolume.c: (gst_volume_interface_supported),
(gst_volume_base_init), (gst_volume_class_init),
(volume_process_double), (volume_process_float),
(volume_transform_ip), (plugin_init):
memset buffers to zero if we get a GAP buffer. We usually see a
buffer as one unit so let's handle it as one and don't care about
volume changes while processing one buffer.
Also clean up some stuff a bit.
Original commit message from CVS:
* gst/audioconvert/gstaudioconvert.c: (gst_audio_convert_init),
(gst_audio_convert_create_silence_buffer),
(gst_audio_convert_transform):
Make audioconvert GAP-aware by outputting silence buffers when the
input has the GAP flag set. This is up to 8x faster.
Based on a patch by Stefan Kost. Fixes bug #517813.
Original commit message from CVS:
* gst/volume/gstvolume.c: (volume_process_double):
Use oil_scalarmultiply_f64_ns() for double processing when it's
available at compile time.
Original commit message from CVS:
* gst/playback/gststreamselector.c: (gst_selector_pad_event),
(gst_selector_pad_chain):
* gst/playback/gststreamselector.h:
Revert change that caused regression until a real fix is found.
Fixes#522203.
