The DTS typefinder may return a lower probability for frames that start
at non-zero offsets and where there's no second frame sync in the first
buffer. It's fairly unlikely that we'll acidentally identify PCM data
as DTS, so we don't do additional checks for now.
https://bugzilla.gnome.org/show_bug.cgi?id=636234
Use new dts audio typefinder from -base to check if the PCM data
contains a dts stream. This way we recognise more varieties more
reliably and also detect the dts stream if there isn't a frame
sync right at the start of the data.
Fixes#413942.
Handle large, invalid or otherwise unusual chunk sizes.
Verify some chunk sizes to be at least the size they are
expected to be and round up some sizes to even number for
e.g. offset administration, which must also be properly
tracked in push mode.
Standard pull mode loop based SEEK handling fails in push mode,
so convert the SEEK event appropriately and dispatch to upstream.
Also cater for NEWSEGMENT event handling, and properly inform
downstream and application of SEEKABLE capabilities, depending
on scheduling mode and upstream.
Link to properties. Correct titles for examples. Document a few trivial cases. Keep lists in section file and docs/plugins/Makefile.am alphabetically ordered.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (uint64_ceiling_scale_int),
(uint64_ceiling_scale), (gst_wavparse_calculate_duration),
(gst_wavparse_stream_headers):
Fix the scaling code.
Fix parsing of the INFO chunks, we were reading the wrong number of
bytes. Fixes#561580.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
Handle the acid chunk and send tempo as part of tags. Other fields are
interesting too, but need more tag-definitions. Fixes#545433.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
Refactor wavparse. Call _reset() from dispose() and move old code from
dispose into reset. This way we don't leak taglists when we abort
parsing. Fix some comments. Move code for skipping a chunk into extra
function. Replace chunk sizes with a const to ease readability.
Original commit message from CVS:
Based on a patch by:
Victor STINNER <victor dot stinner at haypocalc dot com>
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers):
Set variable to NULL after freeing it to prevent double frees
or make failures by another use of it afterwards more obvious
and fix use of it after the freeing.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_loop):
* gst/wavparse/gstwavparse.c: (gst_wavparse_chain):
* sys/ximage/gstximagesrc.c: (composite_pixel):
Fix 'xyz may be used uninitialized' compiler warnings caused
by broken g_assert_not_reached() macro in GLib-2.15.x (it's
not really nice to abort in any case). Fixes#505745.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_srcpad_event):
Fix seeking in .wav files again (#501775). Some people seem to think
they don't need to test their changes when they're just 'reflowing'
some code.
Original commit message from CVS:
* gst/avi/gstavidemux.c:
* gst/wavparse/gstwavparse.c:
Return the result in _activate_pull(). Don't ref element there.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers),
(gst_wavparse_pad_convert), (gst_wavparse_pad_query),
(gst_wavparse_srcpad_event):
Ref the element when we should, but not when we its not needed. Reflow
the event_handling to not leak the event.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
Return FALSE if we can't handle a query instead of changing the
format. Ignore fact when dealing with mpeg audio.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_loop),
(gst_wavparse_chain):
Don't push EOS from the chain function, the element
driving the pipeline is responsible for this. The bug
this was meant to fix seems to be queue not forwarding
EOS in all cases (see #476514).
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_eos),
(gst_wavparse_loop), (gst_wavparse_chain):
Add EOS logic for the push-based mode too. Fixes#476514.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
* gst/wavparse/gstwavparse.h:
Implement seek-query. Refactor duration calculations. Appropriate use
of uint64_scale_int and uint64_scale. Move repeadedly calculated stuff
out of loops.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers):
Skip LIST chunks before the fmt chunk (fixes#437499). Also fix
streaming mode regression for file from #343837 with 'bext' chunk
before the 'fmt' chunk.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_headers), (gst_wavparse_stream_data):
Only set DISCONT when there actually is a discont or when we just
started.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_headers), (gst_wavparse_add_src_pad),
(gst_wavparse_stream_data):
* gst/wavparse/gstwavparse.h:
Be a bit more clever when dealing with VBR files with FACT tags, we
don't want to timestamp buffers in that case but the estimated BPS can
be used for seeking.
Only send close segment in the streaming thread.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_headers), (gst_wavparse_stream_data):
* gst/wavparse/gstwavparse.h:
Apply DISCONT to buffers.
Only apply timestamp to the first sample after a DISCONT, too many VBR
files cause random jitter in the timestamps. Fixes#433119.
Original commit message from CVS:
Patch by: Vincent Torri <vtorri at univ-evry dot fr>
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers):
Use correct format strings for integer types.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_base_init),
(gst_wavparse_create_sourcepad):
Use gst_riff_create_audio_template_caps () instead of the local caps.
This makes updates of the local caps unecessary whenever libgstriff
gets support for new formats.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (uint64_ceiling_scale_int),
(gst_wavparse_perform_seek), (gst_wavparse_stream_headers),
(gst_wavparse_stream_data):
Relax the audio/mpeg caps again and add FIXME: comment.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (uint64_ceiling_scale_int),
(gst_wavparse_perform_seek), (gst_wavparse_stream_headers),
(gst_wavparse_stream_data):
* gst/wavparse/gstwavparse.h:
More sanity check for the header fields. Fix type for 'rate' header
field.
Original commit message from CVS:
* gst/auparse/gstauparse.c: (gst_au_parse_parse_header):
limit caps to the formats we announce in the template
* gst/wavparse/gstwavparse.c: (uint64_ceiling_scale_int),
(gst_wavparse_perform_seek), (gst_wavparse_stream_headers),
(gst_wavparse_add_src_pad), (gst_wavparse_stream_data):
fix some crashers/asserts when dealing with broken files
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
Support audio/x-raw-float in wav files. This only works with
plugins-base CVS, using an older version doesn't have any
disadvantages though.
Original commit message from CVS:
* gst/wavenc/gstwavenc.c: (gst_wavenc_create_header_buf),
(gst_wavenc_sink_setcaps):
Correctly handle width!=depth input.
* gst/wavparse/gstwavparse.c:
Already export in the caps that width==8 uses unsigned samples and
everything else uses signed samples.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
Add support for wav files containing audio/x-raw-int with random
depths between 1 and 32 bits.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers):
Streaming mode fixes: don't unref buffer we don't own any longer;
remove bogus adapter flush. Fixes#419338.
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:
* gst/wavparse/gstwavparse.c: (gst_wavparse_class_init),
(gst_wavparse_reset), (gst_wavparse_init),
(gst_wavparse_destroy_sourcepad), (gst_wavparse_fmt),
(gst_wavparse_parse_file_header), (gst_wavparse_stream_init),
(gst_wavparse_perform_seek), (gst_wavparse_peek_chunk_info),
(gst_wavparse_stream_headers), (gst_wavparse_parse_stream_init),
(gst_wavparse_add_src_pad), (gst_wavparse_stream_data),
(gst_wavparse_loop), (gst_wavparse_chain),
(gst_wavparse_pad_convert), (gst_wavparse_pad_query),
(gst_wavparse_srcpad_event), (gst_wavparse_change_state),
(plugin_init):
* gst/wavparse/gstwavparse.h:
Update docs.
Use boilerplate.
Various code cleanups.
When the bitrate is not known (bps == 0 or compressed formats) let
downstream element guestimate the duration and position and don't
generate timestamps or durations. Fixes#405213.
Fix EOS and ERROR conditions in chain mode, we just need to forward the
error flowreturn upstream.
Original commit message from CVS:
Based on patch by: Jonathan Matthew <jonathan at kaolin wh9 net>
* gst/wavparse/gstwavparse.c: (gst_wavparse_parse_stream_init),
(gst_wavparse_stream_data):
Fix massive memory leak when operating in streaming mode due to
GST_BUFFER_MALLOCDATA() not being set on newly-created buffers.
Fixes#407057.
Original commit message from CVS:
* gst/id3demux/id3v2frames.c: (parse_picture_frame):
We require a -base more recent than 0.10.9, so it's safe to use
GST_TYPE_TAG_IMAGE_TYPE unconditionally now.
* ext/dv/gstdvdec.c: (gst_dvdec_sink_event):
* ext/jpeg/gstjpegdec.c: (gst_jpeg_dec_sink_event):
Use _newsegment_full() now that we depend on a recent enough core.
* gst/wavparse/gstwavparse.c:
Remove cruft that we don't need any longer now that we depend on
a recent enough -base.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_data):
Fix seeking some more, mostly for speed changes.
Original commit message from CVS:
Patch by: Edward Hervey <edward@fluendo.com>
* configure.ac:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_data):
Send the newsegment event in the streaming thread.
Fixes#347529
Original commit message from CVS:
* gst/wavparse/gstwavparse.c:
#define 'fact' RIFF chunk if we are not compiling against
-base CVS (we don't want to depend on -base CVS for this
one define only, and also not for release order reasons).
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_reset),
(gst_wavparse_other), (gst_wavparse_perform_seek),
(gst_wavparse_get_upstream_size), (gst_wavparse_stream_headers),
(gst_wavparse_add_src_pad), (gst_wavparse_stream_data),
(gst_wavparse_pad_query):
* gst/wavparse/gstwavparse.h:
Use information from 'fact' chunk for length calculation of compressed
samples. Calculate bps if bogus value is found in wav header (embeded
mp2/mp3).
Original commit message from CVS:
Patch by: Mark Nauwelaerts <manauw at skynet be>
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek):
Make sure we don't read beyond the end of the file (#345232).
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_peek_chunk_info),
(gst_wavparse_peek_chunk), (gst_wavparse_stream_headers),
(gst_wavparse_chain):
When operating chain-based, don't make any assumptions about the
chunking of the incoming data and make streaming work on days other
than the second Thursday after a full moon. Also fix up debug
messages here and there and make use of the most excellent new
gst_pad_query_peer_duration() utility function.
Skip any 'bext' chunks in front of the 'fmt ' chunk. Fixes#343837.
* gst/wavparse/gstwavparse.h:
Remove trailing comma after last enum value, some compilers don't
like that.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_fmt),
(gst_wavparse_perform_seek), (gst_wavparse_stream_headers):
Fix use of uninitialised values if we're NOT seeking in ready.
Fix typos.
Original commit message from CVS:
* ext\jpeg\smokecodec.c:
use of GST_DEBUG instead of DEBUG(a...) for WIN32
* ext\speex\gstspeexenc.c: (gst_speexenc_set_header_on_caps):
move first instruction after all variables declarations
* gst\alpha\gstalpha.c:
* gst\effectv\gstshagadelic.c:
* gst\smpte\paint.c:
* gst\videofilter\gstvideobalance.c:
define M_PI if it's not defined (it's not defined on WIN32)
* gst\cutter\gstcutter.c: (gst_cutter_chain):
* gst\id3demux\id3v2frames.c: (parse_relative_volume_adjustment_two):
* gst\level\gstlevel.c: (gst_level_set_property), (gst_level_transform_ip):
* gst\matroska\matroska-demux.c: (gst_matroska_demux_parse_info),
(gst_matroska_demux_video_caps):
* gst\matroska\matroska-mux.c: (gst_matroska_mux_start), (gst_matroska_mux_finish):
* gst\wavparse\gstwavparse.c: (gst_wavparse_stream_data):
use gst_guint64_to_gdouble for conversions
* gst\goom\filters.c: (setPixelRGB_):
fix a debug which was using undefined variable
* gst\level\gstlevel.c: (gst_level_set_caps), (gst_level_transform_ip):
* gst\matroska\ebml-read.c: (gst_ebml_read_sint):
replace LL suffix with L suffix (LL isn't supported by MSVC6.0)
* win32/vs6:
add vs6 projects files for most of plugins-good
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_get_upstream_size),
(gst_wavparse_stream_headers), (gst_wavparse_stream_data):
Don't try to read beyond the end of the file just because
the header claims a bigger size (like with truncated files).
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_headers), (gst_wavparse_add_src_pad),
(gst_wavparse_stream_data), (gst_wavparse_loop):
* gst/wavparse/gstwavparse.h:
Delay source pad creation until we have the first chunk of
media data, so the we can examine the data and adjust the
caps accordingly if required. This makes playback of .wav
files with DTS-declared-as-PCM content work (#313266).
Original commit message from CVS:
Patch by: Tommi Myöhänen <ext-tommi dot myohanen at nokia dot com>
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek):
Fix block alignment calculation. Alignment should be done before
adding the byte offset where the data starts (#335231).
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_reset),
(gst_wavparse_change_state):
Implement seek in READY (re-fixes #327658)
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_perform_seek),
(gst_wavparse_stream_headers), (gst_wavparse_stream_data),
(gst_wavparse_pad_convert), (gst_wavparse_srcpad_event),
(gst_wavparse_sink_activate), (gst_wavparse_sink_activate_pull):
Use DEBUG_OBJECT more.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_stream_headers):
Pass extra_data to gst_riff_create_audio_caps(), so that
WAVEFORMATEX stuff works. Post audio codec name and post
it as taglist on the bus. Allow up to 8 channesl for raw
PCM in the source pad template caps.
Original commit message from CVS:
* gst/wavparse/gstwavparse.c: (gst_wavparse_pad_convert):
Fix conversion from TIME to BYTES format (fixes#326864;
patch by: Tommi Myöhänen <ext-tommi dot myohanen at nokia dot com>)
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_reset),
(gst_avi_demux_handle_src_event), (gst_avi_demux_parse_stream),
(gst_avi_demux_parse_odml), (gst_avi_demux_peek_tag),
(gst_avi_demux_next_data_buffer), (gst_avi_demux_stream_scan),
(gst_avi_demux_stream_header), (gst_avi_demux_loop):
Construct index for indexless files.
Make sure pad/buffers are correctly reset to NULL once we don't need
them anymore, else we get lovely segfaults/assertions.
* gst/wavparse/gstwavparse.c:
Yes, you can have 96KHz audio and wma in wav :(
