Push mode seeking uses same index data as pull mode, and stores
offset to data in chunk, whereas push mode operates in chunks,
and as such needs offset consistently corresponding to chunk headers.
Also fix determining best matching stream for incoming newsegment event,
as well as setting some stream state accordingly.
The audio packets in AVI are generally muxed ~0.5s before the
corresponding video packet. This changes causes downstream to only
receive packets with roughly corresponding timestamps.
Make sure we reset the demuxer correctly wrt parsing the index.
Don't leak pending seek events.
Rename some methods to reflect what they do and to avoid confusion with similar
method names.
Try to make the seeking threadsafe by protecting the setup code with a lock.
Make sure we post errors when a seek fails.
When we have not parsed any indexes yet, we don't know the length of the streams
and we must take the length given in the avih as a fallback.
Avoid some typechecking.
When we receive an UNEXPECTED from a stream, move to the next stream and only go
EOS when all streams are EOS. When selecting a stream to push, ignore streams
that went EOS.
Fixes#607949
When we are in push mode, we can encounter RIFF and idx tags in the data chunk
when we are dealing with ODML files. In these cases, simply skip the chunks and
continue streaming instead of going EOS.
Greatly simplify the IDIT chunk handling by using sscanf
instead of 'manually' parsing. Also replaces strncasecmp and
is_alpha/is_digit with glib versions.
Set the current_entry to 0 (instead of -1) in push mode so that we correctly
calculate the current frame number and timestamp.
Add some more debug info and fic the duration debug.
... but disable this for now. Although it basically works fine,
user experience might be shaky (depending on taste), since there
is no keyframe info in push mode.
The performance category is meant to be used to audit codepaths that lead to bad
performance (e.g. copies, conversion that can be avoided).
Remove the event category which is not used.
For vbr audio streams we need to use the number of blocks to calculate the
timestamps.
When the allocation of additional index memory fails, don't throw away what
we had before.
Various cleanups.
Implement scanning of the file when we can parse the index.
Some refactoring of common code.
Cleanups and comments.
Remove some reimplemented code.
Remove index massage code and put a FIXME where we should do something
equivalent later.
Remove some duplicate counters.
Be smarter when updateing the current the timestamp and offset in the stream
because we can reuse previously calculated values when simply go forward one
step.
Correctly set metadata on outgoing buffers.
Add a new function and datastructure to parse and hold the index entries on a
per stream base. Also avoid doing too much work trying to figure out the
timestamps and durations as we can trivially do that later.
Less information in the entries makes them 2 times smaller and not doing too
much work makes this code about 12 times faster than the regular case.
Hook in the new function alongside the existing function for comparison until
the rest of the code is updated to handle the new index datastructure.
Don't call gst_avi_demux_src_convert() for each single index entry. Not
only do we already have the pointer to the stream context, we also know
the formats we want to convert from and to already, so we may just as
well use optimised conversion routines that bypass some of the checks
and lookups made in gst_avi_demux_src_convert().
In streaming mode, avidemux is not supposed to send an EOS event downstream but
it is supposed to return UNEXPECTED from the chain function instead so that
upstream can do the right EOS handling.
Fix the duration query so that it also works with formats other than
TIME, such as DEFAULT to get the number of frames.
Add a convert function.
Fixes#578052.
Reading integers from random memory addresses will result
in SIGBUS on some architectures if the memory address
is not correctly aligned. This can happen at two
places in avidemux so we should use GST_READ_UINT32_LE
and friends here. Fixes bug #572256.
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/avi/gstavidemux.c:
More logging.
* gst/avi/gstavimux.c:
Handle more metadata fields. Better estimate of metadata size. Don't
merge received tags, if application has specified tags using
GST_TAG_MERGE_REPLACE_ALL. Fixes#563221 for avi.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_stream_scan):
Skip entries for streams that don't have a output pad yet, thereby
avoiding calling pad functions with a NULL pad.
Fixes#556424
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_subindex),
(gst_avi_demux_parse_stream), (gst_avi_demux_parse_index):
* gst/avi/gstavidemux.h:
For timestamping audio packets we need to take into account the
amount of blocks in one entry using the blockalign. Fixes some sync
issues with zero-padded audio blocks in the beginning of avi files.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_read_subindexes_pull),
(gst_avi_demux_read_subindexes_push):
Some AVI 2.0 (ODML) files don't respect the 'specifications' completely
and instead of using the 'ix##' nomenclature, use '##ix'.
They're still valid though, this fixes the duration and indexes for
virtually all the ODML files I have.
Original commit message from CVS:
Patch by: Alessandro Decina <alessandro at nnva dot org>
* gst/avi/gstavidemux.c: (gst_avi_demux_riff_parse_vprp):
Fix build of avidemux on big endian architectures.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_stream),
(gst_avi_demux_parse_index), (gst_avi_demux_massage_index),
(gst_avi_demux_calculate_durations_from_index),
(gst_avi_demux_stream_header_push),
(gst_avi_demux_stream_header_pull):
Try to ignore unparsable/unknown streams and give a warning instead of
erroring out. Fixes#537377.
Original commit message from CVS:
* gst/avi/gstavidemux.c:
Fix up my last commit. Use G_GUINT32_FORMAT for the guint32 debug log.
Also downgrade a GST_WARNING to GST_DEBUG and add a comment.
Original commit message from CVS:
* gst/avi/gstavidemux.c:
Revert patch which sends timestamps only on keyframes, as it
breaks playback with current gst-ffmpeg.
Fixes: #515562
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_subindex):
If there's no entries in the subindex, don't try to do anything stupid,
just return.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_process_next_entry),
(gst_avi_demux_stream_data):
GStreamer timestamps are PTS values while AVI only knows about DTS
timestamps. Make sure we only copy the DTS as the buffer timestamp when
we are dealing with a key frame.
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/avi/gstavidemux.c:
* gst/wavparse/gstwavparse.c:
Return the result in _activate_pull(). Don't ref element there.
Original commit message from CVS:
* gst/avi/gstavidemux.c:
* gst/qtdemux/qtdemux.c:
Drop EOS in _handle_src_event(). Fix the refcount in qtdemux that
previous commit messed up.
Original commit message from CVS:
* gst/avi/gstavidemux.c:
* gst/qtdemux/qtdemux.c:
Sync _handle_src_event() with oggdemux. In avidemux also ref the
element when we should, but not when we its not needed.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_base_init),
(gst_avi_demux_reset), (gst_avi_demux_parse_stream):
* gst/avi/gstavidemux.h:
Parse subtitle text streams instead of erroring out (#442034). Still
needs a parser for the subtitles to actually show up.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_push_event),
(gst_avi_demux_loop):
Make _push_event() return TRUE if the event could be pushed on at
least one pad and not only if it could be pushed on all pads,
otherwise we'll end up posting an error message on EOS if one or
more source pads are not connected.
Original commit message from CVS:
2007-04-05 Julien MOUTTE <julien@moutte.net>
* gst/avi/gstavidemux.c: (gst_avi_demux_process_next_entry):
Don't leave the offsets defined by upstream element on the
compressed data buffer we are pushing downstream. Make them
GST_BUFFER_OFFSET_NONE.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_subindex),
(gst_avi_demux_parse_index):
* sys/v4l2/v4l2_calls.c: (gst_v4l2_fill_lists):
* sys/v4l2/v4l2src_calls.c: (gst_v4l2src_grab_frame):
Printf format fixes; also add some missing quotes in translated
strings. Fixes#416728 and #416727.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_stream),
(gst_avi_demux_push_event), (gst_avi_demux_do_seek),
(gst_avi_demux_process_next_entry), (gst_avi_demux_stream_data),
(gst_avi_demux_chain):
Fix stream position reporting after a seek. Fixes#416445.
Original commit message from CVS:
Patch by: René Stadler <mail at renestadler dot de>
* gst/avi/gstavidemux.c: (gst_avi_demux_parse_stream),
(gst_avi_demux_push_event), (gst_avi_demux_process_next_entry),
(gst_avi_demux_stream_data), (gst_avi_demux_chain):
Make avidemux accept optional header chunks in any order.
Fixes#415446.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_class_init),
(gst_avi_demux_reset), (gst_avi_demux_index_entry_for_time),
(gst_avi_demux_handle_src_query), (gst_avi_demux_parse_superindex),
(gst_avi_demux_parse_subindex), (gst_avi_demux_parse_stream),
(gst_avi_demux_parse_index), (gst_avi_demux_stream_index),
(gst_avi_demux_sync), (gst_avi_demux_next_data_buffer),
(gst_avi_demux_stream_scan), (gst_avi_demux_massage_index),
(gst_avi_demux_calculate_durations_from_index),
(gst_avi_demux_push_event), (gst_avi_demux_stream_header_pull),
(gst_avi_demux_do_seek), (gst_avi_demux_process_next_entry),
(gst_avi_demux_stream_data), (gst_avi_demux_loop):
* gst/avi/gstavidemux.h:
Save some memory (8%) by repacking the index entry structure (more to
come). Add more FIXMEs to questionable parts.
Original commit message from CVS:
* gst/avi/gstavidemux.c: (gst_avi_demux_stream_header_pull):
Error out properly when pull_range fails while we're reading the
headers, instead of just pausing the task silently. Fixes#399338.
