This was tested with medias recorded from an iPhone XR and an iPhone 13.
Here is how a typical stream looks like in coding order:
┌────────┬─────┬─────┬──────────┐
│ sample | PTS | DTS | keyframe |
├────────┼─────┼─────┼──────────┤
┊ ┊ ┊ ┊ ┊
│ 53 │ 560 │ 510 │ No │
│ 54 │ 540 │ 520 │ No │
│ 55 │ 530 │ 530 │ No │
│ 56 │ 550 │ 540 │ No │
│ 57 │ 600 │ 550 │ Yes │
│ * 58 │ 580 │ 560 │ No │
│ * 59 │ 570 │ 570 │ No │
│ * 60 │ 590 │ 580 │ No │
│ 61 │ 640 │ 590 │ No │
│ 62 │ 620 │ 600 │ No │
┊ ┊ ┊ ┊ ┊
In composition/display order:
┌────────┬─────┬─────┬──────────┐
│ sample | PTS | DTS | keyframe |
├────────┼─────┼─────┼──────────┤
┊ ┊ ┊ ┊ ┊
│ 55 │ 530 │ 530 │ No │
│ 54 │ 540 │ 520 │ No │
│ 56 │ 550 │ 540 │ No │
│ 53 │ 560 │ 510 │ No │
│ * 59 │ 570 │ 570 │ No │
│ * 58 │ 580 │ 560 │ No │
│ * 60 │ 590 │ 580 │ No │
│ 57 │ 600 │ 550 │ Yes │
│ 63 │ 610 │ 610 │ No │
│ 62 │ 620 │ 600 │ No │
┊ ┊ ┊ ┊ ┊
Sample/frame 58, 59 and 60 are B-frames which actually depends on the
key frame (57). Here the key frame is not an IDR but a "CRA" (Clean
Random Access).
Initially, I thought I could rely on the sdtp box (independent and
disposable samples), but unfortunately:
sdtp[54] is_leading:0 sample_depends_on:1 sample_is_depended_on:0 sample_has_redundancy:0
sdtp[55] is_leading:0 sample_depends_on:1 sample_is_depended_on:2 sample_has_redundancy:0
sdtp[56] is_leading:0 sample_depends_on:1 sample_is_depended_on:2 sample_has_redundancy:0
sdtp[57] is_leading:0 sample_depends_on:2 sample_is_depended_on:0 sample_has_redundancy:0
sdtp[58] is_leading:0 sample_depends_on:1 sample_is_depended_on:0 sample_has_redundancy:0
sdtp[59] is_leading:0 sample_depends_on:1 sample_is_depended_on:2 sample_has_redundancy:0
sdtp[60] is_leading:0 sample_depends_on:1 sample_is_depended_on:2 sample_has_redundancy:0
sdtp[61] is_leading:0 sample_depends_on:1 sample_is_depended_on:0 sample_has_redundancy:0
sdtp[62] is_leading:0 sample_depends_on:1 sample_is_depended_on:0 sample_has_redundancy:0
The information that might have been useful here would have been
is_leading, but all the samples are set to 0 so this was unusable.
Instead, we need to rely on sgpd/sbgp tables. In my case the video track
contained 3 sgpd tables with the following grouping types: tscl, sync
and tsas. In the sync table we have the following 2 entries (only):
sgpd.sync[1]: sync nal_unit_type:0x14
sgpd.sync[2]: sync nal_unit_type:0x15
(The count starts at 1 because 0 carries the undefined semantic, we'll
see that later in the reference table).
The NAL unit types presented here correspond to:
libavcodec/hevc.h: HEVC_NAL_IDR_N_LP = 20,
libavcodec/hevc.h: HEVC_NAL_CRA_NUT = 21,
In parallel, the sbgp sync table contains the following:
┌────┬───────┬─────┐
│ id │ count │ gdi │
├────┼───────┼─────┤
│ 0 │ 1 │ 1 │
│ 1 │ 56 │ 0 │
│ 2 │ 1 │ 2 │
│ 3 │ 59 │ 0 │
│ 4 │ 1 │ 2 │
│ 5 │ 59 │ 0 │
│ 6 │ 1 │ 2 │
│ 7 │ 59 │ 0 │
│ 8 │ 1 │ 2 │
│ 9 │ 59 │ 0 │
│ 10 │ 1 │ 2 │
│ 11 │ 11 │ 0 │
└────┴───────┴─────┘
The gdi column (group description index) directly refers to the index in
the sgpd.sync table. This means the first frame is an IDR, then we have
batches of undefined frames interlaced with CRA frames. No IDR ever
appears again (tried on a 30+ seconds sample).
With that information, we can build an heuristic using the presentation
order.
A few things needed to be introduced in this commit:
1. min_sample_duration is extracted from the stts: we need the minimal
step between sample in order to PTS-step backward to a valid point
2. In order to avoid a loop over the ctts table systematically during a
seek, we build an expanded list of sample offsets which will be used
to translate from DTS to PTS
3. An open_key_samples index to keep track of all the non-IDR key
frames; for now it only supports HEVC CRA frames. We should probably
add BLA frames as well, but I don't have any sample so I prefered to
leave that for later
It is entirely possible I missed something obvious in my approach, but I
couldn't come up with a better solution. Also, as mentioned in the diff,
we could optimize is_open_key_sample(), but the linear scaling overhead
should be fine for now since it only happens in seek events.
Fixing this issue prevents sending broken packets to the decoder. With
FFmpeg hevc decoder the frames are skipped, with VideoToolbox the frames
are glitching.
sgpd means Sample Group Description Box.
For now, only the sync grouping type is parsed, but the function can
easily be adjusted to support other flavours.
The sbgp (Sample to Group Box) sync_group table built in previous commit
contains references to this table through the group_description_index
field.
If _FieldBased, _Matrix, _ColorRange, or _ChromaLocation haven't
been set, that absence would be interpreted as 0, leading to those
being set to case 0 instead of default. There is no case 0 for
_Primaries and _Transfer, so those were correctly falling back
to the default case.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>
It appears this is not allowed "Each Segment Index box documents how a (sub)segment is divided into one or more subsegments
(which may themselves be further subdivided using Segment Index boxes)."
Fixes: Null pointer dereference
Fixes: Ticket9517
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
It's needed for avformat_get_mov_video_tags() and avformat_get_mov_audio_tags(),
both public symbols defined in avformat.h
Signed-off-by: James Almer <jamrial@gmail.com>
Fixes: -nan is outside the range of representable values of type 'long'
Fixes: 44614/clusterfuzz-testcase-minimized-ffmpeg_dem_WEBM_DASH_MANIFEST_fuzzer-6216204841254912
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Some of these were made possible by moving several common macros to
libavutil/macros.h.
While just at it, also improve the other headers a bit.
Reviewed-by: Martin Storsjö <martin@martin.st>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It has been included since af5f434f8c
for deprecation reasons, but removing it has been forgotten after
it had served is purpose. So remove it.
For convenience, include version.h instead as LIBAVUTIL_VERSION_INT
is supposed to be used when creating AVClasses.
Reviewed-by: Martin Storsjö <martin@martin.st>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Since the check got simplified and stdbool was no longer necessary
to include, neither is that variable. Silences a warning.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>
The IMF demuxer does not set the DTS and PTS of packets accurately in all
scenarios. Moreover, audio packets are not trimmed when they exceed the
duration of the underlying resource.
imf-cpl-with-repeat FATE ref file is regenerated.
Addresses https://trac.ffmpeg.org/ticket/9611
IMF CPLs can reference thousands of files, which can result in system limits
for the number of open files to be exceeded. The following patch opens and
closes files as needed.
Addresses https://trac.ffmpeg.org/ticket/9623
Trying to be clever about determining between interface version 8
and 8.1 ended up with pre-8.1 versions of AviSynth+ segfaulting.
The amount of time between interface version 8.1 and 9 is small,
so just restrict the frameprop awareness to version 9 and call it
a day.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>
The ID3v2.4.0 standard defines TIT1 as the "Content group description"
tag [1]. This frame is usually referred to as the "Grouping" tag and in
de-facto use under that name by Vorbis and APEv2 [2].
This commit introduces a mapping from "TIT1" to "grouping" in the
id3v2.4 metadata conversion table. This will enable software to access
it using that name. In particular, MPD will now read this tag correctly
when using the ffmpeg decoder plugin.
[1] https://id3.org/id3v2.4.0-frames (4.2.1)
[2] https://picard-docs.musicbrainz.org/en/appendices/tag_mapping.html#grouping-3
Signed-off-by: Wolfgang Müller <wolf@oriole.systems>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
We should use the systems crypto policy by default. If there is no
system policy, gnutls will use the "NORMAL" policy.
Signed-off-by: Andreas Schneider <asn@cryptomilk.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
* Field Order
* Chroma Location
* Color Transfer Characteristics
* Color Range
* Color Primaries
* Matrix Coefficients
The existing TFF/BFF detection is retained as a fallback for
older versions of AviSynth that can't access frame properties.
The other properties have no legacy equivalent to detect them.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>
AviSynth works on frame-based video by default, which can
be either progressive or interlaced. Some filters can break
frames into half-height fields, at which point it considers
the clip to be field-based (avs_is_field_based can be used
to check for this situation).
To properly detect the field order of a typical video clip,
the frame needs to have been weaved back together already,
so avs_is_field_based should actually report 'false' when
checked.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>
Suggested by zhilizhao, vlc project has solved the compatibility by
the same way, so I borrowed the comments from vlc project.
Fixes ticket #9449
Reviewed-by: Marton Balint <cus@passwd.hu>
Signed-off-by: Limin Wang <lance.lmwang@gmail.com>
Zhao Zhili added a ttl upper bound in commit 9daac85da8,
but the check for ttl in url is missing still.
Signed-off-by: Limin Wang <lance.lmwang@gmail.com>
Using Range allows for getting the full file size from the
Content-Range header in the response, even if the server sends
back the response using chunked Transfer-Encoding, which does not
allow using Content-Length.
When Transfer-Encoding:chunked is used, the client must ignore a
Content-Length header, if present. However, it should not ignore a
Content-Range header, which also includes the full size of the
entity.
bca30570d2 added a user option to set max_packet_size replacing
a hardcoded value. This had a side-effect of leaving the field
set to 0 when packet demuxing is carried out from another demuxer
using avpriv functions, which could lead to demux failure.
Hardcoded max_packet_size inside avpriv_mpegts_parse_open to
2048000 to avoid this. Value chosen to be 10x that of default value
to accommodate large payloads.
Fixes: signed integer overflow: 9223372036854775748 + 60 cannot be represented in type 'long'
Fixes: 44417/clusterfuzz-testcase-minimized-ffmpeg_dem_HLS_fuzzer-5802443881971712
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Steven Liu <lingjiujianke@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: Division by 0
Fixes: 44035/clusterfuzz-testcase-minimized-ffmpeg_dem_WEBM_DASH_MANIFEST_fuzzer-4826721386364928
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: pointer index expression with base 0x000000000000 overflowed to 0xffffffffffffffff
Fixes: 44012/clusterfuzz-testcase-minimized-ffmpeg_dem_HLS_fuzzer-5670607746891776
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The same check is added to zlib too, it seems not needed there though
Fixes: Infinite loop
Fixes: 43932/clusterfuzz-testcase-minimized-ffmpeg_dem_MATROSKA_fuzzer-6175167573786624
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Updating a chapter with the same id does not break monotonicity
Fixes: Timeout
Fixes: 43727/clusterfuzz-testcase-minimized-ffmpeg_dem_MOV_fuzzer-4960623367159808
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Before, seeking in hls streams would always seek to the next keyframe
after the given timestamp. With this fix, if seeking in videostream and
AVSEEK_FLAG_BACKWARD is set, seeking will be to the first keyframe of
the segment containing the given timestamp. This fixes#7485.
Signed-off-by: Gustav Grusell <gustav.grusell@gmail.com>
This is by definition the appropriate place for it.
Remove all the now unnecessary libavcodec/internal.h inclusions;
also remove other unnecessary headers from the affected files.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
