avformat/utils: Clear pointer in ff_alloc_extradata() to avoid leaving a stale pointer in memory

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avfilter/buffer: use av_freep() to avoid leaving stale pointers in memory

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/hlsproto: Use av_freep, to avoid leaving stale pointers in memory

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/matroskadec: Use av_freep() to avoid leaving stale pointers in memory

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/mov: use av_freep() to avoid leaving stale pointers in memory

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/mov: Clear array to prevent potential out of array read from av_dlog()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

lavfi: check av_strdup() return value

Signed-off-by: Paul B Mahol <onemda@gmail.com>

mov: Fix negative size calculation in mov_read_default().

The previous code assumed if an atom was marked with a 64-bit
size extension, it actually had that data available. The new
code verfies there's enough data in the atom for this to be
done.

Failure to verify causes total_size > atom.size which will
result in negative size calculations later on.

Found-by: Paul Mehta <paul@paulmehta.com>
Signed-off-by: Dale Curtis <dalecurtis@chromium.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/mov: fix integer overflow in mov_read_udta_string()

Found-by: Paul Mehta <paul@paulmehta.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

mov: Fix overflow and error handling in read_tfra().

Under abnormal conditions the item_count may exceed the max
allocation size on 32-bit systems, this causes the allocated
size to overflow and become too small for the given count.

Additionally, if av_reallocp() fails its allocation, the
fragment_index_count is not correctly decremented.

Ensuring further havoc may be wrought, the error code for
read_tfra() is not checked upon return.

Found-by: Paul Mehta <paul@paulmehta.com>
positive return code and use of _array functions by commiter

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avfilter/vf_idet: Add analyze_interlaced_flag mode

This should allow us to insert idet before scale and let scale have interl=-1 as default in that case

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avfilter/vf_yadif: detect telecine content

Fixes: yadif with interlaced_flag_switch.mpeg
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/id3v1: strip trailing whitespace

ID3v1 fields have a fixed size, and they are padded either with zeros,
or with spaces. Handle the latter case, instead of putting strings with
trailing spaces into the AVDictionary.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

x86/flacdsp: remove unneeded ifdeffery

x86inc can translate r*m into a register or stack on its own

Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>

avformat/dvbsub: increase probe score

Ive not found a single false positive after some moderate testing with
probetest

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/dvbsub: fix indent after previous commit

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/dvbsub: check for 0xf0 earlier in probe, 30% faster

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Merge remote-tracking branch 'origin/master'

* origin/master:
  libavfilter/vf_blend: do not unconditionally compile both filters

Merged-by: Michael Niedermayer <michaelni@gmx.at>

libavfilter/vf_blend: do not unconditionally compile both filters

Signed-off-by: Paul B Mahol <onemda@gmail.com>

avcodec/dvdsubdec: fix out of bounds accesses

The code blindly trusted buffer offsets read from the file in the RLE
decoder. Explicitly check the offset. Also error out on other RLE
decoding errors.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

doc/ffmpeg: mention both ffpreset/avpreset in documentation, remove superfluous example

ffmpeg looks for both .ffpreset and .avpreset files depending on whether the
-[avsf]pre or -pre option is used. Added two sections for each type of preset
including the rules according to which files are searched.

(Notably, the lookup order is swapped for avpreset files, because it first
looks for <codec>_<arg>.avpreset and then for <arg>.avpreset.)

This removes the section explaining -pre only, which was under "Examples",
where it did not really make sense.

Signed-off-by: Werner Robitza <werner.robitza@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '46808fdf04ab113df374157b90b506eb3110daf2'

* commit '46808fdf04ab113df374157b90b506eb3110daf2':
  movenc: Enable editlists by default if delay_moov is enabled

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '9f810a9b374e0ff8e2a19fd8a7347afe2933e229'

* commit '9f810a9b374e0ff8e2a19fd8a7347afe2933e229':
  libavformat: Check for malloc failures in avformat_new_stream

See: a66893ac949864352b36b39e48c4cd72bbd81e54
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'c4aa0f865fb2c3b28a309d1bb27fd1c24b75ba28'

* commit 'c4aa0f865fb2c3b28a309d1bb27fd1c24b75ba28':
  libavformat: Remove a misleading/incorrect comment

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'f4d4e66a24a5c9497a5b6d3c089ac58089a87428'

* commit 'f4d4e66a24a5c9497a5b6d3c089ac58089a87428':
  configure: Remap -L to -libpath for msvc

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '3e160652219ff4da433f5672ae1e5f4956abb815'

* commit '3e160652219ff4da433f5672ae1e5f4956abb815':
  cmdutils: update copyright year to 2015.

Conflicts:
cmdutils.c

See: b8db25a3338b67186837c49580fe538d63dd73c7
Merged-by: Michael Niedermayer <michaelni@gmx.at>

avformat/aiffdec: only read codec tag when there is space in header

Signed-off-by: Peter Ross <pross@xvid.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/utils: add dvbsub to set_codec_from_probe_data()

Fixes part of ticket 4221

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat: add dvbsub demuxer

This is needed for probing some dvbsub streams in mpegts

Fixes part of ticket4221

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/rawdec: Add FF_DEF_RAWSUB_DEMUXER

This is similar to the corresponding audio and video macros

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/rawdec: Make ff_raw_data_read_header() non static

This is similar to the raw video read header

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvenc: omit more metadata elements with specific meaning

Reviewed-by: Thilo Borgmann <thilo.borgmann@mail.de>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Factorize avpriv_mirror() out

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avfilter/vf_sab: fix filtering tiny images

Fixes out of array reads

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

movenc: Enable editlists by default if delay_moov is enabled

Being able to write editlists properly is one of the main points
in the delay_moov flag.

Signed-off-by: Martin Storsjö <martin@martin.st>

libavformat: Check for malloc failures in avformat_new_stream

CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>

libavformat: Remove a misleading/incorrect comment

This comment can be traced back to the initial commit from 2001,
and it seemed to be misleading/incorect already back then. (It
was used for normal, non-raw file formats already then.)

Signed-off-by: Martin Storsjö <martin@martin.st>

configure: Remap -L to -libpath for msvc

This allows using libraries that are detected via pkg-config with
msvc. (The libraries themselves may have to be built with MSVC
though.)

Signed-off-by: Martin Storsjö <martin@martin.st>

cmdutils: update copyright year to 2015.

Signed-off-by: Martin Storsjö <martin@martin.st>

ffserver: reindent rtsp_cmd_setup()

Signed-off-by: Reynaldo H. Verdejo Pinochet <reynaldo@osg.samsung.com>

ffserver: reflow rtsp_cmd_setup()

Signed-off-by: Reynaldo H. Verdejo Pinochet <reynaldo@osg.samsung.com>

ffserver: drop redundant else clause

Signed-off-by: Reynaldo H. Verdejo Pinochet <reynaldo@osg.samsung.com>

ffserver: reindent compute_status()

Signed-off-by: Reynaldo H. Verdejo Pinochet <reynaldo@osg.samsung.com>

ffserver: reflow compute_status()

Signed-off-by: Reynaldo H. Verdejo Pinochet <reynaldo@osg.samsung.com>

avcodec/libxvid: check for av_malloc*() failures

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avcodec/ljpegenc: Check for av_malloc_array() failure

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/westwood_vqa: Remove unneeded cast

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/movenc: fix DELAY_MOOV

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

tests/regression-funcs: Remove hardcoded threads 1

The tests which use encoders which either use slices or store the encoder thread count
keep a hardcoded value of 1

This will help test more threading code like in filters

Found-by: ubitux
Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'f2ad1495f23376ce61542967f4fc14205f284d40'

* commit 'f2ad1495f23376ce61542967f4fc14205f284d40':
  avisynth: Use RTLD_LOCAL instead of RTLD_GLOBAL

Merged-by: Michael Niedermayer <michaelni@gmx.at>

build: require libxcb >= 1.4

Fixes Ticket #4139

Merge commit '470c9db11ff2c3249e995e7ba68e87bb81bf778c'

* commit '470c9db11ff2c3249e995e7ba68e87bb81bf778c':
  sidxindex: Remove parsing that isn't necessary any longer

Merged-by: Michael Niedermayer <michaelni@gmx.at>

configure: bump year

Happy new year!

Merge commit '5cf6bda6e2eae496e8eb2bb06c96852d59a58b8a'

* commit '5cf6bda6e2eae496e8eb2bb06c96852d59a58b8a':
  sidxindex: Don't adjust the Period start time depending on the track start time

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '7a1a63e34fa46af18311c2493fdaec9a93bdb750'

* commit '7a1a63e34fa46af18311c2493fdaec9a93bdb750':
  dashenc: Use pts for MPD timeline timestamps

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'c5e7ea13d2d4da0c5da91973a547afff6fe9e011'

* commit 'c5e7ea13d2d4da0c5da91973a547afff6fe9e011':
  dashenc: Use delay_moov

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'b3b0b35db2f3b61bf2f0f4fa85f5b6267d83c8fe'

* commit 'b3b0b35db2f3b61bf2f0f4fa85f5b6267d83c8fe':
  movenc: Get rid of a hack for updating the dvc1 atom

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '847bf5988fec1d3e65c1d8cf0cdb8caf0cfd0c1b'

* commit '847bf5988fec1d3e65c1d8cf0cdb8caf0cfd0c1b':
  movenc: Add an option for delaying writing the moov with empty_moov

Conflicts:
libavformat/movenc.c
libavformat/version.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>

avformat/movenc: Use cluster timestamps when available in edts

Also print both if they differ

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit 'c725faebda9a516766d94c33b07972ab0f70cf93'

* commit 'c725faebda9a516766d94c33b07972ab0f70cf93':
  movenc: Use start_dts/cts instead of cluster[0] for writing edit lists

Conflicts:
libavformat/movenc.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

avisynth: Use RTLD_LOCAL instead of RTLD_GLOBAL

There shouldn't be any need to add the loaded libraries to the global
symbol namespace.

Signed-off-by: Martin Storsjö <martin@martin.st>

Merge commit '724cbea7193945fe5a5b4dea8ede344803572844'

* commit '724cbea7193945fe5a5b4dea8ede344803572844':
  movenc: Remove an unnecessary condition when flushing fragments

Merged-by: Michael Niedermayer <michaelni@gmx.at>

Merge commit '355d01a1bf55297b1d1f04e4bfbf0ddc93b6247e'

* commit '355d01a1bf55297b1d1f04e4bfbf0ddc93b6247e':
  movenc: Factorize writing ftyp and other identification tags to a separate function

Conflicts:
libavformat/movenc.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

sidxindex: Remove parsing that isn't necessary any longer

When we don't adjust the Period start time, we don't need to
parse the earliest_presentation_time from the sidx boxes either.

Signed-off-by: Martin Storsjö <martin@martin.st>

sidxindex: Don't adjust the Period start time depending on the track start time

This was only necessary to get playback to start with dash.js 1.2.0,
it has been fixed in the git version.

The previous behaviour was incorrect - the Period's start time
is irrespective of the actual first timestamp of the contents
within the period. The Period start time only says when, within the
global timeline, this particular piece should start to be played
back.

Signed-off-by: Martin Storsjö <martin@martin.st>

dashenc: Use pts for MPD timeline timestamps

This should be more correct. This also should give more sensible
switching between video streams with different amount of b-frame
delay.

The current dash.js release (1.2.0) fails to start playback of
such files from the start (if the start pts is > 0), but this has
been fixed in the current git version of dash.js.

Also enable the use of edit lists, so that streams in many cases
start at pts=0.

Signed-off-by: Martin Storsjö <martin@martin.st>

dashenc: Use delay_moov

Signed-off-by: Martin Storsjö <martin@martin.st>

movenc: Get rid of a hack for updating the dvc1 atom

Use the more generic approach with the delay_moov flag, instead of
having a update mechanism specific to this one single atom.

Signed-off-by: Martin Storsjö <martin@martin.st>

movenc: Add an option for delaying writing the moov with empty_moov

This delays writing the moov until the first fragment is written,
or can be flushed by the caller explicitly when wanted. If the first
sample in all streams is available at this point, we can write
a proper editlist at this point, allowing streams to start at
something else than dts=0. For AC3 and DNXHD, a packet is
needed in order to write the moov header properly.

This isn't added to the normal behaviour for empty_moov, since
the behaviour that ftyp+moov is written during avformat_write_header
would be changed. Callers that split the output stream into header+segments
(either by flushing manually, with the custom_frag flag set, or by
just differentiating between data written during avformat_write_header
and the rest) will need to be adjusted to take this option into use.

For handling streams that start at something else than dts=0, an
alternative would be to use different kinds of heuristics for
guessing the start dts (using AVCodecContext delay or has_b_frames
together with the frame rate), but this is not reliable and doesn't
necessarily work well with stream copy, and wouldn't work for getting
the right initialization data for AC3 or DNXHD either.

Signed-off-by: Martin Storsjö <martin@martin.st>

movenc: Use start_dts/cts instead of cluster[0] for writing edit lists

This allows writing edit lists even when track->entry == 0, if
the start times have been set.

Signed-off-by: Martin Storsjö <martin@martin.st>

movenc: Remove an unnecessary condition when flushing fragments

If fragments == 0 it means we haven't written any moov atom yet.
If the empty_moov flag is set, we already have written an empty moov
atom at startup. Thus, the check for empty_moov is redundant.

This is in preparation for allowing writing the moov atom later,
even when using the empty moov flag.

Signed-off-by: Martin Storsjö <martin@martin.st>

movenc: Factorize writing ftyp and other identification tags to a separate function

Signed-off-by: Martin Storsjö <martin@martin.st>

avutil/camellia: cosmetic fixes

Signed-off-by: Giorgio Vazzana <mywing81@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avutil/camellia: use K[2] instead of *K in generate_round_keys()

Additionally, change parameters order.

Signed-off-by: Giorgio Vazzana <mywing81@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avutil/camellia: make LR128() more robust

Signed-off-by: Giorgio Vazzana <mywing81@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avutil/camellia: fix documentation for av_camellia_crypt()

Signed-off-by: Giorgio Vazzana <mywing81@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

lavd/lavfi: allow to extract subcc.

Signed-off-by: Nicolas George <george@nsup.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

libavutil: Added Camellia symmetric block cipher

Signed-off-by: Supraja Meedinti <supraja0493@gmail.com>
Previous version of this patch reviewed-by: Giorgio Vazzana <mywing81@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Changelog: Add cropdetect non 8bpp

Suggested-by: Reimar
Previous patch reviewed-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/mov: reindent after previous commit

avformat/mov: move edit list heuristics into mov_build_index()

mov_read_elst() is now only responsible from storing the table in a data
structure; this is consistent with other table readers functions.

avcodec/mpeg12dec: Check actual aspect ratio instead of aspect_ratio_info

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avcodec/mpeg12dec: Recalculate SAR unconditionally

Fixes Ticket4161

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avcodec/mpeg12dec: Move user data debug code out of unrelated if (buf_size > 29)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvdec: remove duplicated line error

Fix typo regression since 5c37ffca59e7eb1b8768f237deb35a9936c1a0a5

Found-By: cousin_luigi

Merge remote-tracking branch 'cus/stable'

* cus/stable:
  ffplay: dump format before selecting streams
  ffplay: add support for stream specifiers in -ast, -vst, -sst options
  ffplay: remove unused no_background from videostate
  ffplay: remove flushed state from decoder context

Merged-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvdec: use named identifier instead of literal numbers as return values

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvdec: Increase string array size

Fixes parsing httphostheader of Scarlatti\,\ Pieter-Jan\ Belder\ -\ Sonata\ K113\ in\ A\ major\ -\ Alle.flv

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvdec: do not inject dts=0 metadata packets which failed to be parsed into a new data stream

Such data streams (which then contain no other packets except the faulty one)
confuse some user applications, like VLC
Works around vlcticket 12389

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/flvdec: add several error messages to error conditions

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

ffplay: dump format before selecting streams

This helps the user to see the available streams just before the error message
if the stream specifiers do not match.

Signed-off-by: Marton Balint <cus@passwd.hu>

ffplay: add support for stream specifiers in -ast, -vst, -sst options

Also fix the outdated documentation of these options.

Signed-off-by: Marton Balint <cus@passwd.hu>

ffplay: remove unused no_background from videostate

Signed-off-by: Marton Balint <cus@passwd.hu>

ffplay: remove flushed state from decoder context

Signed-off-by: Marton Balint <cus@passwd.hu>

avfilter/vf_idet: Add 9, 12 and 14 bit pixel formats

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avfilter/vf_idet: Use frame_requested instead of prev

This is more robust if the delay is not constant

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

avformat/cdxl: Fix integer overflow of image_size

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

x86/swr: disable pack_8ch functions on msvc/icl x86_32

Until a proper fix is committed.

Signed-off-by: James Almer <jamrial@gmail.com>

x86/swr: add missing alignment check to pack_6ch functions

Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>

avcodec/loco: check the init_get_bits8() return code

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>