All copyright holders have agreed to the relicensing.
Approved-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Approved-by: David Sedacca <sedacca@comcast.net>
Approved-by: Ganesh Ajjanagadde <gajjanag@mit.edu>
Approved-by: Jean First <jeanfirst@gmail.com>
Approved-by: Kyle Swanson <k@ylo.ph>
Approved-by: Michael Niedermayer <michael@niedermayer.cc>
Approved-by: Nicolas George <george@nsup.org>
Approved-by: Paul B Mahol <onemda@gmail.com>
Approved-by: Thilo Borgmann <thilo.borgmann@mail.de>
Expands the parser to also accept the separator '-' in addition to
'+', and take the negative sign into consideration.
The optional sign for the first factor in the expression is already
covered by parsing for an integer.
Signed-off-by: Moritz Barsnick <barsnick@gmx.net>
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
libavfilter/af_asyncts.c:212:9: warning: absolute value function 'labs' given an argument of type 'int64_t' (aka 'long long') but has parameter of type 'long' which may cause truncation of value [-Wabsolute-value]
* commit 'c3f113d58488df7594a489bdbb993a69ad47063c':
vf_hwdownload: allocate the destination frame for the pool size
Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
AVFilterLink.frame_count is supposed to count the number of frames
that were passed on the link, but with min_samples, that number is
not always the same for the source and destination filters.
With the addition of a FIFO on the link, the difference will become
more significant.
Split the variable in two: frame_count_in counts the number of
frames that entered the link, frame_count_out counts the number
of frames that were sent to the destination filter.
Also contains the following changes to the library:
- add ff_ prefix to functions
- remove cplusplus defines.
- add FF_ prefix to contants and some structs
- remove true peak calculation feature, since it uses its own resampler, and
af_loudnorm does not need it.
- remove version info and some fprintf(stderr) functions
- convert to use av_malloc
- always use histogram mode for LRA calculation, otherwise LRA data is slowly
consuming memory making af_loudnorm unfit for 24/7 operation. It also uses a
BSD style linked list implementation which is probably not available on all
platforms. So let's just remove the classic mode which not uses histogram.
- add ff_thread_once for calculating static histogram tables
- convert some functions to void which cannot fail
- remove intrinsics and some unused headers
- add support for planar audio
- remove channel / sample rate changer function, in ffmpeg usually we simply
alloc a new context
- convert some static variables to defines
- declare static histogram variables as aligned
- convert some initalizations to mallocz
- add window size parameter to init function and remove window size setter
function
- convert return codes to AVERROR
- fix indentation
Signed-off-by: Marton Balint <cus@passwd.hu>
2-channels convolution using complex fft
improves speed significantly
not sure if it should be enabled by default
so disable it by default
Signed-off-by: Muhammad Faiz <mfcc64@gmail.com>
Takes a frame associated with a hardware context as input and maps it
to something else (another hardware frame or normal memory) for other
processing. If the frame to map was originally in the target format
(but mapped to something else), the original frame is output.
Also supports mapping backwards, where only the output has a hardware
context. The link immediately before will be supplied with mapped
hardware frames which it can write directly into, and this filter
then unmaps them back to the actual hardware frames.
Also adds a new flag to mark filters which are aware of hwframes and
will perform this task themselves, and marks all appropriate filters
with this flag.
This is required to allow software-mapped hardware frames to work,
because we need to have the frames context available for any later
mapping operation in the filter graph.
The output from the filter graph should only propagate further to an
encoder if the hardware format actually matches the visible format
(mapped frames are valid here and have an hw_frames_ctx, but this
should not be given to the encoder as its hardware context).
Thanks to Mathieu Malaterre <malat@debian.org> for reporting the
Que/Queue typo. (https://bugs.debian.org/839542)
Reviewed-by: Lou Logan <lou@lrcd.com>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
This is a similar filter to f_metadata, only it works on side data. Since
adding side data from a user provided arbitrary binary string is unsafe,
because current code assumes that a side data of a certain kind has the proper
size, this filter only implements selection and deletion. Also, no value
matching support is implemented yet, because there is no uniform way to specify
a side data textually.
Signed-off-by: Marton Balint <cus@passwd.hu>
The parser for the outdef will accept a negative value for the first
named channel's gain. As negative values effectively only invert the
phase of the signal, and not negate the level, the gains' absolute
values must be used to correctly accumulate the levels.
Signed-off-by: Moritz Barsnick <barsnick@gmx.net>
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* commit '5b63b15663d31f50ce45d980b904a68795ad3f7a':
lavfi: set the link hwframes context before configuring the dst input
Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
As all known valid HDCD sample formats and sample rates are now handled
by the filter, remove the scan that "invades the privacy" of the filter graph
and turn off autoconvert by default as requested by Nicolas George.
http://ffmpeg.org/pipermail/ffmpeg-devel/2016-August/197571.html
Signed-off-by: Burt P <pburt0@gmail.com>
I don't have any legitimate 20 or 24-bit HDCD to test. It is known
that the PM Model Two would insert packets into 20 and 24-bit output,
but I have no idea what differences in behavior existed when decoding
20 or 24-bit. For now, as with 16-bit, PE (if enabled) will expand the
top 3dB into 9dB and LLE (gain adjust) will be applied if signaled.
Signed-off-by: Burt P <pburt0@gmail.com>
New versions of hdcd_scan() and hdcd_integrate() that also do the
work of hdcd_scan_stereo() and hdcd_integrate_stereo().
Some code split into previously separate functions to remove
duplication is now merged back into each function in the single
place where it is used.
Signed-off-by: Burt P <pburt0@gmail.com>
The buffer is already being copied anyway, so interlace the planar
format during the copy and remove one use of auto-convert.
Signed-off-by: Burt P <pburt0@gmail.com>
The PM Model Two could output HDCD-encoded audio in CD and all
DVD-Audio sample rates. (44100, 48000, 88200, 96000, 176400, and
192000 Hz)
Signed-off-by: Burt P <pburt0@gmail.com>
This is the assumption that is made in pixel format conversion do
throughout the code (in particular swscale), and BT-specifications
mandate.
Add a warning to inform the user that an automatic selection is being
made.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
If the input has been decoded from a stream which uses edge cropping
then the whole surface need not be valid. This defines an input
region for the scaler so we only use the active area of the frame.
Useful when the amerge filter parameters are generated from a script based on
the number of input streams, by allowing 1 input it does not have to be handled
specially.
The split filter also allows 1 output, so it is more consistent to allow
merging 1 input as well.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Marton Balint <cus@passwd.hu>
Allows to use values returned from API and from ffprobe directly.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Reviewed-by: Pavel Koshevoy <pkoshevoy@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The filter needs input frames with color properties filled out by
the decoder. Since this is not always possible, add input options to
the filter so that user may override color space, color primaries,
transfer characteristics, and color range, as well as a generic option
to set all properties at once.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Whenever a full range video is input, since the YUVJ* formats are not
listed as supported for this filter, a range reduction takes place
through the auto-inserted format filter, forcing the conversion to
operate on a limited range,
However the filter handles full range videos perfectly fine, so adding
support to YUVJ* formats will allow skipping a conversion step, while
providing completely identical results.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This function can potentially allocate memory.
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
IIRC, The spaces are not standard before the #
Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Mostly just re-arranges some code to make it easier to update this
filter and libhdcd together. filter_frame() is much simpler as a
result.
* use the HDCD detection data structure and functions from libhdcd,
moved detection code out of filter_frame()
* moved analyze_mode preparation out of filter_frame() into
hdcd_analyze_prepare(), from libhdcd
* moved some macro definitions to the top so they are all together
Signed-off-by: Burt P <pburt0@gmail.com>
The names are hard to understand and also seem to break build on at least some solaris
versions
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Anton Khirnov:
"[av_frame_copy_props()] potentially contains memory allocation,
so the return value needs to be checked."
Signed-off-by: Burt P <pburt0@gmail.com>
Should fix compilation with non C99 compilers like msvc 2012, where
acosh() is not available.
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
Restore a check added in 440af105f2
but lost sometime after. avfilter_graph_set_auto_convert() will
have an effect once again.
Signed-off-by: Burt P <pburt0@gmail.com>
Also checking AVFilterLink->type is AVMEDIA_TYPE_AUDIO before
calling av_get_sample_fmt_name() on AVFilterLink->format.
Signed-off-by: Burt P <pburt0@gmail.com>
A new mode, selected by filter option, to aid in analysis of HDCD
encoded audio. In this mode the audio is replaced by a solid tone and
the amplitude is adjusted to signal some specified aspect of the process.
The output file can be loaded in an audio editor alongside the original,
where the user can see where different features or states are present.
Signed-off-by: Burt P <pburt0@gmail.com>
In order to avoid possible clipping at the very
beginning. Especially apparent when "alternative"
boundary mode is used.
Adopted from original commit by committer.
HDCD is only encoded in s16@44100Hz. Scan the chain of AVFilterLinks
for any resampling or format conversion/truncation that might cause
problems for the filter and issue warnings.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Issue #5727: gain adjustment should only be applied if matching
gain value from a valid packet in both channels. The existing functions process
each channel separately, so it was not possible.
* New versions of hdcd_process(), hdcd_scan(), hdcd_integrate() named
hdcd_*_stereo() that process both channels together.
* target_gain applied will be the last matching target_gain.
* The old single channel functions remain as an option. They can be
used by: -af hdcd=process_stereo=0.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Used to attempt replication of some results from
http://www.audiomisc.co.uk/HFN/HDCD/Examined.html
May not be generally useful, defaults to off.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* Moves the filter context member out of state and into HDCDContext
* More useful information when an error is detected
* Gives a location near where the error was detected
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The counter is now -1 if the code detect timer was never set,
and 0 if it was set but never expired.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The stereowiden filter uses a buffer, s->buffer[], and a pointer
within the buffer, s->write, to implement inter-channel delays.
The loop which applies the delayed samples turns out to be faulty.
109 for (n = 0; n < in->nb_samples; n++, src += 2, dst += 2) {
110 const float left = src[0], right = src[1];
111 float *read = s->write + 2;
112
113 if (read > s->buffer + s->length)
114 read = s->buffer;
115
116 dst[0] = drymix * left - crossfeed * right - feedback * read[1];
117 dst[1] = drymix * right - crossfeed * left - feedback * read[0];
118
119 s->write[0] = left;
120 s->write[1] = right;
121
122 if (s->write == s->buffer + s->length)
123 s->write = s->buffer;
124 else
125 s->write += 2;
126 }
For one, the buffer gets written past its end in lines 119-120, before
the bound check is done in lines 122-123. This can be easily confirmed
by valgrind.
==3544== Invalid read of size 4
==3544== at 0x593B41: filter_frame (af_stereowiden.c:116)
==3544== Address 0xb1b03c4 is 4 bytes after a block of size 7,680 alloc'd
==3544==
==3544== Invalid read of size 4
==3544== at 0x593B66: filter_frame (af_stereowiden.c:117)
==3544== Address 0xb1b03c0 is 0 bytes after a block of size 7,680 alloc'd
==3544==
==3544== Invalid write of size 4
==3544== at 0x593B79: filter_frame (af_stereowiden.c:119)
==3544== Address 0xb1b03c0 is 0 bytes after a block of size 7,680 alloc'd
==3544==
==3544== Invalid write of size 4
==3544== at 0x593B7D: filter_frame (af_stereowiden.c:120)
==3544== Address 0xb1b03c4 is 4 bytes after a block of size 7,680 alloc'd
Also, using two separate pointers, s->write and read = s->write + 2,
does not seem to be well thought out. To apply the delay of s->buffer[],
it is enough to read the delayed samples at the current position within
the buffer, and then to store new samples at the same current position.
Thus the application of delayed samples can probably be best described
with a single pointer s->cur.
I also introduce a minor change to ensure that the size of s->buffer[]
is always a multiple of 2. Since the delay parameter is a float, it is
otherwise possible to trick the code into allocating off-by-one buffer.
Add an AVOption stats_version with a new header for V2 stats, which
specifies the stats log version and lists the fields that will be
present in the log (to ease parsing).
The primary motivation is to facilitate the addition of optional fields
to the log without breaking backwards compatibility, while making the
logs easier to parse.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The Peak Extend feature could be enabled permanently or only
when needed. This is now reported.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Adds a counter for when the "code detect timer" expired without
finding a valid packet.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Add some comments describing the fields in hdcd_state_t.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
HDCD is now only considered detected if a valid packet
is active in both channels simultaneously.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
There is actually a need for the origin and end point not to be defined.
We can not automatically insert them with the y value of the first and
last point as it will influence the curves in a wrong way.
Fixes#5397
Count and report when a code is signaled but fails to match a known pattern.
For example try Līve - Secret Samadhi.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Only call hdcd_update_info() when the control code changes
instead of every frame, so the counters are more meaningful.
Signed-off-by: Burt P <pburt0@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This is an - once again - updated patch, that uses avio_write instead
of avio_puts to stream clean text output without null characters. Works
now for me as intended.
Changes metadata filter to accept general urls as file argument without
breaking former behaviour. As a byproduct, it also allows for writing to
file "-" if specified as "file:-".
Example:
ffmpeg -i test.wav -filter_complex "silencedetect=n=-40dB:d=0.1,ametadata=mode=print:file='pipe\:4'" -f null
Signed-off-by: Sami Hult <sami.hult@gmail.com>
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The new HDCD filter really does nothing to show that it is working or
that HDCD control information was even detected in the stream. This
patch collects information about the decode, like which features were
used, and reports it to the user at the end.
Also,
* Fixes low-level gain adjustment
* Updates the documentation
Signed-off-by: Burt P <pburt0@gmail.com>
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Even though this is not part of the public API, some external
applications access fields after it, thus breaking after updating from
ffmpeg 3.0 or earlier.
Since it is not public, it can be freely moved to the end to avoid
that problem in the future.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes crash due to unprocessed input being passed through
This fixes the last segfault caused by mixing 3.0 and 3.1 libs and
applications
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
