libswresample/swresample.h
b5875b91
 /*
ba1314c2
  * Copyright (C) 2011-2013 Michael Niedermayer (michaelni@gmx.at)
b5875b91
  *
  * This file is part of libswresample
  *
  * libswresample is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public
  * License as published by the Free Software Foundation; either
  * version 2.1 of the License, or (at your option) any later version.
  *
  * libswresample is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  * Lesser General Public License for more details.
  *
  * You should have received a copy of the GNU Lesser General Public
  * License along with libswresample; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
7cca237d
 #ifndef SWRESAMPLE_SWRESAMPLE_H
 #define SWRESAMPLE_SWRESAMPLE_H
 
68d3144e
 /**
  * @file
5595e7a1
  * @ingroup lswr
68d3144e
  * libswresample public header
  */
 
7cca237d
 /**
58c7bf78
  * @defgroup lswr libswresample
7cca237d
  * @{
  *
58c7bf78
  * Audio resampling, sample format conversion and mixing library.
7cca237d
  *
  * Interaction with lswr is done through SwrContext, which is
  * allocated with swr_alloc() or swr_alloc_set_opts(). It is opaque, so all parameters
  * must be set with the @ref avoptions API.
  *
2711b470
  * The first thing you will need to do in order to use lswr is to allocate
  * SwrContext. This can be done with swr_alloc() or swr_alloc_set_opts(). If you
  * are using the former, you must set options through the @ref avoptions API.
  * The latter function provides the same feature, but it allows you to set some
  * common options in the same statement.
  *
7cca237d
  * For example the following code will setup conversion from planar float sample
  * format to interleaved signed 16-bit integer, downsampling from 48kHz to
  * 44.1kHz and downmixing from 5.1 channels to stereo (using the default mixing
2711b470
  * matrix). This is using the swr_alloc() function.
7cca237d
  * @code
  * SwrContext *swr = swr_alloc();
904c89ac
  * av_opt_set_channel_layout(swr, "in_channel_layout",  AV_CH_LAYOUT_5POINT1, 0);
  * av_opt_set_channel_layout(swr, "out_channel_layout", AV_CH_LAYOUT_STEREO,  0);
7cca237d
  * av_opt_set_int(swr, "in_sample_rate",     48000,                0);
  * av_opt_set_int(swr, "out_sample_rate",    44100,                0);
0212c1c4
  * av_opt_set_sample_fmt(swr, "in_sample_fmt",  AV_SAMPLE_FMT_FLTP, 0);
  * av_opt_set_sample_fmt(swr, "out_sample_fmt", AV_SAMPLE_FMT_S16,  0);
7cca237d
  * @endcode
  *
2711b470
  * The same job can be done using swr_alloc_set_opts() as well:
  * @code
  * SwrContext *swr = swr_alloc_set_opts(NULL,  // we're allocating a new context
  *                       AV_CH_LAYOUT_STEREO,  // out_ch_layout
  *                       AV_SAMPLE_FMT_S16,    // out_sample_fmt
  *                       44100,                // out_sample_rate
  *                       AV_CH_LAYOUT_5POINT1, // in_ch_layout
  *                       AV_SAMPLE_FMT_FLTP,   // in_sample_fmt
  *                       48000,                // in_sample_rate
  *                       0,                    // log_offset
  *                       NULL);                // log_ctx
  * @endcode
  *
7cca237d
  * Once all values have been set, it must be initialized with swr_init(). If
  * you need to change the conversion parameters, you can change the parameters
2711b470
  * using @ref AVOptions, as described above in the first example; or by using
  * swr_alloc_set_opts(), but with the first argument the allocated context.
  * You must then call swr_init() again.
7cca237d
  *
  * The conversion itself is done by repeatedly calling swr_convert().
  * Note that the samples may get buffered in swr if you provide insufficient
  * output space or if sample rate conversion is done, which requires "future"
  * samples. Samples that do not require future input can be retrieved at any
  * time by using swr_convert() (in_count can be set to 0).
  * At the end of conversion the resampling buffer can be flushed by calling
  * swr_convert() with NULL in and 0 in_count.
  *
2711b470
  * The samples used in the conversion process can be managed with the libavutil
  * @ref lavu_sampmanip "samples manipulation" API, including av_samples_alloc()
  * function used in the following example.
  *
7cca237d
  * The delay between input and output, can at any time be found by using
  * swr_get_delay().
  *
  * The following code demonstrates the conversion loop assuming the parameters
  * from above and caller-defined functions get_input() and handle_output():
  * @code
  * uint8_t **input;
  * int in_samples;
  *
  * while (get_input(&input, &in_samples)) {
186942a5
  *     uint8_t *output;
7cca237d
  *     int out_samples = av_rescale_rnd(swr_get_delay(swr, 48000) +
  *                                      in_samples, 44100, 48000, AV_ROUND_UP);
  *     av_samples_alloc(&output, NULL, 2, out_samples,
  *                      AV_SAMPLE_FMT_S16, 0);
  *     out_samples = swr_convert(swr, &output, out_samples,
  *                                      input, in_samples);
  *     handle_output(output, out_samples);
  *     av_freep(&output);
a89e9ad1
  * }
  * @endcode
7cca237d
  *
  * When the conversion is finished, the conversion
  * context and everything associated with it must be freed with swr_free().
2711b470
  * A swr_close() function is also available, but it exists mainly for
  * compatibility with libavresample, and is not required to be called.
  *
7cca237d
  * There will be no memory leak if the data is not completely flushed before
  * swr_free().
  */
b5875b91
 
52b7823b
 #include <stdint.h>
23c0779c
 #include "libavutil/channel_layout.h"
b7d5e016
 #include "libavutil/frame.h"
b5875b91
 #include "libavutil/samplefmt.h"
 
b75c3d2b
 #include "libswresample/version.h"
320b8d63
 
c0d9b026
 /**
  * @name Option constants
  * These constants are used for the @ref avoptions interface for lswr.
  * @{
  *
  */
 
68d3144e
 #define SWR_FLAG_RESAMPLE 1 ///< Force resampling even if equal sample rate
b5875b91
 //TODO use int resample ?
 //long term TODO can we enable this dynamically?
 
fc71434e
 /** Dithering algorithms */
db2eadb2
 enum SwrDitherType {
     SWR_DITHER_NONE = 0,
     SWR_DITHER_RECTANGULAR,
b1ef4dc4
     SWR_DITHER_TRIANGULAR,
f8a237a3
     SWR_DITHER_TRIANGULAR_HIGHPASS,
82742294
 
     SWR_DITHER_NS = 64,         ///< not part of API/ABI
     SWR_DITHER_NS_LIPSHITZ,
     SWR_DITHER_NS_F_WEIGHTED,
     SWR_DITHER_NS_MODIFIED_E_WEIGHTED,
     SWR_DITHER_NS_IMPROVED_E_WEIGHTED,
     SWR_DITHER_NS_SHIBATA,
     SWR_DITHER_NS_LOW_SHIBATA,
     SWR_DITHER_NS_HIGH_SHIBATA,
db2eadb2
     SWR_DITHER_NB,              ///< not part of API/ABI
 };
b5875b91
 
5a5d7074
 /** Resampling Engines */
 enum SwrEngine {
     SWR_ENGINE_SWR,             /**< SW Resampler */
801b3157
     SWR_ENGINE_SOXR,            /**< SoX Resampler */
5a5d7074
     SWR_ENGINE_NB,              ///< not part of API/ABI
 };
 
7e15df75
 /** Resampling Filter Types */
 enum SwrFilterType {
     SWR_FILTER_TYPE_CUBIC,              /**< Cubic */
f3fc103c
     SWR_FILTER_TYPE_BLACKMAN_NUTTALL,   /**< Blackman Nuttall windowed sinc */
     SWR_FILTER_TYPE_KAISER,             /**< Kaiser windowed sinc */
7e15df75
 };
 
c0d9b026
 /**
  * @}
  */
 
77c5f546
 /**
  * The libswresample context. Unlike libavcodec and libavformat, this structure
  * is opaque. This means that if you would like to set options, you must use
  * the @ref avoptions API and cannot directly set values to members of the
  * structure.
  */
1937c4e0
 typedef struct SwrContext SwrContext;
b5875b91
 
 /**
0c583882
  * Get the AVClass for SwrContext. It can be used in combination with
22057e8e
  * AV_OPT_SEARCH_FAKE_OBJ for examining options.
  *
  * @see av_opt_find().
5b58692e
  * @return the AVClass of SwrContext
22057e8e
  */
 const AVClass *swr_get_class(void);
 
 /**
064945b3
  * @name SwrContext constructor functions
  * @{
  */
 
 /**
b5875b91
  * Allocate SwrContext.
68d3144e
  *
  * If you use this function you will need to set the parameters (manually or
  * with swr_alloc_set_opts()) before calling swr_init().
  *
  * @see swr_alloc_set_opts(), swr_init(), swr_free()
  * @return NULL on error, allocated context otherwise
b5875b91
  */
 struct SwrContext *swr_alloc(void);
 
 /**
  * Initialize context after user parameters have been set.
c4ac48c5
  * @note The context must be configured using the AVOption API.
  *
  * @see av_opt_set_int()
  * @see av_opt_set_dict()
68d3144e
  *
5b58692e
  * @param[in,out]   s Swr context to initialize
68d3144e
  * @return AVERROR error code in case of failure.
b5875b91
  */
 int swr_init(struct SwrContext *s);
 
 /**
a66be608
  * Check whether an swr context has been initialized or not.
  *
5b58692e
  * @param[in]       s Swr context to check
  * @see swr_init()
a66be608
  * @return positive if it has been initialized, 0 if not initialized
  */
 int swr_is_initialized(struct SwrContext *s);
 
 /**
68d3144e
  * Allocate SwrContext if needed and set/reset common parameters.
  *
  * This function does not require s to be allocated with swr_alloc(). On the
  * other hand, swr_alloc() can use swr_alloc_set_opts() to set the parameters
  * on the allocated context.
  *
5b58692e
  * @param s               existing Swr context if available, or NULL if not
68d3144e
  * @param out_ch_layout   output channel layout (AV_CH_LAYOUT_*)
58205edc
  * @param out_sample_fmt  output sample format (AV_SAMPLE_FMT_*).
68d3144e
  * @param out_sample_rate output sample rate (frequency in Hz)
  * @param in_ch_layout    input channel layout (AV_CH_LAYOUT_*)
58205edc
  * @param in_sample_fmt   input sample format (AV_SAMPLE_FMT_*).
68d3144e
  * @param in_sample_rate  input sample rate (frequency in Hz)
  * @param log_offset      logging level offset
  * @param log_ctx         parent logging context, can be NULL
  *
  * @see swr_init(), swr_free()
  * @return NULL on error, allocated context otherwise
b5875b91
  */
645c61fb
 struct SwrContext *swr_alloc_set_opts(struct SwrContext *s,
                                       int64_t out_ch_layout, enum AVSampleFormat out_sample_fmt, int out_sample_rate,
                                       int64_t  in_ch_layout, enum AVSampleFormat  in_sample_fmt, int  in_sample_rate,
eafa2b60
                                       int log_offset, void *log_ctx);
b5875b91
 
 /**
064945b3
  * @}
  *
  * @name SwrContext destructor functions
  * @{
  */
 
 /**
68d3144e
  * Free the given SwrContext and set the pointer to NULL.
5b58692e
  *
  * @param[in] s a pointer to a pointer to Swr context
b5875b91
  */
 void swr_free(struct SwrContext **s);
 
 /**
96cb4c87
  * Closes the context so that swr_is_initialized() returns 0.
  *
0c583882
  * The context can be brought back to life by running swr_init(),
96cb4c87
  * swr_init() can also be used without swr_close().
  * This function is mainly provided for simplifying the usecase
0c583882
  * where one tries to support libavresample and libswresample.
5b58692e
  *
  * @param[in,out] s Swr context to be closed
96cb4c87
  */
 void swr_close(struct SwrContext *s);
 
 /**
064945b3
  * @}
  *
  * @name Core conversion functions
  * @{
  */
 
 /** Convert audio.
5cd1580f
  *
68d3144e
  * in and in_count can be set to 0 to flush the last few samples out at the
  * end.
  *
cc17b43d
  * If more input is provided than output space, then the input will be buffered.
  * You can avoid this buffering by using swr_get_out_samples() to retrieve an
  * upper bound on the required number of output samples for the given number of
  * input samples. Conversion will run directly without copying whenever possible.
1b0fcf33
  *
68d3144e
  * @param s         allocated Swr context, with parameters set
  * @param out       output buffers, only the first one need be set in case of packed audio
  * @param out_count amount of space available for output in samples per channel
  * @param in        input buffers, only the first one need to be set in case of packed audio
  * @param in_count  number of input samples available in one channel
  *
7b4e1466
  * @return number of samples output per channel, negative value on error
b5875b91
  */
947215a2
 int swr_convert(struct SwrContext *s, uint8_t **out, int out_count,
                                 const uint8_t **in , int in_count);
b5875b91
 
68d3144e
 /**
72a242c9
  * Convert the next timestamp from input to output
186942a5
  * timestamps are in 1/(in_sample_rate * out_sample_rate) units.
72a242c9
  *
  * @note There are 2 slightly differently behaving modes.
5b58692e
  *       @li When automatic timestamp compensation is not used, (min_compensation >= FLT_MAX)
72a242c9
  *              in this case timestamps will be passed through with delays compensated
5b58692e
  *       @li When automatic timestamp compensation is used, (min_compensation < FLT_MAX)
  *              in this case the output timestamps will match output sample numbers.
  *              See ffmpeg-resampler(1) for the two modes of compensation.
  *
  * @param s[in]     initialized Swr context
  * @param pts[in]   timestamp for the next input sample, INT64_MIN if unknown
  * @see swr_set_compensation(), swr_drop_output(), and swr_inject_silence() are
  *      function used internally for timestamp compensation.
186942a5
  * @return the output timestamp for the next output sample
72a242c9
  */
 int64_t swr_next_pts(struct SwrContext *s, int64_t pts);
 
 /**
064945b3
  * @}
  *
  * @name Low-level option setting functions
  * These functons provide a means to set low-level options that is not possible
  * with the AVOption API.
  * @{
  */
 
 /**
5b58692e
  * Activate resampling compensation ("soft" compensation). This function is
  * internally called when needed in swr_next_pts().
  *
  * @param[in,out] s             allocated Swr context. If it is not initialized,
  *                              or SWR_FLAG_RESAMPLE is not set, swr_init() is
  *                              called with the flag set.
  * @param[in]     sample_delta  delta in PTS per sample
  * @param[in]     compensation_distance number of samples to compensate for
  * @return    >= 0 on success, AVERROR error codes if:
  *            @li @c s is NULL,
  *            @li @c compensation_distance is less than 0,
  *            @li @c compensation_distance is 0 but sample_delta is not,
  *            @li compensation unsupported by resampler, or
  *            @li swr_init() fails when called.
68d3144e
  */
741aca79
 int swr_set_compensation(struct SwrContext *s, int sample_delta, int compensation_distance);
b5875b91
 
eafa2b60
 /**
  * Set a customized input channel mapping.
  *
5b58692e
  * @param[in,out] s           allocated Swr context, not yet initialized
  * @param[in]     channel_map customized input channel mapping (array of channel
  *                            indexes, -1 for a muted channel)
  * @return >= 0 on success, or AVERROR error code in case of failure.
eafa2b60
  */
 int swr_set_channel_mapping(struct SwrContext *s, const int *channel_map);
 
4468ad82
 /**
23c0779c
  * Generate a channel mixing matrix.
  *
  * This function is the one used internally by libswresample for building the
  * default mixing matrix. It is made public just as a utility function for
  * building custom matrices.
  *
  * @param in_layout           input channel layout
  * @param out_layout          output channel layout
  * @param center_mix_level    mix level for the center channel
  * @param surround_mix_level  mix level for the surround channel(s)
  * @param lfe_mix_level       mix level for the low-frequency effects channel
  * @param rematrix_maxval     if 1.0, coefficients will be normalized to prevent
  *                            overflow. if INT_MAX, coefficients will not be
  *                            normalized.
  * @param[out] matrix         mixing coefficients; matrix[i + stride * o] is
  *                            the weight of input channel i in output channel o.
  * @param stride              distance between adjacent input channels in the
  *                            matrix array
  * @param matrix_encoding     matrixed stereo downmix mode (e.g. dplii)
  * @param log_ctx             parent logging context, can be NULL
  * @return                    0 on success, negative AVERROR code on failure
  */
 int swr_build_matrix(uint64_t in_layout, uint64_t out_layout,
                      double center_mix_level, double surround_mix_level,
                      double lfe_mix_level, double rematrix_maxval,
                      double rematrix_volume, double *matrix,
                      int stride, enum AVMatrixEncoding matrix_encoding,
                      void *log_ctx);
 
 /**
560b224f
  * Set a customized remix matrix.
  *
  * @param s       allocated Swr context, not yet initialized
  * @param matrix  remix coefficients; matrix[i + stride * o] is
  *                the weight of input channel i in output channel o
  * @param stride  offset between lines of the matrix
5b58692e
  * @return  >= 0 on success, or AVERROR error code in case of failure.
560b224f
  */
 int swr_set_matrix(struct SwrContext *s, const double *matrix, int stride);
 
 /**
064945b3
  * @}
  *
  * @name Sample handling functions
  * @{
  */
 
 /**
f88f705a
  * Drops the specified number of output samples.
5b58692e
  *
  * This function, along with swr_inject_silence(), is called by swr_next_pts()
  * if needed for "hard" compensation.
  *
  * @param s     allocated Swr context
  * @param count number of samples to be dropped
  *
  * @return >= 0 on success, or a negative AVERROR code on failure
f88f705a
  */
 int swr_drop_output(struct SwrContext *s, int count);
 
 /**
7a59964b
  * Injects the specified number of silence samples.
5b58692e
  *
  * This function, along with swr_drop_output(), is called by swr_next_pts()
  * if needed for "hard" compensation.
  *
  * @param s     allocated Swr context
  * @param count number of samples to be dropped
  *
  * @return >= 0 on success, or a negative AVERROR code on failure
7a59964b
  */
 int swr_inject_silence(struct SwrContext *s, int count);
 
 /**
4def5d2b
  * Gets the delay the next input sample will experience relative to the next output sample.
  *
  * Swresample can buffer data if more input has been provided than available
  * output space, also converting between sample rates needs a delay.
  * This function returns the sum of all such delays.
186942a5
  * The exact delay is not necessarily an integer value in either input or
156a75a4
  * output sample rate. Especially when downsampling by a large value, the
  * output sample rate may be a poor choice to represent the delay, similarly
  * for upsampling and the input sample rate.
4def5d2b
  *
  * @param s     swr context
5b58692e
  * @param base  timebase in which the returned delay will be:
  *              @li if it's set to 1 the returned delay is in seconds
  *              @li if it's set to 1000 the returned delay is in milliseconds
  *              @li if it's set to the input sample rate then the returned
  *                  delay is in input samples
  *              @li if it's set to the output sample rate then the returned
  *                  delay is in output samples
  *              @li if it's the least common multiple of in_sample_rate and
  *                  out_sample_rate then an exact rounding-free delay will be
  *                  returned
  * @returns     the delay in 1 / @c base units.
4def5d2b
  */
 int64_t swr_get_delay(struct SwrContext *s, int64_t base);
 
 /**
cc17b43d
  * Find an upper bound on the number of samples that the next swr_convert
  * call will output, if called with in_samples of input samples. This
  * depends on the internal state, and anything changing the internal state
  * (like further swr_convert() calls) will may change the number of samples
  * swr_get_out_samples() returns for the same number of input samples.
  *
  * @param in_samples    number of input samples.
  * @note any call to swr_inject_silence(), swr_convert(), swr_next_pts()
  *       or swr_set_compensation() invalidates this limit
  * @note it is recommended to pass the correct available buffer size
  *       to all functions like swr_convert() even if swr_get_out_samples()
  *       indicates that less would be used.
  * @returns an upper bound on the number of samples that the next swr_convert
  *          will output or a negative value to indicate an error
  */
 int swr_get_out_samples(struct SwrContext *s, int in_samples);
 
 /**
064945b3
  * @}
  *
  * @name Configuration accessors
  * @{
  */
 
 /**
81f47e27
  * Return the @ref LIBSWRESAMPLE_VERSION_INT constant.
  *
  * This is useful to check if the build-time libswresample has the same version
  * as the run-time one.
  *
  * @returns     the unsigned int-typed version
4468ad82
  */
 unsigned swresample_version(void);
 
 /**
  * Return the swr build-time configuration.
81f47e27
  *
  * @returns     the build-time @c ./configure flags
4468ad82
  */
 const char *swresample_configuration(void);
 
 /**
  * Return the swr license.
81f47e27
  *
  * @returns     the license of libswresample, determined at build-time
4468ad82
  */
 const char *swresample_license(void);
 
7cca237d
 /**
  * @}
b7d5e016
  *
  * @name AVFrame based API
  * @{
  */
 
 /**
  * Convert the samples in the input AVFrame and write them to the output AVFrame.
  *
  * Input and output AVFrames must have channel_layout, sample_rate and format set.
  *
  * If the output AVFrame does not have the data pointers allocated the nb_samples
  * field will be set using av_frame_get_buffer()
  * is called to allocate the frame.
  *
  * The output AVFrame can be NULL or have fewer allocated samples than required.
  * In this case, any remaining samples not written to the output will be added
  * to an internal FIFO buffer, to be returned at the next call to this function
  * or to swr_convert().
  *
  * If converting sample rate, there may be data remaining in the internal
  * resampling delay buffer. swr_get_delay() tells the number of
  * remaining samples. To get this data as output, call this function or
  * swr_convert() with NULL input.
  *
  * If the SwrContext configuration does not match the output and
  * input AVFrame settings the conversion does not take place and depending on
  * which AVFrame is not matching AVERROR_OUTPUT_CHANGED, AVERROR_INPUT_CHANGED
  * or the result of a bitwise-OR of them is returned.
  *
  * @see swr_delay()
  * @see swr_convert()
  * @see swr_get_delay()
  *
  * @param swr             audio resample context
  * @param output          output AVFrame
  * @param input           input AVFrame
  * @return                0 on success, AVERROR on failure or nonmatching
  *                        configuration.
  */
 int swr_convert_frame(SwrContext *swr,
                       AVFrame *output, const AVFrame *input);
 
 /**
  * Configure or reconfigure the SwrContext using the information
  * provided by the AVFrames.
  *
  * The original resampling context is reset even on failure.
  * The function calls swr_close() internally if the context is open.
  *
  * @see swr_close();
  *
  * @param swr             audio resample context
  * @param output          output AVFrame
  * @param input           input AVFrame
  * @return                0 on success, AVERROR on failure.
  */
 int swr_config_frame(SwrContext *swr, const AVFrame *out, const AVFrame *in);
 
 /**
  * @}
064945b3
  * @}
7cca237d
  */
 
7be09a91
 #endif /* SWRESAMPLE_SWRESAMPLE_H */