mirror of
https://github.com/FFmpeg/FFmpeg.git
synced 2024-12-07 11:13:41 +02:00
b4f5201967
If one looks at the many query_formats callbacks in existence, one will immediately recognize that there is one type of default callback for video and a slightly different default callback for audio: It is "return ff_set_common_formats_from_list(ctx, pix_fmts);" for video with a filter-specific pix_fmts list. For audio, it is the same with a filter-specific sample_fmts list together with ff_set_common_all_samplerates() and ff_set_common_all_channel_counts(). This commit allows to remove the boilerplate query_formats callbacks by replacing said callback with a union consisting the old callback and pointers for pixel and sample format arrays. For the not uncommon case in which these lists only contain a single entry (besides the sentinel) enum AVPixelFormat and enum AVSampleFormat fields are also added to the union to store them directly in the AVFilter, thereby avoiding a relocation. The state of said union will be contained in a new, dedicated AVFilter field (the nb_inputs and nb_outputs fields have been shrunk to uint8_t in order to create a hole for this new field; this is no problem, as the maximum of all the nb_inputs is four; for nb_outputs it is only two). The state's default value coincides with the earlier default of query_formats being unset, namely that the filter accepts all formats (and also sample rates and channel counts/layouts for audio) provided that these properties agree coincide for all inputs and outputs. By using different union members for audio and video filters the type-unsafety of using the same functions for audio and video lists will furthermore be more confined to formats.c than before. When the new fields are used, they will also avoid allocations: Currently something nearly equivalent to ff_default_query_formats() is called after every successful call to a query_formats callback; yet in the common case that the newly allocated AVFilterFormats are not used at all (namely if there are no free links) these newly allocated AVFilterFormats are freed again without ever being used. Filters no longer using the callback will not exhibit this any more. Reviewed-by: Paul B Mahol <onemda@gmail.com> Reviewed-by: Nicolas George <george@nsup.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
341 lines
11 KiB
C
341 lines
11 KiB
C
/*
|
|
* Copyright (c) 2011 Stefano Sabatini
|
|
* Copyright (c) 2011 Mina Nagy Zaki
|
|
*
|
|
* This file is part of FFmpeg.
|
|
*
|
|
* FFmpeg 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.
|
|
*
|
|
* FFmpeg 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 FFmpeg; if not, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*/
|
|
|
|
/**
|
|
* @file
|
|
* resampling audio filter
|
|
*/
|
|
|
|
#include "libavutil/avstring.h"
|
|
#include "libavutil/channel_layout.h"
|
|
#include "libavutil/opt.h"
|
|
#include "libavutil/samplefmt.h"
|
|
#include "libavutil/avassert.h"
|
|
#include "libswresample/swresample.h"
|
|
#include "avfilter.h"
|
|
#include "audio.h"
|
|
#include "internal.h"
|
|
|
|
typedef struct AResampleContext {
|
|
const AVClass *class;
|
|
int sample_rate_arg;
|
|
double ratio;
|
|
struct SwrContext *swr;
|
|
int64_t next_pts;
|
|
int more_data;
|
|
} AResampleContext;
|
|
|
|
static av_cold int preinit(AVFilterContext *ctx)
|
|
{
|
|
AResampleContext *aresample = ctx->priv;
|
|
|
|
aresample->next_pts = AV_NOPTS_VALUE;
|
|
aresample->swr = swr_alloc();
|
|
if (!aresample->swr)
|
|
return AVERROR(ENOMEM);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static av_cold void uninit(AVFilterContext *ctx)
|
|
{
|
|
AResampleContext *aresample = ctx->priv;
|
|
swr_free(&aresample->swr);
|
|
}
|
|
|
|
static int query_formats(AVFilterContext *ctx)
|
|
{
|
|
AResampleContext *aresample = ctx->priv;
|
|
enum AVSampleFormat out_format;
|
|
int64_t out_rate, out_layout;
|
|
|
|
AVFilterLink *inlink = ctx->inputs[0];
|
|
AVFilterLink *outlink = ctx->outputs[0];
|
|
|
|
AVFilterFormats *in_formats, *out_formats;
|
|
AVFilterFormats *in_samplerates, *out_samplerates;
|
|
AVFilterChannelLayouts *in_layouts, *out_layouts;
|
|
int ret;
|
|
|
|
if (aresample->sample_rate_arg > 0)
|
|
av_opt_set_int(aresample->swr, "osr", aresample->sample_rate_arg, 0);
|
|
av_opt_get_sample_fmt(aresample->swr, "osf", 0, &out_format);
|
|
av_opt_get_int(aresample->swr, "osr", 0, &out_rate);
|
|
av_opt_get_int(aresample->swr, "ocl", 0, &out_layout);
|
|
|
|
in_formats = ff_all_formats(AVMEDIA_TYPE_AUDIO);
|
|
if ((ret = ff_formats_ref(in_formats, &inlink->outcfg.formats)) < 0)
|
|
return ret;
|
|
|
|
in_samplerates = ff_all_samplerates();
|
|
if ((ret = ff_formats_ref(in_samplerates, &inlink->outcfg.samplerates)) < 0)
|
|
return ret;
|
|
|
|
in_layouts = ff_all_channel_counts();
|
|
if ((ret = ff_channel_layouts_ref(in_layouts, &inlink->outcfg.channel_layouts)) < 0)
|
|
return ret;
|
|
|
|
if(out_rate > 0) {
|
|
int ratelist[] = { out_rate, -1 };
|
|
out_samplerates = ff_make_format_list(ratelist);
|
|
} else {
|
|
out_samplerates = ff_all_samplerates();
|
|
}
|
|
|
|
if ((ret = ff_formats_ref(out_samplerates, &outlink->incfg.samplerates)) < 0)
|
|
return ret;
|
|
|
|
if(out_format != AV_SAMPLE_FMT_NONE) {
|
|
int formatlist[] = { out_format, -1 };
|
|
out_formats = ff_make_format_list(formatlist);
|
|
} else
|
|
out_formats = ff_all_formats(AVMEDIA_TYPE_AUDIO);
|
|
if ((ret = ff_formats_ref(out_formats, &outlink->incfg.formats)) < 0)
|
|
return ret;
|
|
|
|
if(out_layout) {
|
|
int64_t layout_list[] = { out_layout, -1 };
|
|
out_layouts = ff_make_format64_list(layout_list);
|
|
} else
|
|
out_layouts = ff_all_channel_counts();
|
|
|
|
return ff_channel_layouts_ref(out_layouts, &outlink->incfg.channel_layouts);
|
|
}
|
|
|
|
|
|
static int config_output(AVFilterLink *outlink)
|
|
{
|
|
int ret;
|
|
AVFilterContext *ctx = outlink->src;
|
|
AVFilterLink *inlink = ctx->inputs[0];
|
|
AResampleContext *aresample = ctx->priv;
|
|
int64_t out_rate, out_layout;
|
|
enum AVSampleFormat out_format;
|
|
char inchl_buf[128], outchl_buf[128];
|
|
|
|
aresample->swr = swr_alloc_set_opts(aresample->swr,
|
|
outlink->channel_layout, outlink->format, outlink->sample_rate,
|
|
inlink->channel_layout, inlink->format, inlink->sample_rate,
|
|
0, ctx);
|
|
if (!aresample->swr)
|
|
return AVERROR(ENOMEM);
|
|
if (!inlink->channel_layout)
|
|
av_opt_set_int(aresample->swr, "ich", inlink->channels, 0);
|
|
if (!outlink->channel_layout)
|
|
av_opt_set_int(aresample->swr, "och", outlink->channels, 0);
|
|
|
|
ret = swr_init(aresample->swr);
|
|
if (ret < 0)
|
|
return ret;
|
|
|
|
av_opt_get_int(aresample->swr, "osr", 0, &out_rate);
|
|
av_opt_get_int(aresample->swr, "ocl", 0, &out_layout);
|
|
av_opt_get_sample_fmt(aresample->swr, "osf", 0, &out_format);
|
|
outlink->time_base = (AVRational) {1, out_rate};
|
|
|
|
av_assert0(outlink->sample_rate == out_rate);
|
|
av_assert0(outlink->channel_layout == out_layout || !outlink->channel_layout);
|
|
av_assert0(outlink->format == out_format);
|
|
|
|
aresample->ratio = (double)outlink->sample_rate / inlink->sample_rate;
|
|
|
|
av_get_channel_layout_string(inchl_buf, sizeof(inchl_buf), inlink ->channels, inlink ->channel_layout);
|
|
av_get_channel_layout_string(outchl_buf, sizeof(outchl_buf), outlink->channels, outlink->channel_layout);
|
|
|
|
av_log(ctx, AV_LOG_VERBOSE, "ch:%d chl:%s fmt:%s r:%dHz -> ch:%d chl:%s fmt:%s r:%dHz\n",
|
|
inlink ->channels, inchl_buf, av_get_sample_fmt_name(inlink->format), inlink->sample_rate,
|
|
outlink->channels, outchl_buf, av_get_sample_fmt_name(outlink->format), outlink->sample_rate);
|
|
return 0;
|
|
}
|
|
|
|
static int filter_frame(AVFilterLink *inlink, AVFrame *insamplesref)
|
|
{
|
|
AResampleContext *aresample = inlink->dst->priv;
|
|
const int n_in = insamplesref->nb_samples;
|
|
int64_t delay;
|
|
int n_out = n_in * aresample->ratio + 32;
|
|
AVFilterLink *const outlink = inlink->dst->outputs[0];
|
|
AVFrame *outsamplesref;
|
|
int ret;
|
|
|
|
delay = swr_get_delay(aresample->swr, outlink->sample_rate);
|
|
if (delay > 0)
|
|
n_out += FFMIN(delay, FFMAX(4096, n_out));
|
|
|
|
outsamplesref = ff_get_audio_buffer(outlink, n_out);
|
|
|
|
if(!outsamplesref) {
|
|
av_frame_free(&insamplesref);
|
|
return AVERROR(ENOMEM);
|
|
}
|
|
|
|
av_frame_copy_props(outsamplesref, insamplesref);
|
|
outsamplesref->format = outlink->format;
|
|
outsamplesref->channels = outlink->channels;
|
|
outsamplesref->channel_layout = outlink->channel_layout;
|
|
outsamplesref->sample_rate = outlink->sample_rate;
|
|
|
|
if(insamplesref->pts != AV_NOPTS_VALUE) {
|
|
int64_t inpts = av_rescale(insamplesref->pts, inlink->time_base.num * (int64_t)outlink->sample_rate * inlink->sample_rate, inlink->time_base.den);
|
|
int64_t outpts= swr_next_pts(aresample->swr, inpts);
|
|
aresample->next_pts =
|
|
outsamplesref->pts = ROUNDED_DIV(outpts, inlink->sample_rate);
|
|
} else {
|
|
outsamplesref->pts = AV_NOPTS_VALUE;
|
|
}
|
|
n_out = swr_convert(aresample->swr, outsamplesref->extended_data, n_out,
|
|
(void *)insamplesref->extended_data, n_in);
|
|
if (n_out <= 0) {
|
|
av_frame_free(&outsamplesref);
|
|
av_frame_free(&insamplesref);
|
|
return 0;
|
|
}
|
|
|
|
aresample->more_data = outsamplesref->nb_samples == n_out; // Indicate that there is probably more data in our buffers
|
|
|
|
outsamplesref->nb_samples = n_out;
|
|
|
|
ret = ff_filter_frame(outlink, outsamplesref);
|
|
av_frame_free(&insamplesref);
|
|
return ret;
|
|
}
|
|
|
|
static int flush_frame(AVFilterLink *outlink, int final, AVFrame **outsamplesref_ret)
|
|
{
|
|
AVFilterContext *ctx = outlink->src;
|
|
AResampleContext *aresample = ctx->priv;
|
|
AVFilterLink *const inlink = outlink->src->inputs[0];
|
|
AVFrame *outsamplesref;
|
|
int n_out = 4096;
|
|
int64_t pts;
|
|
|
|
outsamplesref = ff_get_audio_buffer(outlink, n_out);
|
|
*outsamplesref_ret = outsamplesref;
|
|
if (!outsamplesref)
|
|
return AVERROR(ENOMEM);
|
|
|
|
pts = swr_next_pts(aresample->swr, INT64_MIN);
|
|
pts = ROUNDED_DIV(pts, inlink->sample_rate);
|
|
|
|
n_out = swr_convert(aresample->swr, outsamplesref->extended_data, n_out, final ? NULL : (void*)outsamplesref->extended_data, 0);
|
|
if (n_out <= 0) {
|
|
av_frame_free(&outsamplesref);
|
|
return (n_out == 0) ? AVERROR_EOF : n_out;
|
|
}
|
|
|
|
outsamplesref->sample_rate = outlink->sample_rate;
|
|
outsamplesref->nb_samples = n_out;
|
|
|
|
outsamplesref->pts = pts;
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int request_frame(AVFilterLink *outlink)
|
|
{
|
|
AVFilterContext *ctx = outlink->src;
|
|
AResampleContext *aresample = ctx->priv;
|
|
int ret;
|
|
|
|
// First try to get data from the internal buffers
|
|
if (aresample->more_data) {
|
|
AVFrame *outsamplesref;
|
|
|
|
if (flush_frame(outlink, 0, &outsamplesref) >= 0) {
|
|
return ff_filter_frame(outlink, outsamplesref);
|
|
}
|
|
}
|
|
aresample->more_data = 0;
|
|
|
|
// Second request more data from the input
|
|
ret = ff_request_frame(ctx->inputs[0]);
|
|
|
|
// Third if we hit the end flush
|
|
if (ret == AVERROR_EOF) {
|
|
AVFrame *outsamplesref;
|
|
|
|
if ((ret = flush_frame(outlink, 1, &outsamplesref)) < 0)
|
|
return ret;
|
|
|
|
return ff_filter_frame(outlink, outsamplesref);
|
|
}
|
|
return ret;
|
|
}
|
|
|
|
static const AVClass *resample_child_class_iterate(void **iter)
|
|
{
|
|
const AVClass *c = *iter ? NULL : swr_get_class();
|
|
*iter = (void*)(uintptr_t)c;
|
|
return c;
|
|
}
|
|
|
|
static void *resample_child_next(void *obj, void *prev)
|
|
{
|
|
AResampleContext *s = obj;
|
|
return prev ? NULL : s->swr;
|
|
}
|
|
|
|
#define OFFSET(x) offsetof(AResampleContext, x)
|
|
#define FLAGS AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
|
|
|
|
static const AVOption options[] = {
|
|
{"sample_rate", NULL, OFFSET(sample_rate_arg), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, FLAGS },
|
|
{NULL}
|
|
};
|
|
|
|
static const AVClass aresample_class = {
|
|
.class_name = "aresample",
|
|
.item_name = av_default_item_name,
|
|
.option = options,
|
|
.version = LIBAVUTIL_VERSION_INT,
|
|
.child_class_iterate = resample_child_class_iterate,
|
|
.child_next = resample_child_next,
|
|
};
|
|
|
|
static const AVFilterPad aresample_inputs[] = {
|
|
{
|
|
.name = "default",
|
|
.type = AVMEDIA_TYPE_AUDIO,
|
|
.filter_frame = filter_frame,
|
|
},
|
|
};
|
|
|
|
static const AVFilterPad aresample_outputs[] = {
|
|
{
|
|
.name = "default",
|
|
.config_props = config_output,
|
|
.request_frame = request_frame,
|
|
.type = AVMEDIA_TYPE_AUDIO,
|
|
},
|
|
};
|
|
|
|
const AVFilter ff_af_aresample = {
|
|
.name = "aresample",
|
|
.description = NULL_IF_CONFIG_SMALL("Resample audio data."),
|
|
.preinit = preinit,
|
|
.uninit = uninit,
|
|
.priv_size = sizeof(AResampleContext),
|
|
.priv_class = &aresample_class,
|
|
FILTER_INPUTS(aresample_inputs),
|
|
FILTER_OUTPUTS(aresample_outputs),
|
|
FILTER_QUERY_FUNC(query_formats),
|
|
};
|