1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
FFmpeg/libavfilter/af_ashowinfo.c
Anton Khirnov 565e4993c6 lavfi: merge start_frame/draw_slice/end_frame
Any alleged performance benefits gained from the split are purely
mythological and do not justify added code complexity.
2012-11-28 08:50:19 +01:00

138 lines
4.1 KiB
C

/*
* Copyright (c) 2011 Stefano Sabatini
*
* This file is part of Libav.
*
* Libav 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.
*
* Libav 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 Libav; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/**
* @file
* filter for showing textual audio frame information
*/
#include <inttypes.h>
#include <stddef.h>
#include "libavutil/adler32.h"
#include "libavutil/channel_layout.h"
#include "libavutil/common.h"
#include "libavutil/mem.h"
#include "libavutil/samplefmt.h"
#include "audio.h"
#include "avfilter.h"
#include "internal.h"
typedef struct AShowInfoContext {
/**
* Scratch space for individual plane checksums for planar audio
*/
uint32_t *plane_checksums;
/**
* Frame counter
*/
uint64_t frame;
} AShowInfoContext;
static int config_input(AVFilterLink *inlink)
{
AShowInfoContext *s = inlink->dst->priv;
int channels = av_get_channel_layout_nb_channels(inlink->channel_layout);
s->plane_checksums = av_malloc(channels * sizeof(*s->plane_checksums));
if (!s->plane_checksums)
return AVERROR(ENOMEM);
return 0;
}
static void uninit(AVFilterContext *ctx)
{
AShowInfoContext *s = ctx->priv;
av_freep(&s->plane_checksums);
}
static int filter_frame(AVFilterLink *inlink, AVFilterBufferRef *buf)
{
AVFilterContext *ctx = inlink->dst;
AShowInfoContext *s = ctx->priv;
char chlayout_str[128];
uint32_t checksum = 0;
int channels = av_get_channel_layout_nb_channels(buf->audio->channel_layout);
int planar = av_sample_fmt_is_planar(buf->format);
int block_align = av_get_bytes_per_sample(buf->format) * (planar ? 1 : channels);
int data_size = buf->audio->nb_samples * block_align;
int planes = planar ? channels : 1;
int i;
for (i = 0; i < planes; i++) {
uint8_t *data = buf->extended_data[i];
s->plane_checksums[i] = av_adler32_update(0, data, data_size);
checksum = i ? av_adler32_update(checksum, data, data_size) :
s->plane_checksums[0];
}
av_get_channel_layout_string(chlayout_str, sizeof(chlayout_str), -1,
buf->audio->channel_layout);
av_log(ctx, AV_LOG_INFO,
"n:%"PRIu64" pts:%"PRId64" pts_time:%f "
"fmt:%s chlayout:%s rate:%d nb_samples:%d "
"checksum:%08X ",
s->frame, buf->pts, buf->pts * av_q2d(inlink->time_base),
av_get_sample_fmt_name(buf->format), chlayout_str,
buf->audio->sample_rate, buf->audio->nb_samples,
checksum);
av_log(ctx, AV_LOG_INFO, "plane_checksums: [ ");
for (i = 0; i < planes; i++)
av_log(ctx, AV_LOG_INFO, "%08X ", s->plane_checksums[i]);
av_log(ctx, AV_LOG_INFO, "]\n");
s->frame++;
return ff_filter_frame(inlink->dst->outputs[0], buf);
}
static const AVFilterPad inputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.get_audio_buffer = ff_null_get_audio_buffer,
.config_props = config_input,
.filter_frame = filter_frame,
.min_perms = AV_PERM_READ,
},
{ NULL },
};
static const AVFilterPad outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
},
{ NULL },
};
AVFilter avfilter_af_ashowinfo = {
.name = "ashowinfo",
.description = NULL_IF_CONFIG_SMALL("Show textual information for each audio frame."),
.priv_size = sizeof(AShowInfoContext),
.uninit = uninit,
.inputs = inputs,
.outputs = outputs,
};