SDL2 播放音频(MP4)

1.简介

这里引入FFmpeg库，获取音频流数据，然后通过FFmpeg将视频流解码成pcm原始数据，再将pcm数据送入到SDL库中实现音频播放。

2.FFmpeg的操作流程

注册API：av_register_all()
构建输入AVFormatContext上下文：avformat_open_input()
查找音视频流信息：avformat_find_stream_info()
查找解码器：avcodec_find_decoder()
打开解码器：avcodec_open2()
然后通过while循环，不停的读取数据：av_read_frame()
帧解码：avcodec_send_packet()和avcodec_receive_frame()
重采样：swr_convert()

3.SDL音频播放流程

SDL播放音频的流程如下：

初始化音频子系统：SDL_Init()。
设置音频参数：SDL_AudioSpec。
设置回调函数：SDL_AudioCallback。
打开音频设备：SDL_OpenAudio()。
打开pcm文件，读取数据。
开始播放：SDL_PauseAudio()。

4.示例

#include <stdio.h>
#include <SDL.h>
#include <memory>

extern "C"
{
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libswscale/swscale.h"
#include "libavutil/imgutils.h"
#include "libswresample/swresample.h"
};

static  Uint8  *audio_chunk;
static  Uint32  audio_len;
static  Uint8  *audio_pos;

void  fill_audio(void *udata, Uint8 *stream, int len)
{
	//SDL 2.0
	SDL_memset(stream, 0, len);
	if (audio_len == 0)		/*  Only  play  if  we  have  data  left  */
		return;
	len = (len > audio_len ? audio_len : len);	/*  Mix  as  much  data  as  possible  */

	SDL_MixAudio(stream, audio_pos, len, SDL_MIX_MAXVOLUME/2);
	audio_pos += len;
	audio_len -= len;	
}

AVFrame *recv(AVCodecContext *codecCtx)
{
	if (!codecCtx)
	{
		return NULL;
	}

	AVFrame *frame = av_frame_alloc();
	int ret = avcodec_receive_frame(codecCtx, frame);

	if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
	{
		av_frame_free(&frame);
		return NULL;
	}
	else if (ret < 0)
	{
		av_frame_free(&frame);
		return NULL;
	}

	return frame;
}

#undef main
int main(int argc, char* argv[])
{
	av_register_all();

	///ffmpeg
	avformat_network_init();

	AVFormatContext* pFormatCtx = NULL;
	const char* inputUrl = "./2.mp4";

	///打开输入的流
	int ret = avformat_open_input(&pFormatCtx, inputUrl, NULL, NULL);
	if (ret != 0)
	{
		printf("Couldn't open input stream.\n");
		return -1;
	}

	//查找流信息
	if (avformat_find_stream_info(pFormatCtx, NULL) < 0)
	{
		printf("Couldn't find stream information.\n");
		return -1;
	}

	//找到音频流索引
	int audio_index = av_find_best_stream(pFormatCtx, AVMEDIA_TYPE_AUDIO, -1, -1, NULL, 0);
	AVStream* st = pFormatCtx->streams[audio_index];
	AVCodec* codec = nullptr;
	//找到解码器
	codec = avcodec_find_decoder(st->codecpar->codec_id);
	if (!codec)
	{
		fprintf(stderr, "Codec not found\n");
		return -1;
	}

	//申请AVCodecContext
	AVCodecContext* pCodecCtx = avcodec_alloc_context3(codec);
	if (!pCodecCtx)
	{
		return -1;
	}

	avcodec_parameters_to_context(pCodecCtx, pFormatCtx->streams[audio_index]->codecpar);

	//打开解码器
	if ((ret = avcodec_open2(pCodecCtx, codec, NULL) < 0))
	{
		return -1;
	}

	AVPacket* pkt = av_packet_alloc();

	//------------SDL----------------
	//Output Info-----------------------------
	printf("---------------- File Information ---------------\n");
	av_dump_format(pFormatCtx, 0, inputUrl, 0);
	printf("-------------------------------------------------\n");

	SwrContext *swrContext = swr_alloc();
	if (!swrContext)
	{
		return -1;
	}

	swrContext = swr_alloc_set_opts(NULL,													//ctx
		AV_CH_LAYOUT_STEREO,																	//输出channel布局
		AV_SAMPLE_FMT_S16,																		 //输出的采样格式
		44100,																									//采样率
		av_get_default_channel_layout(pCodecCtx->channels),						//输入channel布局
		pCodecCtx->sample_fmt,																	//输入的采样格式
		pCodecCtx->sample_rate,																	//输入的采样率
		0, NULL);

	// 初始化重采样上下文
	if (swr_init(swrContext) < 0)
	{
		swr_free(&swrContext);
		return -1;
	}


	if (SDL_Init(SDL_INIT_VIDEO | SDL_INIT_AUDIO | SDL_INIT_TIMER))
	{
		printf("Could not initialize SDL - %s\n", SDL_GetError());
		return -1;
	}

	//SDL_AudioSpec
	SDL_AudioSpec wanted_spec;
	wanted_spec.freq = 44100;
	wanted_spec.format = AUDIO_S16SYS;
	wanted_spec.channels = 2;
	wanted_spec.silence = 0;
	wanted_spec.samples = 1024;
	wanted_spec.callback = fill_audio;
	wanted_spec.userdata = pCodecCtx;


	if (SDL_OpenAudio(&wanted_spec, NULL) < 0)
	{
		printf("can't open audio.\n");
		return -1;
	}

	//Play
	SDL_PauseAudio(0);

	// 分配输出音频数据
	uint8_t	 *out_buffer = nullptr;
	while (av_read_frame(pFormatCtx, pkt) >= 0)
	{
		if (pkt->stream_index == audio_index)
		{
			//一次send 多次recv
			int ret = avcodec_send_packet(pCodecCtx, pkt);
			if (ret < 0)
				continue;

			//释放资源
			av_packet_unref(pkt);

			while (1)
			{
				AVFrame *frame = recv(pCodecCtx);
				if (!frame)
					break;

				//输入的样本数
				int in_nb_samples = frame->nb_samples;//1024

				int out_linesize;
				int dst_nb_samples = av_rescale_rnd(in_nb_samples, 44100, frame->sample_rate, AV_ROUND_UP);

				//输出的样本数
				int out_buffer_size = av_samples_get_buffer_size(NULL, 2, dst_nb_samples, AV_SAMPLE_FMT_S16, 0);

				if(!out_buffer)
					out_buffer = (uint8_t *)av_malloc(out_buffer_size);

				//返回每个通道输出的样本数，错误时为负值
				int sampleCount = swr_convert(swrContext, &out_buffer, dst_nb_samples,
					(const uint8_t**)frame->data, in_nb_samples);

				if (sampleCount <= 0)
				{
					av_frame_free(&frame);
					break;
				}

				int outSize = sampleCount * 2 * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16);

				while (audio_len > 0)//Wait until finish
					SDL_Delay(1);

				//Set audio buffer (PCM data)
				audio_chunk = (Uint8 *)out_buffer;
				//Audio buffer length
				audio_len = outSize;
				audio_pos = audio_chunk;

				av_frame_free(&frame);
			}
		}
		else
		{
			//释放资源
			av_packet_unref(pkt);
		}
	}

	//--------------
	av_free(out_buffer);
	av_packet_free(&pkt);
	swr_close(swrContext);
	swr_free(&swrContext);
	avcodec_close(pCodecCtx);
	avcodec_free_context(&pCodecCtx);
	avformat_close_input(&pFormatCtx);

	SDL_CloseAudio();
	SDL_Quit();
}