最新（2.44）FFmpeg音频播放 ----- 关键点swr_convert

ffmpeg 版本：git clone 于 2014-12-02 ，版本接近2.44，在2.44和2.51之间

SDL版本：SDL 1.2（Centos 6.5软件库的相应版本）

       有些旧的ffmpeg播放音频示例中，会存在一些音频可以播放一些不能播放，其中一个我们需要考虑的原因和该注意的地方就是 av_decode_audiole类似函数所获的的AVFrame的格式是否是我们(SDL)所需要的，本例代码用来解决该问题，关键点在于swr_convert函数，代码及注释如下：

</pre><pre name="code" class="cpp">#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#include <libswscale/swscale.h>
#include <libswresample/swresample.h>
#include <libavutil/avstring.h>
#include <libavutil/pixfmt.h>
#include <libavutil/log.h>
#include <SDL/SDL.h>
#include <SDL/SDL_thread.h>
#include <stdio.h>
#include <math.h>

#define SDL_AUDIO_BUFFER_SIZE 1024 
#define MAX_AUDIOQ_SIZE (1 * 1024 * 1024)
#define FF_ALLOC_EVENT   (SDL_USEREVENT)
#define FF_REFRESH_EVENT (SDL_USEREVENT + 1)
#define FF_QUIT_EVENT (SDL_USEREVENT + 2)

//该字段存在于旧版本的ffmpeg中，此处粘贴过来使用，勿怪！
#define AVCODEC_MAX_AUDIO_FRAME_SIZE 192000 // 1 second of 48khz 32bit audio

typedef struct PacketQueue {
	AVPacketList *first_pkt, *last_pkt;
	int nb_packets;
	int size;
	SDL_mutex *mutex;
	SDL_cond *cond;
} PacketQueue;

typedef struct VideoState {
	char filename[1024];
	AVFormatContext *ic;
	int videoStream, audioStream;
	AVStream *audio_st;
	AVFrame *audio_frame;
	PacketQueue audioq;
	unsigned int audio_buf_size;
	unsigned int audio_buf_index;
	AVPacket audio_pkt;
	uint8_t *audio_pkt_data;
	int audio_pkt_size;
	uint8_t *audio_buf;
	uint8_t *audio_buf1;
	DECLARE_ALIGNED(16,uint8_t,audio_buf2) [AVCODEC_MAX_AUDIO_FRAME_SIZE * 4];
	enum AVSampleFormat audio_src_fmt;
	enum AVSampleFormat audio_tgt_fmt;
	int audio_src_channels;
	int audio_tgt_channels;
	int64_t audio_src_channel_layout;
	int64_t audio_tgt_channel_layout;
	int audio_src_freq;
	int audio_tgt_freq;
	struct SwrContext *swr_ctx;
	SDL_Thread *parse_tid;
	int quit;
} VideoState;

VideoState *global_video_state;

void packet_queue_init(PacketQueue *q) {
	memset(q, 0, sizeof(PacketQueue));
	q->mutex = SDL_CreateMutex();
	q->cond = SDL_CreateCond();
}

int packet_queue_put(PacketQueue *q, AVPacket *pkt) {
	AVPacketList *pkt1;

	pkt1 = (AVPacketList *) av_malloc(sizeof(AVPacketList));
	if (!pkt1) {
		return -1;
	}
	pkt1->pkt = *pkt;
	pkt1->next = NULL;

	SDL_LockMutex(q->mutex);

	if (!q->last_pkt) {
		q->first_pkt = pkt1;
	} else {
		q->last_pkt->next = pkt1;
	}

	q->last_pkt = pkt1;
	q->nb_packets++;
	q->size += pkt1->pkt.size;
	SDL_CondSignal(q->cond);
	SDL_UnlockMutex(q->mutex);
	return 0;
}

static int packet_queue_get(PacketQueue *q, AVPacket *pkt, int block) {
	AVPacketList *pkt1;
	int ret;

	SDL_LockMutex(q->mutex);

	for (;;) {
		if (global_video_state->quit) {
			ret = -1;
			break;
		}

		pkt1 = q->first_pkt;
		if (pkt1) {
			q->first_pkt = pkt1->next;
			if (!q->first_pkt) {
				q->last_pkt = NULL;
			}
			q->nb_packets--;
			q->size -= pkt1->pkt.size;
			*pkt = pkt1->pkt;

			av_free(pkt1);
			ret = 1;
			break;
		} else if (!block) {
			ret = 0;
			break;
		} else {
			SDL_CondWait(q->cond, q->mutex);
		}
	}

	SDL_UnlockMutex(q->mutex);

	return ret;
}

int audio_decode_frame(VideoState *is) {
	int len1, len2, decoded_data_size;
	AVPacket *pkt = &is->audio_pkt;
	int got_frame = 0;
	int64_t dec_channel_layout;
	int wanted_nb_samples, resampled_data_size;

	for (;;) {
		while (is->audio_pkt_size > 0) {
			if (!is->audio_frame) {
				if (!(is->audio_frame = av_frame_alloc())) {
					return AVERROR(ENOMEM);
				}
			} else
				av_frame_unref(is->audio_frame);
			/**
			 * 当AVPacket中装得是音频时，有可能一个AVPacket中有多个AVFrame，
			 * 而某些解码器只会解出第一个AVFrame，这种情况我们必须循环解码出后续AVFrame
			 */
			len1 = avcodec_decode_audio4(is->audio_st->codec, is->audio_frame,
					&got_frame, pkt);
			if (len1 < 0) {
				// error, skip the frame
				is->audio_pkt_size = 0;
				break;
			}

			is->audio_pkt_data += len1;
			is->audio_pkt_size -= len1;

			if (!got_frame)
				continue;
			//执行到这里我们得到了一个AVFrame

			decoded_data_size = av_samples_get_buffer_size(NULL,
					is->audio_frame->channels, is->audio_frame->nb_samples,
					is->audio_frame->format, 1);

			//得到这个AvFrame的声音布局，比如立体声
			dec_channel_layout =
					(is->audio_frame->channel_layout
							&& is->audio_frame->channels
									== av_get_channel_layout_nb_channels(
											is->audio_frame->channel_layout)) ?
							is->audio_frame->channel_layout :
							av_get_default_channel_layout(
									is->audio_frame->channels);

			//这个AVFrame每个声道的采样数
			wanted_nb_samples = is->audio_frame->nb_samples;


			/**
			 * 接下来判断我们之前设置SDL时设置的声音格式(AV_SAMPLE_FMT_S16)，声道布局，
			 * 采样频率，每个AVFrame的每个声道采样数与
			 * 得到的该AVFrame分别是否相同，如有任意不同，我们就需要swr_convert该AvFrame，
			 * 然后才能符合之前设置好的SDL的需要，才能播放
			 */
			if (is->audio_frame->format != is->audio_src_fmt
					|| dec_channel_layout != is->audio_src_channel_layout
					|| is->audio_frame->sample_rate != is->audio_src_freq
					|| (wanted_nb_samples != is->audio_frame->nb_samples
							&& !is->swr_ctx)) {
				if (is->swr_ctx)
					swr_free(&is->swr_ctx);
				is->swr_ctx = swr_alloc_set_opts(NULL,
						is->audio_tgt_channel_layout, is->audio_tgt_fmt,
						is->audio_tgt_freq, dec_channel_layout,
						is->audio_frame->format, is->audio_frame->sample_rate,
						0, NULL);
				if (!is->swr_ctx || swr_init(is->swr_ctx) < 0) {
					fprintf(stderr, "swr_init() failed\n");
					break;
				}
				is->audio_src_channel_layout = dec_channel_layout;
				is->audio_src_channels = is->audio_st->codec->channels;
				is->audio_src_freq = is->audio_st->codec->sample_rate;
				is->audio_src_fmt = is->audio_st->codec->sample_fmt;
			}

			/**
			 * 如果上面if判断失败，就会初始化好swr_ctx，就会如期进行转换
			 */
			if (is->swr_ctx) {
				// const uint8_t *in[] = { is->audio_frame->data[0] };
				const uint8_t **in =
						(const uint8_t **) is->audio_frame->extended_data;
				uint8_t *out[] = { is->audio_buf2 };
				if (wanted_nb_samples != is->audio_frame->nb_samples) {
					fprintf(stdout, "swr_set_compensation \n");
					if (swr_set_compensation(is->swr_ctx,
							(wanted_nb_samples - is->audio_frame->nb_samples)
									* is->audio_tgt_freq
									/ is->audio_frame->sample_rate,
							wanted_nb_samples * is->audio_tgt_freq
									/ is->audio_frame->sample_rate) < 0) {
						fprintf(stderr, "swr_set_compensation() failed\n");
						break;
					}
				}

				/**
				 * 转换该AVFrame到设置好的SDL需要的样子，有些旧的代码示例最主要就是少了这一部分，
				 * 往往一些音频能播，一些不能播，这就是原因，比如有些源文件音频恰巧是AV_SAMPLE_FMT_S16的。
				 * swr_convert 返回的是转换后每个声道(channel)的采样数
				 */
				len2 = swr_convert(is->swr_ctx, out,
						sizeof(is->audio_buf2) / is->audio_tgt_channels
								/ av_get_bytes_per_sample(is->audio_tgt_fmt),
						in, is->audio_frame->nb_samples);
				if (len2 < 0) {
					fprintf(stderr, "swr_convert() failed\n");
					break;
				}
				if (len2
						== sizeof(is->audio_buf2) / is->audio_tgt_channels
								/ av_get_bytes_per_sample(is->audio_tgt_fmt)) {
					fprintf(stderr,
							"warning: audio buffer is probably too small\n");
					swr_init(is->swr_ctx);
				}
				is->audio_buf = is->audio_buf2;

				//每声道采样数 x 声道数 x 每个采样字节数
				resampled_data_size = len2 * is->audio_tgt_channels
						* av_get_bytes_per_sample(is->audio_tgt_fmt);
			} else {
				resampled_data_size = decoded_data_size;
				is->audio_buf = is->audio_frame->data[0];
			}
			// We have data, return it and come back for more later
			return resampled_data_size;
		}

		if (pkt->data)
			av_free_packet(pkt);
		memset(pkt, 0, sizeof(*pkt));
		if (is->quit)
			return -1;
		if (packet_queue_get(&is->audioq, pkt, 1) < 0)
			return -1;

		is->audio_pkt_data = pkt->data;
		is->audio_pkt_size = pkt->size;
	}
}

void audio_callback(void *userdata, Uint8 *stream, int len) {
	VideoState *is = (VideoState *) userdata;
	int len1, audio_data_size;

	while (len > 0) {
		if (is->audio_buf_index >= is->audio_buf_size) {
			audio_data_size = audio_decode_frame(is);

			if (audio_data_size < 0) {
				/* silence */
				is->audio_buf_size = 1024;
				memset(is->audio_buf, 0, is->audio_buf_size);
			} else {
				is->audio_buf_size = audio_data_size;
			}
			is->audio_buf_index = 0;
		}

		len1 = is->audio_buf_size - is->audio_buf_index;
		if (len1 > len) {
			len1 = len;
		}

		memcpy(stream, (uint8_t *) is->audio_buf + is->audio_buf_index, len1);
		len -= len1;
		stream += len1;
		is->audio_buf_index += len1;
	}
}

/**
 * 设置SDL播放声音的参数如声音采样格式，声道布局，静音值
 */
int stream_component_open(VideoState *is, int stream_index) {
	AVFormatContext *ic = is->ic;
	AVCodecContext *codecCtx;
	AVCodec *codec;
	SDL_AudioSpec wanted_spec, spec;
	int64_t wanted_channel_layout = 0;
	int wanted_nb_channels;
	const int next_nb_channels[] = { 0, 0, 1, 6, 2, 6, 4, 6 };

	if (stream_index < 0 || stream_index >= ic->nb_streams) {
		return -1;
	}

	codecCtx = ic->streams[stream_index]->codec;
	wanted_nb_channels = codecCtx->channels;
	if (!wanted_channel_layout
			|| wanted_nb_channels
					!= av_get_channel_layout_nb_channels(
							wanted_channel_layout)) {
		wanted_channel_layout = av_get_default_channel_layout(
				wanted_nb_channels);
		wanted_channel_layout &= ~AV_CH_LAYOUT_STEREO_DOWNMIX;
	}

	wanted_spec.channels = av_get_channel_layout_nb_channels(
			wanted_channel_layout);
	wanted_spec.freq = codecCtx->sample_rate;
	if (wanted_spec.freq <= 0 || wanted_spec.channels <= 0) {
		fprintf(stderr, "Invalid sample rate or channel count!\n");
		return -1;
	}
	wanted_spec.format = AUDIO_S16SYS;
	wanted_spec.silence = 0;
	wanted_spec.samples = SDL_AUDIO_BUFFER_SIZE;
	wanted_spec.callback = audio_callback;
	wanted_spec.userdata = is;

	while (SDL_OpenAudio(&wanted_spec, &spec) < 0) {
		fprintf(stderr, "SDL_OpenAudio (%d channels): %s\n",
				wanted_spec.channels, SDL_GetError());
		wanted_spec.channels = next_nb_channels[FFMIN(7, wanted_spec.channels)];
		if (!wanted_spec.channels) {
			fprintf(stderr,
					"No more channel combinations to tyu, audio open failed\n");
			return -1;
		}
		wanted_channel_layout = av_get_default_channel_layout(
				wanted_spec.channels);
	}

	if (spec.format != AUDIO_S16SYS) {
		fprintf(stderr, "SDL advised audio format %d is not supported!\n",
				spec.format);
		return -1;
	}
	if (spec.channels != wanted_spec.channels) {
		wanted_channel_layout = av_get_default_channel_layout(spec.channels);
		if (!wanted_channel_layout) {
			fprintf(stderr, "SDL advised channel count %d is not supported!\n",
					spec.channels);
			return -1;
		}
	}

	fprintf(stderr, "%d: wanted_spec.format = %d\n", __LINE__,
			wanted_spec.format);
	fprintf(stderr, "%d: wanted_spec.samples = %d\n", __LINE__,
			wanted_spec.samples);
	fprintf(stderr, "%d: wanted_spec.channels = %d\n", __LINE__,
			wanted_spec.channels);
	fprintf(stderr, "%d: wanted_spec.freq = %d\n", __LINE__, wanted_spec.freq);

	fprintf(stderr, "%d: spec.format = %d\n", __LINE__, spec.format);
	fprintf(stderr, "%d: spec.samples = %d\n", __LINE__, spec.samples);
	fprintf(stderr, "%d: spec.channels = %d\n", __LINE__, spec.channels);
	fprintf(stderr, "%d: spec.freq = %d\n", __LINE__, spec.freq);

	is->audio_src_fmt = is->audio_tgt_fmt = AV_SAMPLE_FMT_S16;
	is->audio_src_freq = is->audio_tgt_freq = spec.freq;
	is->audio_src_channel_layout = is->audio_tgt_channel_layout =
			wanted_channel_layout;
	is->audio_src_channels = is->audio_tgt_channels = spec.channels;

	codec = avcodec_find_decoder(codecCtx->codec_id);
	if (!codec || (avcodec_open2(codecCtx, codec, NULL) < 0)) {
		fprintf(stderr, "Unsupported codec!\n");
		return -1;
	}
	ic->streams[stream_index]->discard = AVDISCARD_DEFAULT;
	switch (codecCtx->codec_type) {
	case AVMEDIA_TYPE_AUDIO:
		is->audioStream = stream_index;
		is->audio_st = ic->streams[stream_index];
		is->audio_buf_size = 0;
		is->audio_buf_index = 0;
		memset(&is->audio_pkt, 0, sizeof(is->audio_pkt));
		packet_queue_init(&is->audioq);
		SDL_PauseAudio(0);
		break;
	default:
		break;
	}
}

/**
 * demuxing出AVPacket
 */
static int decode_thread(void *arg) {
	VideoState *is = (VideoState *) arg;
	AVFormatContext *ic = NULL;
	AVPacket pkt1, *packet = &pkt1;
	int ret, i, audio_index = -1;

	is->audioStream = -1;
	global_video_state = is;
	if (avformat_open_input(&ic, is->filename, NULL, NULL) != 0) {
		return -1;
	}
	is->ic = ic;
	if (avformat_find_stream_info(ic, NULL) < 0) {
		return -1;
	}
	av_dump_format(ic, 0, is->filename, 0);
	for (i = 0; i < ic->nb_streams; i++) {
		if (ic->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO
				&& audio_index < 0) {
			audio_index = i;
			break;
		}
	}
	if (audio_index >= 0) {
		stream_component_open(is, audio_index);
	}
	if (is->audioStream < 0) {
		fprintf(stderr, "%s: could not open codecs\n", is->filename);
		goto fail;
	}
	// main decode loop
	for (;;) {
		if (is->quit)
			break;
		if (is->audioq.size > MAX_AUDIOQ_SIZE) {
			SDL_Delay(10);
			continue;
		}
		ret = av_read_frame(is->ic, packet);
		if (ret < 0) {
			if (ret == AVERROR_EOF || url_feof(is->ic->pb)) {
				break;
			}
			if (is->ic->pb && is->ic->pb->error) {
				break;
			}
			continue;
		}

		if (packet->stream_index == is->audioStream) {
			packet_queue_put(&is->audioq, packet);
		} else {
			av_free_packet(packet);
		}
	}

	while (!is->quit) {
		SDL_Delay(100);
	}

	fail: {
		SDL_Event event;
		event.type = FF_QUIT_EVENT;
		event.user.data1 = is;
		SDL_PushEvent(&event);
	}

	return 0;
}

int main(int argc, char *argv[]) {
	SDL_Event event;
	VideoState *is;

	is = (VideoState *) av_mallocz(sizeof(VideoState));

	if (argc < 2) {
		fprintf(stderr, "Usage: test <file>\n");
		exit(1);
	}

	av_register_all();

	if (SDL_Init(SDL_INIT_AUDIO)) {
		fprintf(stderr, "Could not initialize SDL - %s\n", SDL_GetError());
		exit(1);
	}

	av_strlcpy(is->filename, argv[1], sizeof(is->filename));

	is->parse_tid = SDL_CreateThread(decode_thread, is);
	if (!is->parse_tid) {
		av_free(is);
		return -1;
	}

	for (;;) {
		SDL_WaitEvent(&event);
		switch (event.type) {
		case FF_QUIT_EVENT:
		case SDL_QUIT:
			is->quit = 1;
			SDL_Quit();
			exit(0);
			break;
		default:
			break;
		}
	}
	return 0;
}

FFmpeg版本逐渐更新，代码功能更加丰富和易于使用，掌握音视频基础概念结合ffmpeg就可以方便使用！