本文為作者原創，轉載請注明出處：https://www.cnblogs.com/leisure_chn/p/10584925.html
FFmpeg編解碼處理系列筆記：
[0]. FFmpeg時間戳詳解
[1]. FFmpeg編解碼處理1-轉碼全流程簡介
[2]. FFmpeg編解碼處理2-編解碼API詳解
[3]. FFmpeg編解碼處理3-視頻編碼
[4]. FFmpeg編解碼處理4-音頻編碼
基於FFmpeg 4.1版本。
4. 編解碼API詳解
解碼使用 avcodec_send_packet() 和 avcodec_receive_frame() 兩個函數。
編碼使用 avcodec_send_frame() 和 avcodec_receive_packet() 兩個函數。
4.1 API定義
4.1.1 avcodec_send_packet()
/**
* Supply raw packet data as input to a decoder.
*
* Internally, this call will copy relevant AVCodecContext fields, which can
* influence decoding per-packet, and apply them when the packet is actually
* decoded. (For example AVCodecContext.skip_frame, which might direct the
* decoder to drop the frame contained by the packet sent with this function.)
*
* @warning The input buffer, avpkt->data must be AV_INPUT_BUFFER_PADDING_SIZE
* larger than the actual read bytes because some optimized bitstream
* readers read 32 or 64 bits at once and could read over the end.
*
* @warning Do not mix this API with the legacy API (like avcodec_decode_video2())
* on the same AVCodecContext. It will return unexpected results now
* or in future libavcodec versions.
*
* @note The AVCodecContext MUST have been opened with @ref avcodec_open2()
* before packets may be fed to the decoder.
*
* @param avctx codec context
* @param[in] avpkt The input AVPacket. Usually, this will be a single video
* frame, or several complete audio frames.
* Ownership of the packet remains with the caller, and the
* decoder will not write to the packet. The decoder may create
* a reference to the packet data (or copy it if the packet is
* not reference-counted).
* Unlike with older APIs, the packet is always fully consumed,
* and if it contains multiple frames (e.g. some audio codecs),
* will require you to call avcodec_receive_frame() multiple
* times afterwards before you can send a new packet.
* It can be NULL (or an AVPacket with data set to NULL and
* size set to 0); in this case, it is considered a flush
* packet, which signals the end of the stream. Sending the
* first flush packet will return success. Subsequent ones are
* unnecessary and will return AVERROR_EOF. If the decoder
* still has frames buffered, it will return them after sending
* a flush packet.
*
* @return 0 on success, otherwise negative error code:
* AVERROR(EAGAIN): input is not accepted in the current state - user
* must read output with avcodec_receive_frame() (once
* all output is read, the packet should be resent, and
* the call will not fail with EAGAIN).
* AVERROR_EOF: the decoder has been flushed, and no new packets can
* be sent to it (also returned if more than 1 flush
* packet is sent)
* AVERROR(EINVAL): codec not opened, it is an encoder, or requires flush
* AVERROR(ENOMEM): failed to add packet to internal queue, or similar
* other errors: legitimate decoding errors
*/
int avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt);
4.1.2 avcodec_receive_frame()
/**
* Return decoded output data from a decoder.
*
* @param avctx codec context
* @param frame This will be set to a reference-counted video or audio
* frame (depending on the decoder type) allocated by the
* decoder. Note that the function will always call
* av_frame_unref(frame) before doing anything else.
*
* @return
* 0: success, a frame was returned
* AVERROR(EAGAIN): output is not available in this state - user must try
* to send new input
* AVERROR_EOF: the decoder has been fully flushed, and there will be
* no more output frames
* AVERROR(EINVAL): codec not opened, or it is an encoder
* other negative values: legitimate decoding errors
*/
int avcodec_receive_frame(AVCodecContext *avctx, AVFrame *frame);
4.1.3 avcodec_send_frame()
/**
* Supply a raw video or audio frame to the encoder. Use avcodec_receive_packet()
* to retrieve buffered output packets.
*
* @param avctx codec context
* @param[in] frame AVFrame containing the raw audio or video frame to be encoded.
* Ownership of the frame remains with the caller, and the
* encoder will not write to the frame. The encoder may create
* a reference to the frame data (or copy it if the frame is
* not reference-counted).
* It can be NULL, in which case it is considered a flush
* packet. This signals the end of the stream. If the encoder
* still has packets buffered, it will return them after this
* call. Once flushing mode has been entered, additional flush
* packets are ignored, and sending frames will return
* AVERROR_EOF.
*
* For audio:
* If AV_CODEC_CAP_VARIABLE_FRAME_SIZE is set, then each frame
* can have any number of samples.
* If it is not set, frame->nb_samples must be equal to
* avctx->frame_size for all frames except the last.
* The final frame may be smaller than avctx->frame_size.
* @return 0 on success, otherwise negative error code:
* AVERROR(EAGAIN): input is not accepted in the current state - user
* must read output with avcodec_receive_packet() (once
* all output is read, the packet should be resent, and
* the call will not fail with EAGAIN).
* AVERROR_EOF: the encoder has been flushed, and no new frames can
* be sent to it
* AVERROR(EINVAL): codec not opened, refcounted_frames not set, it is a
* decoder, or requires flush
* AVERROR(ENOMEM): failed to add packet to internal queue, or similar
* other errors: legitimate decoding errors
*/
int avcodec_send_frame(AVCodecContext *avctx, const AVFrame *frame);
4.1.4 avcodec_receive_packet()
/**
* Read encoded data from the encoder.
*
* @param avctx codec context
* @param avpkt This will be set to a reference-counted packet allocated by the
* encoder. Note that the function will always call
* av_frame_unref(frame) before doing anything else.
* @return 0 on success, otherwise negative error code:
* AVERROR(EAGAIN): output is not available in the current state - user
* must try to send input
* AVERROR_EOF: the encoder has been fully flushed, and there will be
* no more output packets
* AVERROR(EINVAL): codec not opened, or it is an encoder
* other errors: legitimate decoding errors
*/
int avcodec_receive_packet(AVCodecContext *avctx, AVPacket *avpkt);
4.2 API使用說明
4.2.1 解碼API使用詳解
關於 avcodec_send_packet() 與 avcodec_receive_frame() 的使用說明：
[1] 按 dts 遞增的順序向解碼器送入編碼幀 packet，解碼器按 pts 遞增的順序輸出原始幀 frame，實際上解碼器不關注輸入 packe t的 dts(錯值都沒關系)，它只管依次處理收到的 packet，按需緩沖和解碼
[2] avcodec_receive_frame() 輸出 frame 時，會根據各種因素設置好 frame->best_effort_timestamp(文檔明確說明)，實測 frame->pts 也會被設置(通常直接拷貝自對應的 packet.pts，文檔未明確說明)用戶應確保 avcodec_send_packet() 發送的 packet 具有正確的 pts，編碼幀 packet 與原始幀 frame 間的對應關系通過 pts 確定
[3] avcodec_receive_frame() 輸出 frame 時，frame->pkt_dts 拷貝自當前avcodec_send_packet() 發送的 packet 中的 dts，如果當前 packet 為 NULL(flush packet)，解碼器進入 flush 模式，當前及剩余的 frame->pkt_dts 值總為 AV_NOPTS_VALUE。因為解碼器中有緩存幀，當前輸出的 frame 並不是由當前輸入的 packet 解碼得到的，所以這個 frame->pkt_dts 沒什么實際意義，可以不必關注
[4] avcodec_send_packet() 發送第一個 NULL 會返回成功，后續的 NULL 會返回 AVERROR_EOF
[5] avcodec_send_packet() 多次發送 NULL 並不會導致解碼器中緩存的幀丟失，使用 avcodec_flush_buffers() 可以立即丟掉解碼器中緩存幀。因此播放完畢時應 avcodec_send_packet(NULL) 來取完緩存的幀，而 SEEK 操作或切換流時應調用 avcodec_flush_buffers() 來直接丟棄緩存幀
[6] 解碼器通常的沖洗方法：調用一次 avcodec_send_packet(NULL)(返回成功)，然后不停調用 avcodec_receive_frame() 直到其返回 AVERROR_EOF，取出所有緩存幀，avcodec_receive_frame() 返回 AVERROR_EOF 這一次是沒有有效數據的，僅僅獲取到一個結束標志
4.2.2 編碼API使用詳解
關於 avcodec_send_frame() 與 avcodec_receive_packet() 的使用說明：
[1] 按 pts 遞增的順序向編碼器送入原始幀 frame，編碼器按 dts 遞增的順序輸出編碼幀 packet，實際上編碼器關注輸入 frame 的 pts 不關注其 dts，它只管依次處理收到的 frame，按需緩沖和編碼
[2] avcodec_receive_packet() 輸出 packet 時，會設置 packet.dts，從 0 開始，每次輸出的 packet 的 dts 加 1，這是視頻層的 dts，用戶寫輸出前應將其轉換為容器層的 dts
[3] avcodec_receive_packet() 輸出 packet 時，packet.pts 拷貝自對應的 frame.pts，這是視頻層的 pts，用戶寫輸出前應將其轉換為容器層的 pts
[4] avcodec_send_frame() 發送 NULL frame 時，編碼器進入 flush 模式
[5] avcodec_send_frame() 發送第一個 NULL 會返回成功，后續的 NULL 會返回 AVERROR_EOF
[6] avcodec_send_frame() 多次發送 NULL 並不會導致編碼器中緩存的幀丟失，使用 avcodec_flush_buffers() 可以立即丟掉編碼器中緩存幀。因此編碼完畢時應使用 avcodec_send_frame(NULL) 來取完緩存的幀，而SEEK操作或切換流時應調用 avcodec_flush_buffers() 來直接丟棄緩存幀
[7] 編碼器通常的沖洗方法：調用一次 avcodec_send_frame(NULL)(返回成功)，然后不停調用 avcodec_receive_packet() 直到其返回 AVERROR_EOF，取出所有緩存幀，avcodec_receive_packet() 返回 AVERROR_EOF 這一次是沒有有效數據的，僅僅獲取到一個結束標志
[8] 對音頻來說，如果 AV_CODEC_CAP_VARIABLE_FRAME_SIZE(在 AVCodecContext.codec.capabilities 變量中，只讀)標志有效，表示編碼器支持可變尺寸音頻幀，送入編碼器的音頻幀可以包含任意數量的采樣點。如果此標志無效，則每一個音頻幀的采樣點數目(frame->nb_samples)必須等於編碼器設定的音頻幀尺寸(avctx->frame_size)，最后一幀除外，最后一幀音頻幀采樣點數可以小於 avctx->frame_size
4.3 API使用例程
4.3.1 解碼API例程
// retrun 0: got a frame success
// AVERROR(EAGAIN): need more packet
// AVERROR_EOF: end of file, decoder has been flushed
// <0: error
int av_decode_frame(AVCodecContext *dec_ctx, AVPacket *packet, bool *new_packet, AVFrame *frame)
{
int ret = AVERROR(EAGAIN);
while (1)
{
// 2. 從解碼器接收frame
if (dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO)
{
// 2.1 一個視頻packet含一個視頻frame
// 解碼器緩存一定數量的packet后，才有解碼后的frame輸出
// frame輸出順序是按pts的順序，如IBBPBBP
// frame->pkt_pos變量是此frame對應的packet在視頻文件中的偏移地址，值同pkt.pos
ret = avcodec_receive_frame(dec_ctx, frame);
if (ret >= 0)
{
if (frame->pts == AV_NOPTS_VALUE)
{
frame->pts = frame->best_effort_timestamp;
printf("set video pts %d\n", frame->pts);
}
}
}
else if (dec_ctx->codec_type == AVMEDIA_TYPE_AUDIO)
{
// 2.2 一個音頻packet含一至多個音頻frame，每次avcodec_receive_frame()返回一個frame，此函數返回。
// 下次進來此函數，繼續獲取一個frame，直到avcodec_receive_frame()返回AVERROR(EAGAIN)，
// 表示解碼器需要填入新的音頻packet
ret = avcodec_receive_frame(dec_ctx, frame);
if (ret >= 0)
{
if (frame->pts == AV_NOPTS_VALUE)
{
frame->pts = frame->best_effort_timestamp;
printf("set audio pts %d\n", frame->pts);
}
}
}
if (ret >= 0) // 成功解碼得到一個視頻幀或一個音頻幀，則返回
{
return ret;
}
else if (ret == AVERROR_EOF) // 解碼器已沖洗，解碼中所有幀已取出
{
avcodec_flush_buffers(dec_ctx);
return ret;
}
else if (ret == AVERROR(EAGAIN))// 解碼器需要喂數據
{
if (!(*new_packet)) // 本函數中已向解碼器喂過數據，因此需要從文件讀取新數據
{
//av_log(NULL, AV_LOG_INFO, "decoder need more packet\n");
return ret;
}
}
else // 錯誤
{
av_log(NULL, AV_LOG_ERROR, "decoder error %d\n", ret);
return ret;
}
/*
if (packet == NULL || (packet->data == NULL && packet->size == 0))
{
// 復位解碼器內部狀態/刷新內部緩沖區。當seek操作或切換流時應調用此函數。
avcodec_flush_buffers(dec_ctx);
}
*/
// 1. 將packet發送給解碼器
// 發送packet的順序是按dts遞增的順序，如IPBBPBB
// pkt.pos變量可以標識當前packet在視頻文件中的地址偏移
// 發送第一個 flush packet 會返回成功，后續的 flush packet 會返回AVERROR_EOF
ret = avcodec_send_packet(dec_ctx, packet);
*new_packet = false;
if (ret != 0)
{
av_log(NULL, AV_LOG_ERROR, "avcodec_send_packet() error, return %d\n", ret);
return ret;
}
}
return -1;
}
4.3.2 編碼API例程
int av_encode_frame(AVCodecContext *enc_ctx, AVFrame *frame, AVPacket *packet)
{
int ret = -1;
// 第一次發送flush packet會返回成功，進入沖洗模式，可調用avcodec_receive_packet()
// 將編碼器中緩存的幀(可能不止一個)取出來
// 后續再發送flush packet將返回AVERROR_EOF
ret = avcodec_send_frame(enc_ctx, frame);
if (ret == AVERROR_EOF)
{
//av_log(NULL, AV_LOG_INFO, "avcodec_send_frame() encoder flushed\n");
}
else if (ret == AVERROR(EAGAIN))
{
//av_log(NULL, AV_LOG_INFO, "avcodec_send_frame() need output read out\n");
}
else if (ret < 0)
{
//av_log(NULL, AV_LOG_INFO, "avcodec_send_frame() error %d\n", ret);
return ret;
}
ret = avcodec_receive_packet(enc_ctx, packet);
if (ret == AVERROR_EOF)
{
av_log(NULL, AV_LOG_INFO, "avcodec_recieve_packet() encoder flushed\n");
}
else if (ret == AVERROR(EAGAIN))
{
//av_log(NULL, AV_LOG_INFO, "avcodec_recieve_packet() need more input\n");
}
return ret;
}