AVFrame关于视音频数据存储

1、概述

这几天用ffmpeg实现简单音频转码，在写程序的过程中发现音频在AVFrame中存储与视频很相似，音频要复杂一些，本人记性不好，所以在这里记录下来，以作备忘。

2、2个数据成员

在AVFrame中有2个很重要的数据成员，一个是data，一个是linesize。data中存储的是未编码的源始数据（不论视音频），linesize中存储的是每行data中数据大小。

data的定义如下：

/**
     * pointer to the picture/channel planes.
     * This might be different from the first allocated byte
     *
     * Some decoders access areas outside 0,0 - width,height, please
     * see avcodec_align_dimensions2(). Some filters and swscale can read
     * up to 16 bytes beyond the planes, if these filters are to be used,
     * then 16 extra bytes must be allocated.
     */
    uint8_t *data[AV_NUM_DATA_POINTERS];

linesize定义如下：

  /**
     * For video, size in bytes of each picture line.
     * For audio, size in bytes of each plane.
     *
     * For audio, only linesize[0] may be set. For planar audio, each channel
     * plane must be the same size.
     *
     * For video the linesizes should be multiplies of the CPUs alignment
     * preference, this is 16 or 32 for modern desktop CPUs.
     * Some code requires such alignment other code can be slower without
     * correct alignment, for yet other it makes no difference.
     *
     * @note The linesize may be larger than the size of usable data -- there
     * may be extra padding present for performance reasons.
     */
    int linesize[AV_NUM_DATA_POINTERS];

注意：当为音频的时候linesize，只有linesize[0]才是有效值，因为左右一样大。

3、存储方式

1、视频

视频相对简单的多，以yuv420为例，图像数据在AVFrame中存储是这样的：

data[0]存储Y

data[1 ]存储U

data[2 ]存储V

而他们相对应的大小为：

linesize[0]为Y的大小

linesize[1]为U的大小

linesize[2]为V的大小

2、音频

音频数据则要复杂一些，在音频中分为平面和非平面数据类型，下面是音频数据类型的定义：

/**
 * Audio Sample Formats
 *
 * @par
 * The data described by the sample format is always in native-endian order.
 * Sample values can be expressed by native C types, hence the lack of a signed
 * 24-bit sample format even though it is a common raw audio data format.
 *
 * @par
 * The floating-point formats are based on full volume being in the range
 * [-1.0, 1.0]. Any values outside this range are beyond full volume level.
 *
 * @par
 * The data layout as used in av_samples_fill_arrays() and elsewhere in FFmpeg
 * (such as AVFrame in libavcodec) is as follows:
 *
 * For planar sample formats, each audio channel is in a separate data plane,
 * and linesize is the buffer size, in bytes, for a single plane. All data
 * planes must be the same size. For packed sample formats, only the first data
 * plane is used, and samples for each channel are interleaved. In this case,
 * linesize is the buffer size, in bytes, for the 1 plane.
 */
enum AVSampleFormat {
    AV_SAMPLE_FMT_NONE = -1,
    AV_SAMPLE_FMT_U8,          ///< unsigned 8 bits
    AV_SAMPLE_FMT_S16,         ///< signed 16 bits
    AV_SAMPLE_FMT_S32,         ///< signed 32 bits
    AV_SAMPLE_FMT_FLT,         ///< float
    AV_SAMPLE_FMT_DBL,         ///< double

    AV_SAMPLE_FMT_U8P,         ///< unsigned 8 bits, planar
    AV_SAMPLE_FMT_S16P,        ///< signed 16 bits, planar
    AV_SAMPLE_FMT_S32P,        ///< signed 32 bits, planar
    AV_SAMPLE_FMT_FLTP,        ///< float, planar
    AV_SAMPLE_FMT_DBLP,        ///< double, planar

    AV_SAMPLE_FMT_NB           ///< Number of sample formats. DO NOT USE if linking dynamically
};

定义中最后带p的为平面数据类型，可以用av_sample_fmt_is_planar来判断此数据类型是否是平面数据类型。

先说非平面数据：

以一个双声道（左右）音频来说，存储格式可能就为LRLRLR.........（左声道在前还是右声道在前没有认真研究过），数据都装在data[0]中，而大小则为linesize[0]。

平面数据：

就有点像视频部分的YUV数据，同样对双声道音频PCM数据，以S16P为例，存储就可能是
plane 0: LLLLLLLLLLLLLLLLLLLLLLLLLL...
plane 1: RRRRRRRRRRRRRRRRRRRR....

对应的存储则为：

data[0]存储plane 0

data[1]存储plane 1

对应的大小则都为linesize[0]，可以用av_get_bytes_per_sample(out_stream->codec->sample_fmt) * out_frame->nb_samples来算出plane的大小。