avcodec/dump_extradata_bsf: don't allocate a new AVPacket per filtered packet
[ffmpeg.git] / libavcodec / dca.h
index d60b282..e96c589 100644 (file)
@@ -4,6 +4,7 @@
  * Copyright (C) 2004 Benjamin Zores
  * Copyright (C) 2006 Benjamin Larsson
  * Copyright (C) 2007 Konstantin Shishkov
+ * Copyright (C) 2016 foo86
  *
  * This file is part of FFmpeg.
  *
 #define AVCODEC_DCA_H
 
 #include <stdint.h>
-#include "libavutil/internal.h"
 
-/** DCA syncwords, also used for bitstream type detection */
-#define DCA_MARKER_RAW_BE 0x7FFE8001
-#define DCA_MARKER_RAW_LE 0xFE7F0180
-#define DCA_MARKER_14B_BE 0x1FFFE800
-#define DCA_MARKER_14B_LE 0xFF1F00E8
+#include "libavutil/common.h"
+#include "libavutil/intreadwrite.h"
 
-/** DCA-HD specific block starts with this marker. */
-#define DCA_HD_MARKER     0x64582025
+#include "get_bits.h"
+#include "internal.h"
+
+#define DCA_CORE_FRAME_HEADER_SIZE      18
+
+enum DCAParseError {
+    DCA_PARSE_ERROR_SYNC_WORD       = -1,
+    DCA_PARSE_ERROR_DEFICIT_SAMPLES = -2,
+    DCA_PARSE_ERROR_PCM_BLOCKS      = -3,
+    DCA_PARSE_ERROR_FRAME_SIZE      = -4,
+    DCA_PARSE_ERROR_AMODE           = -5,
+    DCA_PARSE_ERROR_SAMPLE_RATE     = -6,
+    DCA_PARSE_ERROR_RESERVED_BIT    = -7,
+    DCA_PARSE_ERROR_LFE_FLAG        = -8,
+    DCA_PARSE_ERROR_PCM_RES         = -9,
+};
+
+typedef struct DCACoreFrameHeader {
+    uint8_t     normal_frame;       ///< Frame type
+    uint8_t     deficit_samples;    ///< Deficit sample count
+    uint8_t     crc_present;        ///< CRC present flag
+    uint8_t     npcmblocks;         ///< Number of PCM sample blocks
+    uint16_t    frame_size;         ///< Primary frame byte size
+    uint8_t     audio_mode;         ///< Audio channel arrangement
+    uint8_t     sr_code;            ///< Core audio sampling frequency
+    uint8_t     br_code;            ///< Transmission bit rate
+    uint8_t     drc_present;        ///< Embedded dynamic range flag
+    uint8_t     ts_present;         ///< Embedded time stamp flag
+    uint8_t     aux_present;        ///< Auxiliary data flag
+    uint8_t     hdcd_master;        ///< HDCD mastering flag
+    uint8_t     ext_audio_type;     ///< Extension audio descriptor flag
+    uint8_t     ext_audio_present;  ///< Extended coding flag
+    uint8_t     sync_ssf;           ///< Audio sync word insertion flag
+    uint8_t     lfe_present;        ///< Low frequency effects flag
+    uint8_t     predictor_history;  ///< Predictor history flag switch
+    uint8_t     filter_perfect;     ///< Multirate interpolator switch
+    uint8_t     encoder_rev;        ///< Encoder software revision
+    uint8_t     copy_hist;          ///< Copy history
+    uint8_t     pcmr_code;          ///< Source PCM resolution
+    uint8_t     sumdiff_front;      ///< Front sum/difference flag
+    uint8_t     sumdiff_surround;   ///< Surround sum/difference flag
+    uint8_t     dn_code;            ///< Dialog normalization / unspecified
+} DCACoreFrameHeader;
+
+enum DCASpeaker {
+    DCA_SPEAKER_C,    DCA_SPEAKER_L,    DCA_SPEAKER_R,    DCA_SPEAKER_Ls,
+    DCA_SPEAKER_Rs,   DCA_SPEAKER_LFE1, DCA_SPEAKER_Cs,   DCA_SPEAKER_Lsr,
+    DCA_SPEAKER_Rsr,  DCA_SPEAKER_Lss,  DCA_SPEAKER_Rss,  DCA_SPEAKER_Lc,
+    DCA_SPEAKER_Rc,   DCA_SPEAKER_Lh,   DCA_SPEAKER_Ch,   DCA_SPEAKER_Rh,
+    DCA_SPEAKER_LFE2, DCA_SPEAKER_Lw,   DCA_SPEAKER_Rw,   DCA_SPEAKER_Oh,
+    DCA_SPEAKER_Lhs,  DCA_SPEAKER_Rhs,  DCA_SPEAKER_Chr,  DCA_SPEAKER_Lhr,
+    DCA_SPEAKER_Rhr,  DCA_SPEAKER_Cl,   DCA_SPEAKER_Ll,   DCA_SPEAKER_Rl,
+    DCA_SPEAKER_RSV1, DCA_SPEAKER_RSV2, DCA_SPEAKER_RSV3, DCA_SPEAKER_RSV4,
+
+    DCA_SPEAKER_COUNT
+};
+
+enum DCASpeakerMask {
+    DCA_SPEAKER_MASK_C     = 0x00000001,
+    DCA_SPEAKER_MASK_L     = 0x00000002,
+    DCA_SPEAKER_MASK_R     = 0x00000004,
+    DCA_SPEAKER_MASK_Ls    = 0x00000008,
+    DCA_SPEAKER_MASK_Rs    = 0x00000010,
+    DCA_SPEAKER_MASK_LFE1  = 0x00000020,
+    DCA_SPEAKER_MASK_Cs    = 0x00000040,
+    DCA_SPEAKER_MASK_Lsr   = 0x00000080,
+    DCA_SPEAKER_MASK_Rsr   = 0x00000100,
+    DCA_SPEAKER_MASK_Lss   = 0x00000200,
+    DCA_SPEAKER_MASK_Rss   = 0x00000400,
+    DCA_SPEAKER_MASK_Lc    = 0x00000800,
+    DCA_SPEAKER_MASK_Rc    = 0x00001000,
+    DCA_SPEAKER_MASK_Lh    = 0x00002000,
+    DCA_SPEAKER_MASK_Ch    = 0x00004000,
+    DCA_SPEAKER_MASK_Rh    = 0x00008000,
+    DCA_SPEAKER_MASK_LFE2  = 0x00010000,
+    DCA_SPEAKER_MASK_Lw    = 0x00020000,
+    DCA_SPEAKER_MASK_Rw    = 0x00040000,
+    DCA_SPEAKER_MASK_Oh    = 0x00080000,
+    DCA_SPEAKER_MASK_Lhs   = 0x00100000,
+    DCA_SPEAKER_MASK_Rhs   = 0x00200000,
+    DCA_SPEAKER_MASK_Chr   = 0x00400000,
+    DCA_SPEAKER_MASK_Lhr   = 0x00800000,
+    DCA_SPEAKER_MASK_Rhr   = 0x01000000,
+    DCA_SPEAKER_MASK_Cl    = 0x02000000,
+    DCA_SPEAKER_MASK_Ll    = 0x04000000,
+    DCA_SPEAKER_MASK_Rl    = 0x08000000,
+};
+
+#define DCA_SPEAKER_LAYOUT_MONO         (DCA_SPEAKER_MASK_C)
+#define DCA_SPEAKER_LAYOUT_STEREO       (DCA_SPEAKER_MASK_L | DCA_SPEAKER_MASK_R)
+#define DCA_SPEAKER_LAYOUT_2POINT1      (DCA_SPEAKER_LAYOUT_STEREO | DCA_SPEAKER_MASK_LFE1)
+#define DCA_SPEAKER_LAYOUT_3_0          (DCA_SPEAKER_LAYOUT_STEREO | DCA_SPEAKER_MASK_C)
+#define DCA_SPEAKER_LAYOUT_2_1          (DCA_SPEAKER_LAYOUT_STEREO | DCA_SPEAKER_MASK_Cs)
+#define DCA_SPEAKER_LAYOUT_3_1          (DCA_SPEAKER_LAYOUT_3_0 | DCA_SPEAKER_MASK_Cs)
+#define DCA_SPEAKER_LAYOUT_2_2          (DCA_SPEAKER_LAYOUT_STEREO | DCA_SPEAKER_MASK_Ls | DCA_SPEAKER_MASK_Rs)
+#define DCA_SPEAKER_LAYOUT_5POINT0      (DCA_SPEAKER_LAYOUT_3_0 | DCA_SPEAKER_MASK_Ls | DCA_SPEAKER_MASK_Rs)
+#define DCA_SPEAKER_LAYOUT_5POINT1      (DCA_SPEAKER_LAYOUT_5POINT0 | DCA_SPEAKER_MASK_LFE1)
+#define DCA_SPEAKER_LAYOUT_7POINT0_WIDE (DCA_SPEAKER_LAYOUT_5POINT0 | DCA_SPEAKER_MASK_Lw | DCA_SPEAKER_MASK_Rw)
+#define DCA_SPEAKER_LAYOUT_7POINT1_WIDE (DCA_SPEAKER_LAYOUT_7POINT0_WIDE | DCA_SPEAKER_MASK_LFE1)
+
+#define DCA_HAS_STEREO(mask) \
+    ((mask & DCA_SPEAKER_LAYOUT_STEREO) == DCA_SPEAKER_LAYOUT_STEREO)
+
+enum DCASpeakerPair {
+    DCA_SPEAKER_PAIR_C      = 0x0001,
+    DCA_SPEAKER_PAIR_LR     = 0x0002,
+    DCA_SPEAKER_PAIR_LsRs   = 0x0004,
+    DCA_SPEAKER_PAIR_LFE1   = 0x0008,
+    DCA_SPEAKER_PAIR_Cs     = 0x0010,
+    DCA_SPEAKER_PAIR_LhRh   = 0x0020,
+    DCA_SPEAKER_PAIR_LsrRsr = 0x0040,
+    DCA_SPEAKER_PAIR_Ch     = 0x0080,
+    DCA_SPEAKER_PAIR_Oh     = 0x0100,
+    DCA_SPEAKER_PAIR_LcRc   = 0x0200,
+    DCA_SPEAKER_PAIR_LwRw   = 0x0400,
+    DCA_SPEAKER_PAIR_LssRss = 0x0800,
+    DCA_SPEAKER_PAIR_LFE2   = 0x1000,
+    DCA_SPEAKER_PAIR_LhsRhs = 0x2000,
+    DCA_SPEAKER_PAIR_Chr    = 0x4000,
+    DCA_SPEAKER_PAIR_LhrRhr = 0x8000
+};
+
+/**
+ * Return number of individual channels in DCASpeakerPair mask
+ */
+static inline int ff_dca_count_chs_for_mask(unsigned int mask)
+{
+    return av_popcount((mask & 0xffff) | ((mask & 0xae66) << 16));
+}
+
+enum DCARepresentationType {
+    DCA_REPR_TYPE_LtRt = 2,
+    DCA_REPR_TYPE_LhRh = 3
+};
+
+enum DCAExtensionMask {
+    DCA_CSS_CORE   = 0x001,
+    DCA_CSS_XXCH   = 0x002,
+    DCA_CSS_X96    = 0x004,
+    DCA_CSS_XCH    = 0x008,
+    DCA_CSS_MASK   = 0x00f,
+    DCA_EXSS_CORE  = 0x010,
+    DCA_EXSS_XBR   = 0x020,
+    DCA_EXSS_XXCH  = 0x040,
+    DCA_EXSS_X96   = 0x080,
+    DCA_EXSS_LBR   = 0x100,
+    DCA_EXSS_XLL   = 0x200,
+    DCA_EXSS_RSV1  = 0x400,
+    DCA_EXSS_RSV2  = 0x800,
+    DCA_EXSS_MASK  = 0xff0,
+};
+
+enum DCADownMixType {
+    DCA_DMIX_TYPE_1_0,
+    DCA_DMIX_TYPE_LoRo,
+    DCA_DMIX_TYPE_LtRt,
+    DCA_DMIX_TYPE_3_0,
+    DCA_DMIX_TYPE_2_1,
+    DCA_DMIX_TYPE_2_2,
+    DCA_DMIX_TYPE_3_1,
+
+    DCA_DMIX_TYPE_COUNT
+};
+
+extern av_export_avcodec const uint32_t avpriv_dca_sample_rates[16];
+
+extern const uint32_t ff_dca_sampling_freqs[16];
+extern const uint8_t ff_dca_freq_ranges[16];
+extern const uint8_t ff_dca_bits_per_sample[8];
 
-extern av_export const uint32_t avpriv_dca_sample_rates[16];
 
 /**
  * Convert bitstream to one representation based on sync marker
  */
-int ff_dca_convert_bitstream(const uint8_t *src, int src_size, uint8_t *dst,
-                             int max_size);
+int avpriv_dca_convert_bitstream(const uint8_t *src, int src_size, uint8_t *dst,
+                                 int max_size);
+
+/**
+ * Parse and validate core frame header
+ * @param[out] h    Pointer to struct where header info is written.
+ * @param[in]  buf  Pointer to the data buffer
+ * @param[in]  size Size of the data buffer
+ * @return 0 on success, negative AVERROR code on failure
+ */
+int avpriv_dca_parse_core_frame_header(DCACoreFrameHeader *h, const uint8_t *buf, int size);
+
+/**
+ * Parse and validate core frame header
+ * @param[out] h   Pointer to struct where header info is written.
+ * @param[in]  gbc BitContext containing the first 120 bits of the frame.
+ * @return 0 on success, negative DCA_PARSE_ERROR_ code on failure
+ */
+int ff_dca_parse_core_frame_header(DCACoreFrameHeader *h, GetBitContext *gb);
 
 #endif /* AVCODEC_DCA_H */