mpeg4video: Add Studio DPCM support

3 files changed, 146 insertions, 20 deletions

diff --git a/libavcodec/mpeg4videodec.c b/libavcodec/mpeg4videodec.c
index 1686ed1001..4c77081237 100644
--- a/libavcodec/mpeg4videodec.c
+++ b/libavcodec/mpeg4videodec.c
@@ -24,6 +24,7 @@
 
 #include "libavutil/internal.h"
 #include "libavutil/opt.h"
+#include "libavutil/pixdesc.h"
 #include "error_resilience.h"
 #include "hwaccel.h"
 #include "idctdsp.h"
@@ -36,6 +37,7 @@
 #include "profiles.h"
 #include "thread.h"
 #include "xvididct.h"
+#include "unary.h"
 
 /* The defines below define the number of bits that are read at once for
  * reading vlc values. Changing these may improve speed and data cache needs
@@ -1923,10 +1925,91 @@ static int mpeg4_decode_studio_block(MpegEncContext *s, int32_t block[64], int n
     return 0;
 }
 
+static int mpeg4_decode_dpcm_macroblock(MpegEncContext *s, int16_t macroblock[256], int n)
+{
+    int i, j, w, h, idx = 0;
+    int block_mean, rice_parameter, rice_prefix_code, rice_suffix_code,
+        dpcm_residual, left, top, topleft, min_left_top, max_left_top, p, p2, output;
+    h = 16 >> (n ? s->chroma_y_shift : 0);
+    w = 16 >> (n ? s->chroma_x_shift : 0);
+
+    block_mean = get_bits(&s->gb, s->avctx->bits_per_raw_sample);
+    if (block_mean == 0){
+        av_log(s->avctx, AV_LOG_ERROR, "Forbidden block_mean\n");
+        return AVERROR_INVALIDDATA;
+    }
+    s->last_dc[n] = block_mean * (1 << (s->dct_precision + s->intra_dc_precision));
+
+    rice_parameter = get_bits(&s->gb, 4);
+    if (rice_parameter == 0) {
+        av_log(s->avctx, AV_LOG_ERROR, "Forbidden rice_parameter\n");
+        return AVERROR_INVALIDDATA;
+    }
+
+    if (rice_parameter == 15)
+        rice_parameter = 0;
+
+    if (rice_parameter > 11) {
+        av_log(s->avctx, AV_LOG_ERROR, "Forbidden rice_parameter\n");
+        return AVERROR_INVALIDDATA;
+    }
+
+    for (i = 0; i < h; i++) {
+        output = 1 << (s->avctx->bits_per_raw_sample - 1);
+        top = 1 << (s->avctx->bits_per_raw_sample - 1);
+
+        for (j = 0; j < w; j++) {
+            left = output;
+            topleft = top;
+
+            rice_prefix_code = get_unary(&s->gb, 1, 12);
+
+            /* Escape */
+            if (rice_prefix_code == 11)
+                dpcm_residual = get_bits(&s->gb, s->avctx->bits_per_raw_sample);
+            else {
+                rice_suffix_code = get_bitsz(&s->gb, rice_parameter);
+                dpcm_residual = (rice_prefix_code << rice_parameter) + rice_suffix_code;
+            }
+
+            /* Map to a signed residual */
+            if (dpcm_residual & 1)
+                dpcm_residual = (-1 * dpcm_residual) >> 1;
+            else
+                dpcm_residual = (dpcm_residual >> 1);
+
+            if (i != 0)
+                top = macroblock[idx-w];
+
+            p = left + top - topleft;
+            min_left_top = FFMIN(left, top);
+            if (p < min_left_top)
+                p = min_left_top;
+
+            max_left_top = FFMAX(left, top);
+            if (p > max_left_top)
+                p = max_left_top;
+
+            p2 = (FFMIN(min_left_top, topleft) + FFMAX(max_left_top, topleft)) >> 1;
+            if (p2 == p)
+                p2 = block_mean;
+
+            if (p2 > p)
+                dpcm_residual *= -1;
+
+            macroblock[idx++] = output = (dpcm_residual + p) & ((1 << s->avctx->bits_per_raw_sample) - 1);
+        }
+    }
+
+    return 0;
+}
+
 static int mpeg4_decode_studio_mb(MpegEncContext *s, int16_t block_[12][64])
 {
     int i;
 
+    s->dpcm_direction = 0;
+
     /* StudioMacroblock */
     /* Assumes I-VOP */
     s->mb_intra = 1;
@@ -1945,9 +2028,11 @@ static int mpeg4_decode_studio_mb(MpegEncContext *s, int16_t block_[12][64])
     } else {
         /* DPCM */
         check_marker(s->avctx, &s->gb, "DPCM block start");
-        avpriv_request_sample(s->avctx, "DPCM encoded block");
-        next_start_code_studio(&s->gb);
-        return SLICE_ERROR;
+        s->dpcm_direction = get_bits1(&s->gb) ? -1 : 1;
+        for (i = 0; i < 3; i++) {
+            if (mpeg4_decode_dpcm_macroblock(s, (*s->dpcm_macroblock)[i], i) < 0)
+                return AVERROR_INVALIDDATA;
+        }
     }
 
     if (get_bits_left(&s->gb) >= 24 && show_bits(&s->gb, 23) == 0) {
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index f75c0fd9b3..d4d3bea649 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -390,6 +390,8 @@ static int init_duplicate_context(MpegEncContext *s)
     }
 
     FF_ALLOCZ_OR_GOTO(s->avctx, s->block32, sizeof(*s->block32), fail)
+    s->dpcm_direction = 0;
+    FF_ALLOCZ_OR_GOTO(s->avctx, s->dpcm_macroblock, sizeof(*s->dpcm_macroblock), fail)
 
     if (s->avctx->codec_tag == AV_RL32("VCR2")) {
         // exchange uv
@@ -427,6 +429,7 @@ static void free_duplicate_context(MpegEncContext *s)
     av_freep(&s->me.score_map);
     av_freep(&s->blocks);
     av_freep(&s->block32);
+    av_freep(&s->dpcm_macroblock);
     av_freep(&s->ac_val_base);
     s->block = NULL;
 }
@@ -445,6 +448,8 @@ static void backup_duplicate_context(MpegEncContext *bak, MpegEncContext *src)
     COPY(blocks);
     COPY(block);
     COPY(block32);
+    COPY(dpcm_macroblock);
+    COPY(dpcm_direction);
     COPY(start_mb_y);
     COPY(end_mb_y);
     COPY(me.map_generation);
@@ -820,6 +825,8 @@ static void clear_context(MpegEncContext *s)
     s->blocks = NULL;
     s->block32 = NULL;
     memset(s->pblocks, 0, sizeof(s->pblocks));
+    s->dpcm_direction = 0;
+    s->dpcm_macroblock = NULL;
     s->ac_val_base = NULL;
     s->ac_val[0] =
     s->ac_val[1] =
@@ -2132,23 +2139,55 @@ void mpv_reconstruct_mb_internal(MpegEncContext *s, int16_t block[12][64],
                TODO: Integrate 10-bit properly into mpegvideo.c so that ER works properly */
             if (s->avctx->bits_per_raw_sample > 8){
                 const int act_block_size = block_size * 2;
-                s->idsp.idct_put(dest_y,                           dct_linesize, (int16_t*)(*s->block32)[0]);
-                s->idsp.idct_put(dest_y              + act_block_size, dct_linesize, (int16_t*)(*s->block32)[1]);
-                s->idsp.idct_put(dest_y + dct_offset,              dct_linesize, (int16_t*)(*s->block32)[2]);
-                s->idsp.idct_put(dest_y + dct_offset + act_block_size, dct_linesize, (int16_t*)(*s->block32)[3]);
-
-                dct_linesize = uvlinesize << s->interlaced_dct;
-                dct_offset   = s->interlaced_dct ? uvlinesize : uvlinesize*block_size;
-
-                s->idsp.idct_put(dest_cb,              dct_linesize, (int16_t*)(*s->block32)[4]);
-                s->idsp.idct_put(dest_cr,              dct_linesize, (int16_t*)(*s->block32)[5]);
-                s->idsp.idct_put(dest_cb + dct_offset, dct_linesize, (int16_t*)(*s->block32)[6]);
-                s->idsp.idct_put(dest_cr + dct_offset, dct_linesize, (int16_t*)(*s->block32)[7]);
-                if(!s->chroma_x_shift){//Chroma444
-                    s->idsp.idct_put(dest_cb + act_block_size,              dct_linesize, (int16_t*)(*s->block32)[8]);
-                    s->idsp.idct_put(dest_cr + act_block_size,              dct_linesize, (int16_t*)(*s->block32)[9]);
-                    s->idsp.idct_put(dest_cb + act_block_size + dct_offset, dct_linesize, (int16_t*)(*s->block32)[10]);
-                    s->idsp.idct_put(dest_cr + act_block_size + dct_offset, dct_linesize, (int16_t*)(*s->block32)[11]);
+
+                if(s->dpcm_direction == 0) {
+                    s->idsp.idct_put(dest_y,                           dct_linesize, (int16_t*)(*s->block32)[0]);
+                    s->idsp.idct_put(dest_y              + act_block_size, dct_linesize, (int16_t*)(*s->block32)[1]);
+                    s->idsp.idct_put(dest_y + dct_offset,              dct_linesize, (int16_t*)(*s->block32)[2]);
+                    s->idsp.idct_put(dest_y + dct_offset + act_block_size, dct_linesize, (int16_t*)(*s->block32)[3]);
+
+                    dct_linesize = uvlinesize << s->interlaced_dct;
+                    dct_offset   = s->interlaced_dct ? uvlinesize : uvlinesize*block_size;
+
+                    s->idsp.idct_put(dest_cb,              dct_linesize, (int16_t*)(*s->block32)[4]);
+                    s->idsp.idct_put(dest_cr,              dct_linesize, (int16_t*)(*s->block32)[5]);
+                    s->idsp.idct_put(dest_cb + dct_offset, dct_linesize, (int16_t*)(*s->block32)[6]);
+                    s->idsp.idct_put(dest_cr + dct_offset, dct_linesize, (int16_t*)(*s->block32)[7]);
+                    if(!s->chroma_x_shift){//Chroma444
+                        s->idsp.idct_put(dest_cb + act_block_size,              dct_linesize, (int16_t*)(*s->block32)[8]);
+                        s->idsp.idct_put(dest_cr + act_block_size,              dct_linesize, (int16_t*)(*s->block32)[9]);
+                        s->idsp.idct_put(dest_cb + act_block_size + dct_offset, dct_linesize, (int16_t*)(*s->block32)[10]);
+                        s->idsp.idct_put(dest_cr + act_block_size + dct_offset, dct_linesize, (int16_t*)(*s->block32)[11]);
+                    }
+                } else if(s->dpcm_direction == 1) {
+                    int i, w, h;
+                    uint16_t *dest_pcm[3] = {(uint16_t*)dest_y, (uint16_t*)dest_cb, (uint16_t*)dest_cr};
+                    int linesize[3] = {dct_linesize, uvlinesize, uvlinesize};
+                    for(i = 0; i < 3; i++) {
+                        int idx = 0;
+                        int vsub = i ? s->chroma_y_shift : 0;
+                        int hsub = i ? s->chroma_x_shift : 0;
+                        for(h = 0; h < (16 >> vsub); h++){
+                            for(w = 0; w < (16 >> hsub); w++)
+                                dest_pcm[i][w] = (*s->dpcm_macroblock)[i][idx++];
+                            dest_pcm[i] += linesize[i] / 2;
+                        }
+                    }
+                } else if(s->dpcm_direction == -1) {
+                    int i, w, h;
+                    uint16_t *dest_pcm[3] = {(uint16_t*)dest_y, (uint16_t*)dest_cb, (uint16_t*)dest_cr};
+                    int linesize[3] = {dct_linesize, uvlinesize, uvlinesize};
+                    for(i = 0; i < 3; i++) {
+                        int idx = 0;
+                        int vsub = i ? s->chroma_y_shift : 0;
+                        int hsub = i ? s->chroma_x_shift : 0;
+                        dest_pcm[i] += (linesize[i] / 2) * ((16 >> vsub) - 1);
+                        for(h = (16 >> vsub)-1; h >= 1; h--){
+                            for(w = (16 >> hsub)-1; w >= 1; w--)
+                                dest_pcm[i][w] = (*s->dpcm_macroblock)[i][idx++];
+                            dest_pcm[i] -= linesize[i] / 2;
+                        }
+                    }
                 }
             }
             /* dct only in intra block */
diff --git a/libavcodec/mpegvideo.h b/libavcodec/mpegvideo.h
index e16deb64e7..bbc6b5646a 100644
--- a/libavcodec/mpegvideo.h
+++ b/libavcodec/mpegvideo.h
@@ -509,6 +509,8 @@ typedef struct MpegEncContext {
     int (*decode_mb)(struct MpegEncContext *s, int16_t block[12][64]); // used by some codecs to avoid a switch()
 
     int32_t (*block32)[12][64];
+    int dpcm_direction;          // 0 = DCT, 1 = DPCM top to bottom scan, -1 = DPCM bottom to top scan
+    int16_t (*dpcm_macroblock)[3][256];
 
 #define SLICE_OK         0
 #define SLICE_ERROR     -1