lavc: add a decoder option for configuring side data preference

This and the following commits fix #10857

5 files changed, 85 insertions, 1 deletions

diff --git a/doc/APIchanges b/doc/APIchanges
index 650c135b38..cf58c8c5f0 100644
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -2,6 +2,9 @@ The last version increases of all libraries were on 2024-03-07
 
 API changes, most recent first:
 
+2024-03-08 - xxxxxxxxxx - lavc 61.1.100 - avcodec.h
+  Add AVCodecContext.[nb_]side_data_prefer_packet.
+
 2024-03-08 - xxxxxxxxxx - lavu 59.1.100 - opt.h
   Add AV_OPT_TYPE_FLAG_ARRAY and AVOptionArrayDef.
 
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index fb0612ed26..21fc74707f 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -2028,6 +2028,40 @@ typedef struct AVCodecContext {
      *   an error.
      */
     int64_t frame_num;
+
+    /**
+     * Decoding only. May be set by the caller before avcodec_open2() to an
+     * av_malloc()'ed array (or via AVOptions). Owned and freed by the decoder
+     * afterwards.
+     *
+     * Side data attached to decoded frames may come from several sources:
+     * 1. coded_side_data, which the decoder will for certain types translate
+     *    from packet-type to frame-type and attach to frames;
+     * 2. side data attached to an AVPacket sent for decoding (same
+     *    considerations as above);
+     * 3. extracted from the coded bytestream.
+     * The first two cases are supplied by the caller and typically come from a
+     * container.
+     *
+     * This array configures decoder behaviour in cases when side data of the
+     * same type is present both in the coded bytestream and in the
+     * user-supplied side data (items 1. and 2. above). In all cases, at most
+     * one instance of each side data type will be attached to output frames. By
+     * default it will be the bytestream side data. Adding an
+     * AVPacketSideDataType value to this array will flip the preference for
+     * this type, thus making the decoder prefer user-supplied side data over
+     * bytestream. In case side data of the same type is present both in
+     * coded_data and attacked to a packet, the packet instance always has
+     * priority.
+     *
+     * The array may also contain a single -1, in which case the preference is
+     * switched for all side data types.
+     */
+    int        *side_data_prefer_packet;
+    /**
+     * Number of entries in side_data_prefer_packet.
+     */
+    unsigned nb_side_data_prefer_packet;
 } AVCodecContext;
 
 /**
diff --git a/libavcodec/decode.c b/libavcodec/decode.c
index f0260cfa07..0f3eb79b89 100644
--- a/libavcodec/decode.c
+++ b/libavcodec/decode.c
@@ -65,6 +65,12 @@ typedef struct DecodeContext {
     int64_t pts_correction_num_faulty_dts; /// Number of incorrect DTS values so far
     int64_t pts_correction_last_pts;       /// PTS of the last frame
     int64_t pts_correction_last_dts;       /// DTS of the last frame
+
+    /**
+     * Bitmask indicating for which side data types we prefer user-supplied
+     * (global or attached to packets) side data over bytestream.
+     */
+    uint64_t side_data_pref_mask;
 } DecodeContext;
 
 static DecodeContext *decode_ctx(AVCodecInternal *avci)
@@ -1739,6 +1745,35 @@ int ff_decode_preinit(AVCodecContext *avctx)
         avctx->export_side_data |= AV_CODEC_EXPORT_DATA_MVS;
     }
 
+    if (avctx->nb_side_data_prefer_packet == 1 &&
+        avctx->side_data_prefer_packet[0] == -1)
+        dc->side_data_pref_mask = ~0ULL;
+    else {
+        for (unsigned i = 0; i < avctx->nb_side_data_prefer_packet; i++) {
+            int val = avctx->side_data_prefer_packet[i];
+
+            if (val < 0 || val >= AV_PKT_DATA_NB) {
+                av_log(avctx, AV_LOG_ERROR, "Invalid side data type: %d\n", val);
+                return AVERROR(EINVAL);
+            }
+
+            for (unsigned j = 0; j < FF_ARRAY_ELEMS(sd_global_map); j++) {
+                if (sd_global_map[j].packet == val) {
+                    val = sd_global_map[j].frame;
+
+                    // this code will need to be changed when we have more than
+                    // 64 frame side data types
+                    if (val >= 64) {
+                        av_log(avctx, AV_LOG_ERROR, "Side data type too big\n");
+                        return AVERROR_BUG;
+                    }
+
+                    dc->side_data_pref_mask |= 1ULL << val;
+                }
+            }
+        }
+    }
+
     avci->in_pkt         = av_packet_alloc();
     avci->last_pkt_props = av_packet_alloc();
     if (!avci->in_pkt || !avci->last_pkt_props)
diff --git a/libavcodec/options_table.h b/libavcodec/options_table.h
index 7155a335c1..7a2ef3474e 100644
--- a/libavcodec/options_table.h
+++ b/libavcodec/options_table.h
@@ -42,6 +42,8 @@
 #define D AV_OPT_FLAG_DECODING_PARAM
 #define CC AV_OPT_FLAG_CHILD_CONSTS
 
+#define AR AV_OPT_TYPE_FLAG_ARRAY
+
 #define AV_CODEC_DEFAULT_BITRATE 200*1000
 
 static const AVOption avcodec_options[] = {
@@ -395,6 +397,16 @@ static const AVOption avcodec_options[] = {
 {"unsafe_output", "allow potentially unsafe hwaccel frame output that might require special care to process successfully", 0, AV_OPT_TYPE_CONST, {.i64 = AV_HWACCEL_FLAG_UNSAFE_OUTPUT }, INT_MIN, INT_MAX, V | D, .unit = "hwaccel_flags"},
 {"extra_hw_frames", "Number of extra hardware frames to allocate for the user", OFFSET(extra_hw_frames), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, INT_MAX, V|D },
 {"discard_damaged_percentage", "Percentage of damaged samples to discard a frame", OFFSET(discard_damaged_percentage), AV_OPT_TYPE_INT, {.i64 = 95 }, 0, 100, V|D },
+{"side_data_prefer_packet", "Comma-separated list of side data types for which user-supplied (container) data is preferred over coded bytestream",
+    OFFSET(side_data_prefer_packet), AV_OPT_TYPE_INT | AR, .min = -1, .max = INT_MAX, .flags = V|A|S|D, .unit = "side_data_pkt" },
+    {"replaygain",                  .default_val.i64 = AV_PKT_DATA_REPLAYGAIN,                  .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"displaymatrix",               .default_val.i64 = AV_PKT_DATA_DISPLAYMATRIX,               .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"spherical",                   .default_val.i64 = AV_PKT_DATA_SPHERICAL,                   .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"stereo3d",                    .default_val.i64 = AV_PKT_DATA_STEREO3D,                    .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"audio_service_type",          .default_val.i64 = AV_PKT_DATA_AUDIO_SERVICE_TYPE,          .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"mastering_display_metadata",  .default_val.i64 = AV_PKT_DATA_MASTERING_DISPLAY_METADATA,  .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"content_light_level",         .default_val.i64 = AV_PKT_DATA_CONTENT_LIGHT_LEVEL,         .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
+    {"icc_profile",                 .default_val.i64 = AV_PKT_DATA_ICC_PROFILE,                 .type = AV_OPT_TYPE_CONST, .flags = A|D, .unit = "side_data_pkt" },
 {NULL},
 };
 
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 8c3d476003..5957009457 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -29,7 +29,7 @@
 
 #include "version_major.h"
 
-#define LIBAVCODEC_VERSION_MINOR   0
+#define LIBAVCODEC_VERSION_MINOR   1
 #define LIBAVCODEC_VERSION_MICRO 100
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \