[FFmpeg-devel] [PATCH 1/4 v2] avformat/mov: ignore item boxes for animated heif

[FFmpeg-devel] [PATCH 1/4 v2] avformat/mov: ignore item boxes for animated heif

[James Almer]

Fixes a regression since d9fed9df2a, where the single animated stream would
be exported twice as two independent streams.

Signed-off-by: James Almer <jamrial@gmail.com>
---
 libavformat/isom.h |   1 +
 libavformat/mov.c  | 147 ++++++++++++++++++++++++++++++---------------
 2 files changed, 99 insertions(+), 49 deletions(-)

diff --git a/libavformat/isom.h b/libavformat/isom.h
index 2cf456fee1..21caaac256 100644
--- a/libavformat/isom.h
+++ b/libavformat/isom.h
@@ -280,6 +280,7 @@ typedef struct MOVContext {
     int64_t duration;     ///< duration of the longest track
     int found_moov;       ///< 'moov' atom has been found
     int found_iloc;       ///< 'iloc' atom has been found
+    int found_iinf;       ///< 'iinf' atom has been found
     int found_mdat;       ///< 'mdat' atom has been found
     int found_hdlr_mdta;  ///< 'hdlr' atom with type 'mdta' has been found
     int trak_index;       ///< Index of the current 'trak'
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 4cffd6c7db..03ca62895e 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -81,6 +81,7 @@ typedef struct MOVParseTableEntry {
 
 static int mov_read_default(MOVContext *c, AVIOContext *pb, MOVAtom atom);
 static int mov_read_mfra(MOVContext *c, AVIOContext *f);
+static void mov_free_stream_context(AVFormatContext *s, AVStream *st);
 static int64_t add_ctts_entry(MOVCtts** ctts_data, unsigned int* ctts_count, unsigned int* allocated_size,
                               int count, int duration);
 
@@ -4639,6 +4640,23 @@ static int mov_read_trak(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     MOVStreamContext *sc;
     int ret;
 
+    if (c->found_iinf) {
+        // * For animated heif, if the iinf box showed up before the moov
+        //   box, we need to clear all the streams read in the former.
+        for (int i = c->heif_info_size - 1; i >= 0; i--) {
+            HEIFItem *item = &c->heif_info[i];
+
+            if (!item->st)
+                continue;
+
+            mov_free_stream_context(c->fc, item->st);
+            ff_remove_stream(c->fc, item->st);
+        }
+        av_freep(&c->heif_info);
+        c->heif_info_size = 0;
+        c->found_iinf = c->found_iloc = 0;
+    }
+
     st = avformat_new_stream(c->fc, NULL);
     if (!st) return AVERROR(ENOMEM);
     st->id = -1;
@@ -7768,8 +7786,9 @@ static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     uint64_t base_offset, extent_offset, extent_length;
     uint8_t value;
 
-    if (c->found_iloc) {
-        av_log(c->fc, AV_LOG_INFO, "Duplicate iloc box found\n");
+    if (c->found_moov) {
+        // * For animated heif, we don't care about the iloc box as all the
+        //   necessary information can be found in the moov box.
         return 0;
     }
 
@@ -7891,6 +7910,16 @@ static int mov_read_iinf(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     int entry_count;
     int version, ret;
 
+    if (c->found_iinf) {
+        av_log(c->fc, AV_LOG_WARNING, "Duplicate iinf box found\n");
+        return 0;
+    }
+    if (c->found_moov) {
+        // * For animated heif, we don't care about the iinf box as all the
+        //   necessary information can be found in the moov box.
+        return 0;
+    }
+
     version = avio_r8(pb);
     avio_rb24(pb);  // flags.
     entry_count = version ? avio_rb32(pb) : avio_rb16(pb);
@@ -7914,6 +7943,7 @@ static int mov_read_iinf(MOVContext *c, AVIOContext *pb, MOVAtom atom)
             return ret;
     }
 
+    c->found_iinf = 1;
     return 0;
 }
 
@@ -7927,6 +7957,13 @@ static int mov_read_iref(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 static int mov_read_ispe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 {
     uint32_t width, height;
+
+    if (c->found_moov) {
+        // * For animated heif, we don't care about the ispe box as all the
+        //   necessary information can be found in the moov box.
+        return 0;
+    }
+
     avio_r8(pb);  /* version */
     avio_rb24(pb);  /* flags */
     width  = avio_rb32(pb);
@@ -7961,6 +7998,12 @@ static int mov_read_iprp(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     int version, flags;
     int ret;
 
+    if (c->found_moov) {
+        // * For animated heif, we don't care about the iprp box as all the
+        //   necessary information can be found in the moov box.
+        return 0;
+    }
+
     a.size = avio_rb32(pb);
     a.type = avio_rl32(pb);
 
@@ -8581,6 +8624,58 @@ static void mov_free_encryption_index(MOVEncryptionIndex **index) {
     av_freep(index);
 }
 
+static void mov_free_stream_context(AVFormatContext *s, AVStream *st)
+{
+    MOVStreamContext *sc = st->priv_data;
+
+    if (!sc)
+        return;
+
+    av_freep(&sc->ctts_data);
+    for (int i = 0; i < sc->drefs_count; i++) {
+        av_freep(&sc->drefs[i].path);
+        av_freep(&sc->drefs[i].dir);
+    }
+    av_freep(&sc->drefs);
+
+    sc->drefs_count = 0;
+
+    if (!sc->pb_is_copied)
+        ff_format_io_close(s, &sc->pb);
+
+    sc->pb = NULL;
+    av_freep(&sc->chunk_offsets);
+    av_freep(&sc->stsc_data);
+    av_freep(&sc->sample_sizes);
+    av_freep(&sc->keyframes);
+    av_freep(&sc->stts_data);
+    av_freep(&sc->sdtp_data);
+    av_freep(&sc->stps_data);
+    av_freep(&sc->elst_data);
+    av_freep(&sc->rap_group);
+    av_freep(&sc->sync_group);
+    av_freep(&sc->sgpd_sync);
+    av_freep(&sc->sample_offsets);
+    av_freep(&sc->open_key_samples);
+    av_freep(&sc->display_matrix);
+    av_freep(&sc->index_ranges);
+
+    if (sc->extradata)
+        for (int i = 0; i < sc->stsd_count; i++)
+            av_free(sc->extradata[i]);
+    av_freep(&sc->extradata);
+    av_freep(&sc->extradata_size);
+
+    mov_free_encryption_index(&sc->cenc.encryption_index);
+    av_encryption_info_free(sc->cenc.default_encrypted_sample);
+    av_aes_ctr_free(sc->cenc.aes_ctr);
+
+    av_freep(&sc->stereo3d);
+    av_freep(&sc->spherical);
+    av_freep(&sc->mastering);
+    av_freep(&sc->coll);
+}
+
 static int mov_read_close(AVFormatContext *s)
 {
     MOVContext *mov = s->priv_data;
@@ -8588,54 +8683,8 @@ static int mov_read_close(AVFormatContext *s)
 
     for (i = 0; i < s->nb_streams; i++) {
         AVStream *st = s->streams[i];
-        MOVStreamContext *sc = st->priv_data;
-
-        if (!sc)
-            continue;
-
-        av_freep(&sc->ctts_data);
-        for (j = 0; j < sc->drefs_count; j++) {
-            av_freep(&sc->drefs[j].path);
-            av_freep(&sc->drefs[j].dir);
-        }
-        av_freep(&sc->drefs);
-
-        sc->drefs_count = 0;
 
-        if (!sc->pb_is_copied)
-            ff_format_io_close(s, &sc->pb);
-
-        sc->pb = NULL;
-        av_freep(&sc->chunk_offsets);
-        av_freep(&sc->stsc_data);
-        av_freep(&sc->sample_sizes);
-        av_freep(&sc->keyframes);
-        av_freep(&sc->stts_data);
-        av_freep(&sc->sdtp_data);
-        av_freep(&sc->stps_data);
-        av_freep(&sc->elst_data);
-        av_freep(&sc->rap_group);
-        av_freep(&sc->sync_group);
-        av_freep(&sc->sgpd_sync);
-        av_freep(&sc->sample_offsets);
-        av_freep(&sc->open_key_samples);
-        av_freep(&sc->display_matrix);
-        av_freep(&sc->index_ranges);
-
-        if (sc->extradata)
-            for (j = 0; j < sc->stsd_count; j++)
-                av_free(sc->extradata[j]);
-        av_freep(&sc->extradata);
-        av_freep(&sc->extradata_size);
-
-        mov_free_encryption_index(&sc->cenc.encryption_index);
-        av_encryption_info_free(sc->cenc.default_encrypted_sample);
-        av_aes_ctr_free(sc->cenc.aes_ctr);
-
-        av_freep(&sc->stereo3d);
-        av_freep(&sc->spherical);
-        av_freep(&sc->mastering);
-        av_freep(&sc->coll);
+        mov_free_stream_context(s, st);
     }
 
     av_freep(&mov->dv_demux);
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH 2/4 v4] avformat: add a Tile Grid stream group type

[James Almer]

This will be used to support tiled image formats like HEIF.

Signed-off-by: James Almer <jamrial@gmail.com>
---
 libavformat/avformat.c |  8 ++++
 libavformat/avformat.h | 86 ++++++++++++++++++++++++++++++++++++++++++
 libavformat/dump.c     | 36 ++++++++++++++++++
 libavformat/options.c  | 32 ++++++++++++++++
 4 files changed, 162 insertions(+)

diff --git a/libavformat/avformat.c b/libavformat/avformat.c
index 882927f7b1..58dfa0746a 100644
--- a/libavformat/avformat.c
+++ b/libavformat/avformat.c
@@ -100,6 +100,14 @@ void ff_free_stream_group(AVStreamGroup **pstg)
         av_iamf_mix_presentation_free(&stg->params.iamf_mix_presentation);
         break;
     }
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID: {
+        if (stg->params.tile_grid) {
+            av_freep(&stg->params.tile_grid->tile_width);
+            av_freep(&stg->params.tile_grid->tile_height);
+        }
+        av_freep(&stg->params.tile_grid);
+        break;
+    }
     default:
         break;
     }
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index 5d0fe82250..ab9a3fc6be 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -1022,11 +1022,96 @@ enum AVStreamGroupParamsType {
     AV_STREAM_GROUP_PARAMS_NONE,
     AV_STREAM_GROUP_PARAMS_IAMF_AUDIO_ELEMENT,
     AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION,
+    AV_STREAM_GROUP_PARAMS_TILE_GRID,
 };
 
 struct AVIAMFAudioElement;
 struct AVIAMFMixPresentation;
 
+/**
+ * AVTileGrid holds information on how to combine several independent images in
+ * a single grid for presentation.
+ *
+ * Its size is not a part of the ABI. No new fields may be added to this struct
+ * without a major version bump.
+ */
+typedef struct AVTileGrid {
+    const AVClass *av_class;
+
+    /**
+     * Amount of rows in the grid.
+     *
+     * Must be > 0.
+     */
+    int tile_rows;
+    /**
+     * Amount of columns in the grid.
+     *
+     * Must be > 0.
+     */
+    int tile_cols;
+
+    /**
+     * A @ref tile_rows * @ref tile_cols sized array of width values for each
+     * tile in the grid, in row major order.
+     * The sum of tile width values must be consistent across all rows.
+     *
+     * Must be allocated with the av_malloc() family of functions, and will be
+     * freed by av_tile_grid_free().
+     */
+    int *tile_width;
+
+    /**
+     * A @ref tile_rows * @ref tile_cols sized array of height values for each
+     * tile in the grid, in row major order.
+     * The sum of tile height values must be consistent across all columns.
+     *
+     * Must be allocated with the av_malloc() family of functions, and will be
+     * freed by av_tile_grid_free().
+     */
+    int *tile_height;
+
+    /**
+     * Offset in pixels from the left edge of the grid where the actual image
+     * meant for presentation starts.
+     *
+     * This field must be >= 0 and <= the sum of values for a row in the
+     * tile_width array minus @ref output_width.
+     */
+    int horizontal_offset;
+    /**
+     * Offset in pixels from the top edge of the grid where the actual image meant
+     * for presentation starts.
+     *
+     * This field must be >= 0 and <= the sum of values for a column in the
+     * tile_height array minus @ref output_height.
+     */
+    int vertical_offset;
+
+    /**
+     * Width of the final image for presentation.
+     *
+     * Must be > 0 and <= the sum of values for a row in the tile_width array
+     * minus @ref horizontal_offset.
+     * When it's not equal to the sum of values for a row in the tile_width array,
+     * the result of said sum minus output_width minus @ref horizontal_offset is
+     * the amount of pixels to be cropped from the right edge of the final image
+     * before presentation.
+     */
+    int output_width;
+    /**
+     * Height of the final image for presentation.
+     *
+     * Must be > 0 and <= the sum of values for a column in the tile_height array
+     * minus @ref vertical_offset.
+     * When it's not equal to the sum of values for a column in the tile_height,
+     * the result of said sum minus output_height minus @ref vertical_offset is
+     * the amount of pixels to be cropped from the bottom edge of the final image
+     * before presentation.
+     */
+    int output_height;
+} AVTileGrid;
+
 typedef struct AVStreamGroup {
     /**
      * A class for @ref avoptions. Set by avformat_stream_group_create().
@@ -1062,6 +1147,7 @@ typedef struct AVStreamGroup {
     union {
         struct AVIAMFAudioElement *iamf_audio_element;
         struct AVIAMFMixPresentation *iamf_mix_presentation;
+        struct AVTileGrid *tile_grid;
     } params;
 
     /**
diff --git a/libavformat/dump.c b/libavformat/dump.c
index aff51b43f6..33932f39ae 100644
--- a/libavformat/dump.c
+++ b/libavformat/dump.c
@@ -22,6 +22,7 @@
 #include <stdio.h>
 #include <stdint.h>
 
+#include "libavutil/avstring.h"
 #include "libavutil/channel_layout.h"
 #include "libavutil/display.h"
 #include "libavutil/iamf.h"
@@ -720,6 +721,41 @@ static void dump_stream_group(const AVFormatContext *ic, uint8_t *printed,
             }
         }
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID: {
+        const AVTileGrid *tile_grid = stg->params.tile_grid;
+        AVCodecContext *avctx = avcodec_alloc_context3(NULL);
+        const char *ptr = NULL;
+        av_log(NULL, AV_LOG_INFO, " Tile Grid:");
+        av_log(NULL, AV_LOG_VERBOSE, " %d rows, %d columns,", tile_grid->tile_rows, tile_grid->tile_cols);
+        if (avctx && stg->nb_streams && !avcodec_parameters_to_context(avctx, stg->streams[0]->codecpar)) {
+            int size = tile_grid->tile_rows * tile_grid->tile_cols;
+            int coded_width = 0, coded_height = 0;
+            avctx->width  = tile_grid->output_width;
+            avctx->height = tile_grid->output_height;
+            for (int j = 0; j < tile_grid->tile_cols; j++)
+                coded_width += tile_grid->tile_width[j];
+            for (int j = 0; j < size; j += tile_grid->tile_cols)
+                coded_height += tile_grid->tile_height[j];
+            avctx->coded_width  = coded_width;
+            avctx->coded_height = coded_height;
+            if (ic->dump_separator)
+                av_opt_set(avctx, "dump_separator", ic->dump_separator, 0);
+            buf[0] = 0;
+            avcodec_string(buf, sizeof(buf), avctx, is_output);
+            ptr = av_stristr(buf, " ");
+        }
+        avcodec_free_context(&avctx);
+        if (ptr)
+            av_log(NULL, AV_LOG_INFO, "%s", ptr);
+        av_log(NULL, AV_LOG_INFO, "\n");
+        dump_metadata(NULL, stg->metadata, "    ", AV_LOG_INFO);
+        for (int i = 0; i < stg->nb_streams; i++) {
+            const AVStream *st = stg->streams[i];
+            dump_stream_format(ic, st->index, i, index, is_output, AV_LOG_VERBOSE);
+            printed[st->index] = 1;
+        }
+        break;
+    }
     }
     default:
         break;
diff --git a/libavformat/options.c b/libavformat/options.c
index 75ec86ce05..31620013c6 100644
--- a/libavformat/options.c
+++ b/libavformat/options.c
@@ -337,6 +337,26 @@ fail:
     return NULL;
 }
 
+#define FLAGS AV_OPT_FLAG_ENCODING_PARAM | AV_OPT_FLAG_VIDEO_PARAM
+#define OFFSET(x) offsetof(AVTileGrid, x)
+static const AVOption tile_grid_options[] = {
+    { "tile_rows", NULL, OFFSET(tile_rows), AV_OPT_TYPE_INT, { .i64 = 1 }, 1, INT_MAX, FLAGS },
+    { "tile_cols", NULL, OFFSET(tile_cols), AV_OPT_TYPE_INT, { .i64 = 1 }, 1, INT_MAX, FLAGS },
+    { "horizontal_offset", NULL, OFFSET(horizontal_offset), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
+    { "vertical_offset", NULL, OFFSET(vertical_offset),     AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
+    { "output_size", "size of the output image", OFFSET(output_width), AV_OPT_TYPE_IMAGE_SIZE,
+        { .str = NULL }, 0, INT_MAX, FLAGS },
+    { NULL },
+};
+#undef FLAGS
+#undef OFFSET
+
+static const AVClass tile_grid_class = {
+    .class_name = "AVTileGrid",
+    .version    = LIBAVUTIL_VERSION_INT,
+    .option     = tile_grid_options,
+};
+
 static void *stream_group_child_next(void *obj, void *prev)
 {
     AVStreamGroup *stg = obj;
@@ -346,6 +366,8 @@ static void *stream_group_child_next(void *obj, void *prev)
             return stg->params.iamf_audio_element;
         case AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION:
             return stg->params.iamf_mix_presentation;
+        case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+            return stg->params.tile_grid;
         default:
             break;
         }
@@ -368,6 +390,9 @@ static const AVClass *stream_group_child_iterate(void **opaque)
     case AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION:
         ret = av_iamf_mix_presentation_get_class();
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+        ret = &tile_grid_class;
+        break;
     default:
         break;
     }
@@ -429,6 +454,13 @@ AVStreamGroup *avformat_stream_group_create(AVFormatContext *s,
         if (!stg->params.iamf_mix_presentation)
             goto fail;
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+        stg->params.tile_grid = av_mallocz(sizeof(*stg->params.tile_grid));
+        if (!stg->params.tile_grid)
+            goto fail;
+        stg->params.tile_grid->av_class = &tile_grid_class;
+        av_opt_set_defaults(stg->params.tile_grid);
+        break;
     default:
         goto fail;
     }
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH 3/4 v2] avformat/mov: add support for tile HEIF still images

[James Almer]

Export each tile as its own stream, and the tiling information as a Stream
Group of type TILE_GRID.
This also enables exporting other stream items like thumbnails, which may be
present in non tiled HEIF images too. For those, the primary stream will be
tagged with the default disposition.

Based on a patch by Swaraj Hota

Signed-off-by: James Almer <jamrial@gmail.com>
---
 libavcodec/packet.h    |   9 ++
 libavformat/avformat.h |   6 +
 libavformat/dump.c     |   6 +
 libavformat/isom.h     |   7 +-
 libavformat/mov.c      | 320 ++++++++++++++++++++++++++++++++++++-----
 5 files changed, 312 insertions(+), 36 deletions(-)

diff --git a/libavcodec/packet.h b/libavcodec/packet.h
index 2c57d262c6..48ca799334 100644
--- a/libavcodec/packet.h
+++ b/libavcodec/packet.h
@@ -323,6 +323,15 @@ enum AVPacketSideDataType {
      */
     AV_PKT_DATA_IAMF_RECON_GAIN_INFO_PARAM,
 
+    /**
+     * Tile info for image reconstruction, e.g HEIF.
+     * @code
+     * u32le tile number in row major order [0..nb_tiles-1]
+     * u32le nb_tiles
+     * @endcode
+     */
+    AV_PKT_DATA_TILE_INFO,
+
     /**
      * The number of side data types.
      * This is not part of the public API/ABI in the sense that it may
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index ab9a3fc6be..cf4e72e11d 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -811,6 +811,12 @@ typedef struct AVIndexEntry {
  * The video stream contains still images.
  */
 #define AV_DISPOSITION_STILL_IMAGE          (1 << 20)
+/**
+ * The video stream is intended to be merged with another stream before
+ * presentation.
+ * Used for example to signal the stream contains a tile from a HEIF grid.
+ */
+#define AV_DISPOSITION_TILE                 (1 << 21)
 
 /**
  * @return The AV_DISPOSITION_* flag corresponding to disp or a negative error
diff --git a/libavformat/dump.c b/libavformat/dump.c
index 33932f39ae..5a8b5d7160 100644
--- a/libavformat/dump.c
+++ b/libavformat/dump.c
@@ -514,6 +514,10 @@ static void dump_sidedata(void *ctx, const AVStream *st, const char *indent,
             av_log(ctx, log_level, "SMPTE ST 12-1:2014: ");
             dump_s12m_timecode(ctx, st, sd, log_level);
             break;
+        case AV_PKT_DATA_TILE_INFO:
+            av_log(ctx, log_level, "HEIF tile info: tile %u/%u",
+                   AV_RL32(sd->data) + 1U, AV_RL32(sd->data + 4));
+            break;
         default:
             av_log(ctx, log_level, "unknown side data type %d "
                    "(%"SIZE_SPECIFIER" bytes)", sd->type, sd->size);
@@ -640,6 +644,8 @@ static void dump_stream_format(const AVFormatContext *ic, int i,
         av_log(NULL, log_level, " (still image)");
     if (st->disposition & AV_DISPOSITION_NON_DIEGETIC)
         av_log(NULL, log_level, " (non-diegetic)");
+    if (st->disposition & AV_DISPOSITION_TILE)
+        av_log(NULL, log_level, " (tile)");
     av_log(NULL, log_level, "\n");
 
     dump_metadata(NULL, st->metadata, extra_indent, log_level);
diff --git a/libavformat/isom.h b/libavformat/isom.h
index 21caaac256..c6a8507a68 100644
--- a/libavformat/isom.h
+++ b/libavformat/isom.h
@@ -267,10 +267,10 @@ typedef struct HEIFItem {
     int item_id;
     int64_t extent_length;
     int64_t extent_offset;
-    int64_t size;
     int width;
     int height;
     int type;
+    int is_idat_relative;
 } HEIFItem;
 
 typedef struct MOVContext {
@@ -336,6 +336,11 @@ typedef struct MOVContext {
     int cur_item_id;
     HEIFItem *heif_info;
     int heif_info_size;
+    int grid_item_id;
+    int thmb_item_id;
+    int16_t *tile_id_list;
+    int nb_tiles;
+    int64_t idat_offset;
     int interleaved_read;
 } MOVContext;
 
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 03ca62895e..2dce370b5d 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -185,6 +185,30 @@ static int mov_read_mac_string(MOVContext *c, AVIOContext *pb, int len,
     return p - dst;
 }
 
+static AVStream *get_curr_st(MOVContext *c)
+{
+    AVStream *st = NULL;
+
+    if (c->fc->nb_streams < 1)
+        return NULL;
+
+    for (int i = 0; i < c->heif_info_size; i++) {
+        HEIFItem *item = &c->heif_info[i];
+
+        if (!item->st)
+            continue;
+        if (item->st->id != c->cur_item_id)
+            continue;
+
+        st = item->st;
+        break;
+    }
+    if (!st)
+        st = c->fc->streams[c->fc->nb_streams-1];
+
+    return st;
+}
+
 static int mov_read_covr(MOVContext *c, AVIOContext *pb, int type, int len)
 {
     AVStream *st;
@@ -1767,9 +1791,9 @@ static int mov_read_colr(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     uint16_t color_primaries, color_trc, color_matrix;
     int ret;
 
-    if (c->fc->nb_streams < 1)
+    st = get_curr_st(c);
+    if (!st)
         return 0;
-    st = c->fc->streams[c->fc->nb_streams - 1];
 
     ret = ffio_read_size(pb, color_parameter_type, 4);
     if (ret < 0)
@@ -2117,9 +2141,9 @@ static int mov_read_glbl(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     AVStream *st;
     int ret;
 
-    if (c->fc->nb_streams < 1)
+    st = get_curr_st(c);
+    if (!st)
         return 0;
-    st = c->fc->streams[c->fc->nb_streams-1];
 
     if ((uint64_t)atom.size > (1<<30))
         return AVERROR_INVALIDDATA;
@@ -4946,12 +4970,10 @@ static int heif_add_stream(MOVContext *c, HEIFItem *item)
     st->codecpar->codec_type = AVMEDIA_TYPE_VIDEO;
     st->codecpar->codec_id = mov_codec_id(st, item->type);
     sc->ffindex = st->index;
-    c->trak_index = st->index;
     st->avg_frame_rate.num = st->avg_frame_rate.den = 1;
     st->time_base.num = st->time_base.den = 1;
     st->nb_frames = 1;
     sc->time_scale = 1;
-    sc = st->priv_data;
     sc->pb = c->fc->pb;
     sc->pb_is_copied = 1;
 
@@ -7779,11 +7801,60 @@ static int mov_read_pitm(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     return atom.size;
 }
 
+static int mov_read_idat(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    c->idat_offset = avio_tell(pb);
+    return 0;
+}
+
+static int read_image_grid(AVFormatContext *s, AVTileGrid *tile_grid, HEIFItem *item) {
+    MOVContext *c = s->priv_data;
+    int64_t offset = 0, pos = avio_tell(s->pb);
+    uint8_t flags;
+
+    if (!(s->pb->seekable & AVIO_SEEKABLE_NORMAL)) {
+        av_log(c->fc, AV_LOG_INFO, "grid box with non seekable input\n");
+        return AVERROR_PATCHWELCOME;
+    }
+    if (item->is_idat_relative) {
+        if (!c->idat_offset) {
+            av_log(c->fc, AV_LOG_ERROR, "missing idat box required by the image grid\n");
+            return AVERROR_INVALIDDATA;
+        }
+        offset = c->idat_offset;
+    }
+
+    avio_seek(s->pb, item->extent_offset + offset, SEEK_SET);
+
+    avio_r8(s->pb);    /* version */
+    flags = avio_r8(s->pb);
+
+    tile_grid->tile_rows = avio_r8(s->pb) + 1;
+    tile_grid->tile_cols = avio_r8(s->pb) + 1;
+    /* actual width and height of output image */
+    tile_grid->output_width  = (flags & 1) ? avio_rb32(s->pb) : avio_rb16(s->pb);
+    tile_grid->output_height = (flags & 1) ? avio_rb32(s->pb) : avio_rb16(s->pb);
+
+    tile_grid->tile_width  = av_calloc(tile_grid->tile_rows * tile_grid->tile_cols,
+                                       sizeof(*tile_grid->tile_width));
+    tile_grid->tile_height = av_calloc(tile_grid->tile_rows * tile_grid->tile_cols,
+                                       sizeof(*tile_grid->tile_height));
+    if (!tile_grid->tile_width || !tile_grid->tile_width)
+        return AVERROR(ENOMEM);
+
+    av_log(c->fc, AV_LOG_TRACE, "grid: grid_rows %d grid_cols %d output_width %d output_height %d\n",
+           tile_grid->tile_rows, tile_grid->tile_cols, tile_grid->output_width, tile_grid->output_height);
+
+    avio_seek(s->pb, pos, SEEK_SET);
+
+    return 0;
+}
+
 static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 {
     int version, offset_size, length_size, base_offset_size, index_size;
     int item_count, extent_count;
-    uint64_t base_offset, extent_offset, extent_length;
+    int64_t base_offset, extent_offset, extent_length;
     uint8_t value;
 
     if (c->found_moov) {
@@ -7830,6 +7901,7 @@ static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
         avio_rb16(pb);  // data_reference_index.
         if (rb_size(pb, &base_offset, base_offset_size) < 0)
             return AVERROR_INVALIDDATA;
+        av_log(c->fc, AV_LOG_TRACE, "iloc: base_offset %"PRId64"\n", base_offset);
         extent_count = avio_rb16(pb);
         if (extent_count > 1) {
             // For still AVIF images, we only support one extent item.
@@ -7840,6 +7912,8 @@ static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
             if (rb_size(pb, &extent_offset, offset_size) < 0 ||
                 rb_size(pb, &extent_length, length_size) < 0)
                 return AVERROR_INVALIDDATA;
+            if (offset_type == 1)
+                c->heif_info[i].is_idat_relative = 1;
             c->heif_info[i].extent_length = extent_length;
             c->heif_info[i].extent_offset = base_offset + extent_offset;
             av_log(c->fc, AV_LOG_TRACE, "iloc: item_idx %d, offset_type %d, "
@@ -7878,10 +7952,6 @@ static int mov_read_infe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     av_log(c->fc, AV_LOG_TRACE, "infe: item_id %d, item_type %s, item_name %s\n",
            item_id, av_fourcc2str(item_type), item_name);
 
-    // Skip all but the primary item until support is added
-    if (item_id != c->primary_item_id)
-        return 0;
-
     if (size > 0)
         avio_skip(pb, size);
 
@@ -7895,6 +7965,9 @@ static int mov_read_infe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
         if (ret < 0)
             return ret;
         break;
+    case MKTAG('g','r','i','d'):
+        c->grid_item_id = item_id;
+        break;
     default:
         av_log(c->fc, AV_LOG_TRACE, "infe: ignoring item_type %s\n", av_fourcc2str(item_type));
         break;
@@ -7954,6 +8027,59 @@ static int mov_read_iref(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     return mov_read_default(c, pb, atom);
 }
 
+static int mov_read_dimg(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    int entries, i;
+    int from_item_id = avio_rb16(pb);
+
+    if (c->grid_item_id < 0) {
+        av_log(c->fc, AV_LOG_ERROR, "Missing grid information\n");
+        return AVERROR_INVALIDDATA;
+    }
+    if (from_item_id != c->grid_item_id) {
+        avpriv_request_sample(c->fc, "Derived item of type other than 'grid'");
+        return AVERROR_PATCHWELCOME;
+    }
+    entries = avio_rb16(pb);
+    c->tile_id_list = av_malloc_array(entries, sizeof(*c->tile_id_list));
+    if (!c->tile_id_list)
+        return AVERROR(ENOMEM);
+    /* 'to' item ids */
+    for (i = 0; i < entries; i++)
+        c->tile_id_list[i] = avio_rb16(pb);
+    c->nb_tiles = entries;
+
+    av_log(c->fc, AV_LOG_TRACE, "dimg: from_item_id %d, entries %d\n",
+           from_item_id, entries);
+
+    return 0;
+}
+
+static int mov_read_thmb(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    int entries;
+    int to_item_id, from_item_id = avio_rb16(pb);
+
+    entries = avio_rb16(pb);
+    if (entries > 1) {
+        avpriv_request_sample(c->fc, "More than one thmb entry");
+        return AVERROR_PATCHWELCOME;
+    }
+    /* 'to' item ids */
+    to_item_id = avio_rb16(pb);
+
+    if (to_item_id != c->primary_item_id ||
+        to_item_id != c->grid_item_id)
+        return 0;
+
+    c->thmb_item_id = from_item_id;
+
+    av_log(c->fc, AV_LOG_TRACE, "thmb: from_item_id %d, entries %d\n",
+           from_item_id, entries);
+
+    return 0;
+}
+
 static int mov_read_ispe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 {
     uint32_t width, height;
@@ -7969,15 +8095,17 @@ static int mov_read_ispe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     width  = avio_rb32(pb);
     height = avio_rb32(pb);
 
-    av_log(c->fc, AV_LOG_TRACE, "ispe: item_id %d, width %u, height %u\n",
+    av_log(c->fc, AV_LOG_TRACE, "ispe: cur_item_id %d, width %u, height %u\n",
            c->cur_item_id, width, height);
 
     for (int i = 0; i < c->heif_info_size; i++) {
-        if (c->heif_info[i].item_id == c->cur_item_id) {
-            c->heif_info[i].width  = width;
-            c->heif_info[i].height = height;
-            break;
-        }
+        HEIFItem *item = &c->heif_info[i];
+        if (item->item_id != c->cur_item_id)
+            continue;
+
+        item->width  = width;
+        item->height = height;
+        break;
     }
 
     return 0;
@@ -8074,10 +8202,6 @@ static int mov_read_iprp(MOVContext *c, AVIOContext *pb, MOVAtom atom)
             av_log(c->fc, AV_LOG_TRACE, "ipma: property_index %d, item_id %d, item_type %s\n",
                    index + 1, item_id, av_fourcc2str(ref->type));
 
-            // Skip properties referencing items other than the primary item until support is added
-            if (item_id != c->primary_item_id)
-                continue;
-
             c->cur_item_id = item_id;
 
             ret = mov_read_default(c, &ref->b.pub,
@@ -8205,6 +8329,9 @@ static const MOVParseTableEntry mov_default_parse_table[] = {
 { MKTAG('p','c','m','C'), mov_read_pcmc }, /* PCM configuration box */
 { MKTAG('p','i','t','m'), mov_read_pitm },
 { MKTAG('e','v','c','C'), mov_read_glbl },
+{ MKTAG('d','i','m','g'), mov_read_dimg },
+{ MKTAG('t','h','m','b'), mov_read_thmb },
+{ MKTAG('i','d','a','t'), mov_read_idat },
 { MKTAG('i','r','e','f'), mov_read_iref },
 { MKTAG('i','s','p','e'), mov_read_ispe },
 { MKTAG('i','p','r','p'), mov_read_iprp },
@@ -8713,6 +8840,7 @@ static int mov_read_close(AVFormatContext *s)
     av_freep(&mov->aes_decrypt);
     av_freep(&mov->chapter_tracks);
     av_freep(&mov->heif_info);
+    av_freep(&mov->tile_id_list);
 
     return 0;
 }
@@ -8852,6 +8980,118 @@ fail:
     return ret;
 }
 
+static int mov_parse_tiles(AVFormatContext *s)
+{
+    MOVContext *mov = s->priv_data;
+    AVStreamGroup *stg = avformat_stream_group_create(s, AV_STREAM_GROUP_PARAMS_TILE_GRID, NULL);
+    AVTileGrid *tile_grid;
+    int err;
+
+    if (!stg)
+        return AVERROR(ENOMEM);
+
+    tile_grid = stg->params.tile_grid;
+
+    av_assert0(mov->grid_item_id >= 0);
+    for (int i = 0; i < mov->heif_info_size; i++) {
+        HEIFItem *item = &mov->heif_info[i];
+
+        if (item->item_id != mov->grid_item_id)
+            continue;
+        err = read_image_grid(s, tile_grid, item);
+        if (err < 0)
+            return err;
+        stg->id = item->item_id;
+        break;
+    }
+
+    for (int i = 0; i < mov->nb_tiles; i++) {
+        int tile_id = mov->tile_id_list[i];
+
+        for (int j = 0; j < mov->heif_info_size; j++) {
+            HEIFItem *item = &mov->heif_info[j];
+            AVStream *st = item->st;
+            AVPacketSideData *sd;
+            MOVStreamContext *sc;
+            int64_t offset = 0;
+
+            if (item->item_id != tile_id)
+                continue;
+            if (!st) {
+                av_log(s, AV_LOG_ERROR, "HEIF tile %d doesn't reference a stream\n", tile_id);
+                return AVERROR_INVALIDDATA;
+            }
+            if (item->is_idat_relative) {
+                if (!mov->idat_offset) {
+                    av_log(s, AV_LOG_ERROR, "Missing idat box for HEIF tile %d\n", tile_id);
+                    return AVERROR_INVALIDDATA;
+                }
+                offset = mov->idat_offset;
+            }
+
+            tile_grid->tile_width[i]  = item->width;
+            tile_grid->tile_height[i] = item->height;
+            st->codecpar->width  = item->width;
+            st->codecpar->height = item->height;
+
+            err = avformat_stream_group_add_stream(stg, st);
+            if (err == AVERROR(EEXIST))
+                return AVERROR_INVALIDDATA;
+            else if (err < 0)
+                return err;
+
+            sc = st->priv_data;
+            sc->sample_sizes[0]  = item->extent_length;
+            sc->chunk_offsets[0] = item->extent_offset + offset;
+
+            st->disposition |= AV_DISPOSITION_TILE;
+
+            mov_build_index(mov, st);
+
+            sd = av_packet_side_data_new(&st->codecpar->coded_side_data,
+                                         &st->codecpar->nb_coded_side_data,
+                                         AV_PKT_DATA_TILE_INFO,
+                                         sizeof(uint32_t) * 2, 0);
+            if (!sd)
+                return AVERROR(ENOMEM);
+
+            AV_WL32(sd->data, i);
+            AV_WL32(sd->data + 4, mov->nb_tiles);
+            break;
+        }
+    }
+
+    for (int i = 0; i < mov->heif_info_size; i++) {
+        HEIFItem *item = &mov->heif_info[i];
+        AVStream *st = item->st;
+        MOVStreamContext *sc;
+        int64_t offset = 0;
+
+        if (item->item_id != mov->thmb_item_id)
+            continue;
+
+        if (!st) {
+            av_log(s, AV_LOG_ERROR, "HEIF thumbnail doesn't reference a stream\n");
+            return AVERROR_INVALIDDATA;
+        }
+        if (item->is_idat_relative) {
+            if (!mov->idat_offset) {
+                av_log(s, AV_LOG_ERROR, "Missing idat box for HEIF thumbnail\n");
+                return AVERROR_INVALIDDATA;
+            }
+            offset = mov->idat_offset;
+        }
+
+        sc = st->priv_data;
+        sc->sample_sizes[0]  = item->extent_length;
+        sc->chunk_offsets[0] = item->extent_offset + offset;
+
+        mov_build_index(mov, st);
+    }
+
+    return 0;
+}
+
 static int mov_read_header(AVFormatContext *s)
 {
     MOVContext *mov = s->priv_data;
@@ -8868,6 +9108,9 @@ static int mov_read_header(AVFormatContext *s)
 
     mov->fc = s;
     mov->trak_index = -1;
+    mov->grid_item_id = -1;
+    mov->thmb_item_id = -1;
+    mov->primary_item_id = -1;
     /* .mov and .mp4 aren't streamable anyway (only progressive download if moov is before mdat) */
     if (pb->seekable & AVIO_SEEKABLE_NORMAL)
         atom.size = avio_size(pb);
@@ -8890,23 +9133,30 @@ static int mov_read_header(AVFormatContext *s)
     av_log(mov->fc, AV_LOG_TRACE, "on_parse_exit_offset=%"PRId64"\n", avio_tell(pb));
 
     if (mov->found_iloc) {
-        for (i = 0; i < mov->heif_info_size; i++) {
-            HEIFItem *item = &mov->heif_info[i];
-            MOVStreamContext *sc;
-            AVStream *st;
+        if (mov->nb_tiles) {
+            err = mov_parse_tiles(s);
+            if (err < 0)
+                return err;
+        } else
+            for (i = 0; i < mov->heif_info_size; i++) {
+                HEIFItem *item = &mov->heif_info[i];
+                AVStream *st = item->st;
+                MOVStreamContext *sc;
 
-            if (!item->st)
-                continue;
+                if (!st)
+                    continue;
 
-            st = item->st;
-            sc = st->priv_data;
-            st->codecpar->width  = item->width;
-            st->codecpar->height = item->height;
-            sc->sample_sizes[0]  = item->extent_length;
-            sc->chunk_offsets[0] = item->extent_offset;
+                sc = st->priv_data;
+                st->codecpar->width  = item->width;
+                st->codecpar->height = item->height;
+                sc->sample_sizes[0]  = item->extent_length;
+                sc->chunk_offsets[0] = item->extent_offset;
 
-            mov_build_index(mov, st);
-        }
+                if (item->item_id == mov->primary_item_id)
+                    st->disposition |= AV_DISPOSITION_DEFAULT;
+
+                mov_build_index(mov, st);
+            }
     }
 
     if (pb->seekable & AVIO_SEEKABLE_NORMAL) {
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH 4/4 v2] fate/mov: test remuxing all stream heif items

[James Almer]

Signed-off-by: James Almer <jamrial@gmail.com>
---
 tests/fate/mov.mak                                       | 2 +-
 tests/ref/fate/mov-heic-demux-still-image-multiple-items | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/tests/fate/mov.mak b/tests/fate/mov.mak
index f202f36d96..f549ae33d7 100644
--- a/tests/fate/mov.mak
+++ b/tests/fate/mov.mak
@@ -156,7 +156,7 @@ fate-mov-heic-demux-still-image-1-item: CMD = framemd5 -i $(TARGET_SAMPLES)/heif
 
 FATE_MOV_FFMPEG-$(call FRAMEMD5, MOV, HEVC, HEVC_PARSER) \
                            += fate-mov-heic-demux-still-image-multiple-items
-fate-mov-heic-demux-still-image-multiple-items: CMD = framemd5 -i $(TARGET_SAMPLES)/heif-conformance/C003.heic -c:v copy
+fate-mov-heic-demux-still-image-multiple-items: CMD = framemd5 -i $(TARGET_SAMPLES)/heif-conformance/C003.heic -c:v copy -map 0
 
 # Resulting remux should have:
 # 1. first audio stream with AV_DISPOSITION_HEARING_IMPAIRED
diff --git a/tests/ref/fate/mov-heic-demux-still-image-multiple-items b/tests/ref/fate/mov-heic-demux-still-image-multiple-items
index c850c1ff9c..753cef267a 100644
--- a/tests/ref/fate/mov-heic-demux-still-image-multiple-items
+++ b/tests/ref/fate/mov-heic-demux-still-image-multiple-items
@@ -2,10 +2,17 @@
 #version: 2
 #hash: MD5
 #extradata 0,                             100, 5444bf01e03182c73ae957179d560f4d
+#extradata 1,                             100, 5444bf01e03182c73ae957179d560f4d
 #tb 0: 1/1
 #media_type 0: video
 #codec_id 0: hevc
 #dimensions 0: 1280x720
 #sar 0: 0/1
+#tb 1: 1/1
+#media_type 1: video
+#codec_id 1: hevc
+#dimensions 1: 1280x720
+#sar 1: 0/1
 #stream#, dts,        pts, duration,     size, hash
 0,          0,          0,        1,   111554, 03ceabfab39afd2e2e796b9362111f32
+1,          0,          0,        1,   112393, daa001d351c088a5bc328459e2501c95
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH 2/4 v5] avformat: add a Tile Grid stream group type

[James Almer]

This will be used to support tiled image formats like HEIF.

Signed-off-by: James Almer <jamrial@gmail.com>
---
Removed tile dimension arrays, as the same information can be fetched from the
streams within the group.

 libavformat/avformat.c |  4 +++
 libavformat/avformat.h | 65 ++++++++++++++++++++++++++++++++++++++++++
 libavformat/dump.c     | 36 +++++++++++++++++++++++
 libavformat/options.c  | 32 +++++++++++++++++++++
 4 files changed, 137 insertions(+)

diff --git a/libavformat/avformat.c b/libavformat/avformat.c
index 882927f7b1..c253bdbbad 100644
--- a/libavformat/avformat.c
+++ b/libavformat/avformat.c
@@ -100,6 +100,10 @@ void ff_free_stream_group(AVStreamGroup **pstg)
         av_iamf_mix_presentation_free(&stg->params.iamf_mix_presentation);
         break;
     }
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+        av_opt_free(stg->params.tile_grid);
+        av_freep(&stg->params.tile_grid);
+        break;
     default:
         break;
     }
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index 5d0fe82250..184bb5b2ec 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -1018,10 +1018,74 @@ typedef struct AVStream {
     int pts_wrap_bits;
 } AVStream;
 
+/**
+ * AVStreamGroupTileGrid holds information on how to combine several independent images in
+ * a single grid for presentation.
+ *
+ * Its size is not a part of the ABI. No new fields may be added to this struct
+ * without a major version bump.
+ */
+typedef struct AVStreamGroupTileGrid {
+    const AVClass *av_class;
+
+    /**
+     * Amount of rows in the grid.
+     *
+     * Must be > 0.
+     */
+    int tile_rows;
+    /**
+     * Amount of columns in the grid.
+     *
+     * Must be > 0.
+     */
+    int tile_cols;
+
+    /**
+     * Offset in pixels from the left edge of the grid where the actual image
+     * meant for presentation starts.
+     *
+     * This field must be >= 0 and <= the sum of widths of all images in a row.
+     */
+    int horizontal_offset;
+    /**
+     * Offset in pixels from the top edge of the grid where the actual image
+     * meant for presentation starts.
+     *
+     * This field must be >= 0 and <= the sum of heights of all images in a
+     * column.
+     */
+    int vertical_offset;
+
+    /**
+     * Width of the final image for presentation.
+     *
+     * Must be > 0 and <= the sum of widths of all images in a row minus
+     * @ref horizontal_offset.
+     * When it's not equal the sum of widths of all images in a row, the result
+     * of said sum minus output_width minus @ref horizontal_offset is the
+     * amount of pixels to be cropped from the right edge of the final image
+     * before presentation.
+     */
+    int output_width;
+    /**
+     * Height of the final image for presentation.
+     *
+     * Must be > 0 and <= the sum of heights of all images in a column minus
+     * @ref vertical_offset.
+     * When it's not equal the sum of heights of all images in a column, the
+     * result of said sum minus output_height minus @ref vertical_offset is the
+     * amount of pixels to be cropped from the right edge of the final image
+     * before presentation.
+     */
+    int output_height;
+} AVStreamGroupTileGrid;
+
 enum AVStreamGroupParamsType {
     AV_STREAM_GROUP_PARAMS_NONE,
     AV_STREAM_GROUP_PARAMS_IAMF_AUDIO_ELEMENT,
     AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION,
+    AV_STREAM_GROUP_PARAMS_TILE_GRID,
 };
 
 struct AVIAMFAudioElement;
@@ -1062,6 +1126,7 @@ typedef struct AVStreamGroup {
     union {
         struct AVIAMFAudioElement *iamf_audio_element;
         struct AVIAMFMixPresentation *iamf_mix_presentation;
+        struct AVStreamGroupTileGrid *tile_grid;
     } params;
 
     /**
diff --git a/libavformat/dump.c b/libavformat/dump.c
index aff51b43f6..ef4a8df6fe 100644
--- a/libavformat/dump.c
+++ b/libavformat/dump.c
@@ -22,6 +22,7 @@
 #include <stdio.h>
 #include <stdint.h>
 
+#include "libavutil/avstring.h"
 #include "libavutil/channel_layout.h"
 #include "libavutil/display.h"
 #include "libavutil/iamf.h"
@@ -720,6 +721,41 @@ static void dump_stream_group(const AVFormatContext *ic, uint8_t *printed,
             }
         }
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID: {
+        const AVStreamGroupTileGrid *tile_grid = stg->params.tile_grid;
+        AVCodecContext *avctx = avcodec_alloc_context3(NULL);
+        const char *ptr = NULL;
+        av_log(NULL, AV_LOG_INFO, " Tile Grid:");
+        av_log(NULL, AV_LOG_VERBOSE, " %d rows, %d columns,", tile_grid->tile_rows, tile_grid->tile_cols);
+        if (avctx && stg->nb_streams && !avcodec_parameters_to_context(avctx, stg->streams[0]->codecpar)) {
+            int size = tile_grid->tile_rows * tile_grid->tile_cols;
+            int coded_width = 0, coded_height = 0;
+            avctx->width  = tile_grid->output_width;
+            avctx->height = tile_grid->output_height;
+            for (int j = 0; j < tile_grid->tile_cols; j++)
+                coded_width += stg->streams[j]->codecpar->width;
+            for (int j = 0; j < size; j += tile_grid->tile_cols)
+                coded_height += stg->streams[j]->codecpar->height;
+            avctx->coded_width  = coded_width;
+            avctx->coded_height = coded_height;
+            if (ic->dump_separator)
+                av_opt_set(avctx, "dump_separator", ic->dump_separator, 0);
+            buf[0] = 0;
+            avcodec_string(buf, sizeof(buf), avctx, is_output);
+            ptr = av_stristr(buf, " ");
+        }
+        avcodec_free_context(&avctx);
+        if (ptr)
+            av_log(NULL, AV_LOG_INFO, "%s", ptr);
+        av_log(NULL, AV_LOG_INFO, "\n");
+        dump_metadata(NULL, stg->metadata, "    ", AV_LOG_INFO);
+        for (int i = 0; i < stg->nb_streams; i++) {
+            const AVStream *st = stg->streams[i];
+            dump_stream_format(ic, st->index, i, index, is_output, AV_LOG_VERBOSE);
+            printed[st->index] = 1;
+        }
+        break;
+    }
     }
     default:
         break;
diff --git a/libavformat/options.c b/libavformat/options.c
index 75ec86ce05..9e83ef0d41 100644
--- a/libavformat/options.c
+++ b/libavformat/options.c
@@ -337,6 +337,26 @@ fail:
     return NULL;
 }
 
+#define FLAGS AV_OPT_FLAG_ENCODING_PARAM | AV_OPT_FLAG_VIDEO_PARAM
+#define OFFSET(x) offsetof(AVStreamGroupTileGrid, x)
+static const AVOption tile_grid_options[] = {
+    { "tile_rows", NULL, OFFSET(tile_rows), AV_OPT_TYPE_INT, { .i64 = 1 }, 1, INT_MAX, FLAGS },
+    { "tile_cols", NULL, OFFSET(tile_cols), AV_OPT_TYPE_INT, { .i64 = 1 }, 1, INT_MAX, FLAGS },
+    { "horizontal_offset", NULL, OFFSET(horizontal_offset), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
+    { "vertical_offset", NULL, OFFSET(vertical_offset),     AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, FLAGS },
+    { "output_size", "size of the output image", OFFSET(output_width), AV_OPT_TYPE_IMAGE_SIZE,
+        { .str = NULL }, 0, INT_MAX, FLAGS },
+    { NULL },
+};
+#undef FLAGS
+#undef OFFSET
+
+static const AVClass tile_grid_class = {
+    .class_name = "AVStreamGroupTileGrid",
+    .version    = LIBAVUTIL_VERSION_INT,
+    .option     = tile_grid_options,
+};
+
 static void *stream_group_child_next(void *obj, void *prev)
 {
     AVStreamGroup *stg = obj;
@@ -346,6 +366,8 @@ static void *stream_group_child_next(void *obj, void *prev)
             return stg->params.iamf_audio_element;
         case AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION:
             return stg->params.iamf_mix_presentation;
+        case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+            return stg->params.tile_grid;
         default:
             break;
         }
@@ -368,6 +390,9 @@ static const AVClass *stream_group_child_iterate(void **opaque)
     case AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION:
         ret = av_iamf_mix_presentation_get_class();
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+        ret = &tile_grid_class;
+        break;
     default:
         break;
     }
@@ -429,6 +454,13 @@ AVStreamGroup *avformat_stream_group_create(AVFormatContext *s,
         if (!stg->params.iamf_mix_presentation)
             goto fail;
         break;
+    case AV_STREAM_GROUP_PARAMS_TILE_GRID:
+        stg->params.tile_grid = av_mallocz(sizeof(*stg->params.tile_grid));
+        if (!stg->params.tile_grid)
+            goto fail;
+        stg->params.tile_grid->av_class = &tile_grid_class;
+        av_opt_set_defaults(stg->params.tile_grid);
+        break;
     default:
         goto fail;
     }
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH 3/4 v3] avformat/mov: add support for tile HEIF still images

[James Almer]

Export each tile as its own stream, and the tiling information as a Stream
Group of type TILE_GRID.
This also enables exporting other stream items like thumbnails, which may be
present in non tiled HEIF images too. For those, the primary stream will be
tagged with the default disposition.

Based on a patch by Swaraj Hota

Signed-off-by: James Almer <jamrial@gmail.com>
---
 libavcodec/packet.h    |   9 ++
 libavformat/avformat.h |   6 +
 libavformat/dump.c     |   6 +
 libavformat/isom.h     |   7 +-
 libavformat/mov.c      | 313 ++++++++++++++++++++++++++++++++++++-----
 5 files changed, 305 insertions(+), 36 deletions(-)

diff --git a/libavcodec/packet.h b/libavcodec/packet.h
index 2c57d262c6..48ca799334 100644
--- a/libavcodec/packet.h
+++ b/libavcodec/packet.h
@@ -323,6 +323,15 @@ enum AVPacketSideDataType {
      */
     AV_PKT_DATA_IAMF_RECON_GAIN_INFO_PARAM,
 
+    /**
+     * Tile info for image reconstruction, e.g HEIF.
+     * @code
+     * u32le tile number in row major order [0..nb_tiles-1]
+     * u32le nb_tiles
+     * @endcode
+     */
+    AV_PKT_DATA_TILE_INFO,
+
     /**
      * The number of side data types.
      * This is not part of the public API/ABI in the sense that it may
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index 184bb5b2ec..8199f7b052 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -811,6 +811,12 @@ typedef struct AVIndexEntry {
  * The video stream contains still images.
  */
 #define AV_DISPOSITION_STILL_IMAGE          (1 << 20)
+/**
+ * The video stream is intended to be merged with another stream before
+ * presentation.
+ * Used for example to signal the stream contains a tile from a HEIF grid.
+ */
+#define AV_DISPOSITION_TILE                 (1 << 21)
 
 /**
  * @return The AV_DISPOSITION_* flag corresponding to disp or a negative error
diff --git a/libavformat/dump.c b/libavformat/dump.c
index ef4a8df6fe..6917df96b9 100644
--- a/libavformat/dump.c
+++ b/libavformat/dump.c
@@ -514,6 +514,10 @@ static void dump_sidedata(void *ctx, const AVStream *st, const char *indent,
             av_log(ctx, log_level, "SMPTE ST 12-1:2014: ");
             dump_s12m_timecode(ctx, st, sd, log_level);
             break;
+        case AV_PKT_DATA_TILE_INFO:
+            av_log(ctx, log_level, "HEIF tile info: tile %u/%u",
+                   AV_RL32(sd->data) + 1U, AV_RL32(sd->data + 4));
+            break;
         default:
             av_log(ctx, log_level, "unknown side data type %d "
                    "(%"SIZE_SPECIFIER" bytes)", sd->type, sd->size);
@@ -640,6 +644,8 @@ static void dump_stream_format(const AVFormatContext *ic, int i,
         av_log(NULL, log_level, " (still image)");
     if (st->disposition & AV_DISPOSITION_NON_DIEGETIC)
         av_log(NULL, log_level, " (non-diegetic)");
+    if (st->disposition & AV_DISPOSITION_TILE)
+        av_log(NULL, log_level, " (tile)");
     av_log(NULL, log_level, "\n");
 
     dump_metadata(NULL, st->metadata, extra_indent, log_level);
diff --git a/libavformat/isom.h b/libavformat/isom.h
index 21caaac256..c6a8507a68 100644
--- a/libavformat/isom.h
+++ b/libavformat/isom.h
@@ -267,10 +267,10 @@ typedef struct HEIFItem {
     int item_id;
     int64_t extent_length;
     int64_t extent_offset;
-    int64_t size;
     int width;
     int height;
     int type;
+    int is_idat_relative;
 } HEIFItem;
 
 typedef struct MOVContext {
@@ -336,6 +336,11 @@ typedef struct MOVContext {
     int cur_item_id;
     HEIFItem *heif_info;
     int heif_info_size;
+    int grid_item_id;
+    int thmb_item_id;
+    int16_t *tile_id_list;
+    int nb_tiles;
+    int64_t idat_offset;
     int interleaved_read;
 } MOVContext;
 
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 03ca62895e..3d4de3028b 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -185,6 +185,30 @@ static int mov_read_mac_string(MOVContext *c, AVIOContext *pb, int len,
     return p - dst;
 }
 
+static AVStream *get_curr_st(MOVContext *c)
+{
+    AVStream *st = NULL;
+
+    if (c->fc->nb_streams < 1)
+        return NULL;
+
+    for (int i = 0; i < c->heif_info_size; i++) {
+        HEIFItem *item = &c->heif_info[i];
+
+        if (!item->st)
+            continue;
+        if (item->st->id != c->cur_item_id)
+            continue;
+
+        st = item->st;
+        break;
+    }
+    if (!st)
+        st = c->fc->streams[c->fc->nb_streams-1];
+
+    return st;
+}
+
 static int mov_read_covr(MOVContext *c, AVIOContext *pb, int type, int len)
 {
     AVStream *st;
@@ -1767,9 +1791,9 @@ static int mov_read_colr(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     uint16_t color_primaries, color_trc, color_matrix;
     int ret;
 
-    if (c->fc->nb_streams < 1)
+    st = get_curr_st(c);
+    if (!st)
         return 0;
-    st = c->fc->streams[c->fc->nb_streams - 1];
 
     ret = ffio_read_size(pb, color_parameter_type, 4);
     if (ret < 0)
@@ -2117,9 +2141,9 @@ static int mov_read_glbl(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     AVStream *st;
     int ret;
 
-    if (c->fc->nb_streams < 1)
+    st = get_curr_st(c);
+    if (!st)
         return 0;
-    st = c->fc->streams[c->fc->nb_streams-1];
 
     if ((uint64_t)atom.size > (1<<30))
         return AVERROR_INVALIDDATA;
@@ -4946,12 +4970,10 @@ static int heif_add_stream(MOVContext *c, HEIFItem *item)
     st->codecpar->codec_type = AVMEDIA_TYPE_VIDEO;
     st->codecpar->codec_id = mov_codec_id(st, item->type);
     sc->ffindex = st->index;
-    c->trak_index = st->index;
     st->avg_frame_rate.num = st->avg_frame_rate.den = 1;
     st->time_base.num = st->time_base.den = 1;
     st->nb_frames = 1;
     sc->time_scale = 1;
-    sc = st->priv_data;
     sc->pb = c->fc->pb;
     sc->pb_is_copied = 1;
 
@@ -7779,11 +7801,55 @@ static int mov_read_pitm(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     return atom.size;
 }
 
+static int mov_read_idat(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    c->idat_offset = avio_tell(pb);
+    return 0;
+}
+
+static int read_image_grid(AVFormatContext *s, AVStreamGroupTileGrid *tile_grid,
+                           HEIFItem *item)
+{
+    MOVContext *c = s->priv_data;
+    int64_t offset = 0, pos = avio_tell(s->pb);
+    uint8_t flags;
+
+    if (!(s->pb->seekable & AVIO_SEEKABLE_NORMAL)) {
+        av_log(c->fc, AV_LOG_INFO, "grid box with non seekable input\n");
+        return AVERROR_PATCHWELCOME;
+    }
+    if (item->is_idat_relative) {
+        if (!c->idat_offset) {
+            av_log(c->fc, AV_LOG_ERROR, "missing idat box required by the image grid\n");
+            return AVERROR_INVALIDDATA;
+        }
+        offset = c->idat_offset;
+    }
+
+    avio_seek(s->pb, item->extent_offset + offset, SEEK_SET);
+
+    avio_r8(s->pb);    /* version */
+    flags = avio_r8(s->pb);
+
+    tile_grid->tile_rows = avio_r8(s->pb) + 1;
+    tile_grid->tile_cols = avio_r8(s->pb) + 1;
+    /* actual width and height of output image */
+    tile_grid->output_width  = (flags & 1) ? avio_rb32(s->pb) : avio_rb16(s->pb);
+    tile_grid->output_height = (flags & 1) ? avio_rb32(s->pb) : avio_rb16(s->pb);
+
+    av_log(c->fc, AV_LOG_TRACE, "grid: grid_rows %d grid_cols %d output_width %d output_height %d\n",
+           tile_grid->tile_rows, tile_grid->tile_cols, tile_grid->output_width, tile_grid->output_height);
+
+    avio_seek(s->pb, pos, SEEK_SET);
+
+    return 0;
+}
+
 static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 {
     int version, offset_size, length_size, base_offset_size, index_size;
     int item_count, extent_count;
-    uint64_t base_offset, extent_offset, extent_length;
+    int64_t base_offset, extent_offset, extent_length;
     uint8_t value;
 
     if (c->found_moov) {
@@ -7830,6 +7896,7 @@ static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
         avio_rb16(pb);  // data_reference_index.
         if (rb_size(pb, &base_offset, base_offset_size) < 0)
             return AVERROR_INVALIDDATA;
+        av_log(c->fc, AV_LOG_TRACE, "iloc: base_offset %"PRId64"\n", base_offset);
         extent_count = avio_rb16(pb);
         if (extent_count > 1) {
             // For still AVIF images, we only support one extent item.
@@ -7840,6 +7907,8 @@ static int mov_read_iloc(MOVContext *c, AVIOContext *pb, MOVAtom atom)
             if (rb_size(pb, &extent_offset, offset_size) < 0 ||
                 rb_size(pb, &extent_length, length_size) < 0)
                 return AVERROR_INVALIDDATA;
+            if (offset_type == 1)
+                c->heif_info[i].is_idat_relative = 1;
             c->heif_info[i].extent_length = extent_length;
             c->heif_info[i].extent_offset = base_offset + extent_offset;
             av_log(c->fc, AV_LOG_TRACE, "iloc: item_idx %d, offset_type %d, "
@@ -7878,10 +7947,6 @@ static int mov_read_infe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     av_log(c->fc, AV_LOG_TRACE, "infe: item_id %d, item_type %s, item_name %s\n",
            item_id, av_fourcc2str(item_type), item_name);
 
-    // Skip all but the primary item until support is added
-    if (item_id != c->primary_item_id)
-        return 0;
-
     if (size > 0)
         avio_skip(pb, size);
 
@@ -7895,6 +7960,9 @@ static int mov_read_infe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
         if (ret < 0)
             return ret;
         break;
+    case MKTAG('g','r','i','d'):
+        c->grid_item_id = item_id;
+        break;
     default:
         av_log(c->fc, AV_LOG_TRACE, "infe: ignoring item_type %s\n", av_fourcc2str(item_type));
         break;
@@ -7954,6 +8022,59 @@ static int mov_read_iref(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     return mov_read_default(c, pb, atom);
 }
 
+static int mov_read_dimg(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    int entries, i;
+    int from_item_id = avio_rb16(pb);
+
+    if (c->grid_item_id < 0) {
+        av_log(c->fc, AV_LOG_ERROR, "Missing grid information\n");
+        return AVERROR_INVALIDDATA;
+    }
+    if (from_item_id != c->grid_item_id) {
+        avpriv_request_sample(c->fc, "Derived item of type other than 'grid'");
+        return AVERROR_PATCHWELCOME;
+    }
+    entries = avio_rb16(pb);
+    c->tile_id_list = av_malloc_array(entries, sizeof(*c->tile_id_list));
+    if (!c->tile_id_list)
+        return AVERROR(ENOMEM);
+    /* 'to' item ids */
+    for (i = 0; i < entries; i++)
+        c->tile_id_list[i] = avio_rb16(pb);
+    c->nb_tiles = entries;
+
+    av_log(c->fc, AV_LOG_TRACE, "dimg: from_item_id %d, entries %d\n",
+           from_item_id, entries);
+
+    return 0;
+}
+
+static int mov_read_thmb(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+    int entries;
+    int to_item_id, from_item_id = avio_rb16(pb);
+
+    entries = avio_rb16(pb);
+    if (entries > 1) {
+        avpriv_request_sample(c->fc, "More than one thmb entry");
+        return AVERROR_PATCHWELCOME;
+    }
+    /* 'to' item ids */
+    to_item_id = avio_rb16(pb);
+
+    if (to_item_id != c->primary_item_id ||
+        to_item_id != c->grid_item_id)
+        return 0;
+
+    c->thmb_item_id = from_item_id;
+
+    av_log(c->fc, AV_LOG_TRACE, "thmb: from_item_id %d, entries %d\n",
+           from_item_id, entries);
+
+    return 0;
+}
+
 static int mov_read_ispe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
 {
     uint32_t width, height;
@@ -7969,15 +8090,17 @@ static int mov_read_ispe(MOVContext *c, AVIOContext *pb, MOVAtom atom)
     width  = avio_rb32(pb);
     height = avio_rb32(pb);
 
-    av_log(c->fc, AV_LOG_TRACE, "ispe: item_id %d, width %u, height %u\n",
+    av_log(c->fc, AV_LOG_TRACE, "ispe: cur_item_id %d, width %u, height %u\n",
            c->cur_item_id, width, height);
 
     for (int i = 0; i < c->heif_info_size; i++) {
-        if (c->heif_info[i].item_id == c->cur_item_id) {
-            c->heif_info[i].width  = width;
-            c->heif_info[i].height = height;
-            break;
-        }
+        HEIFItem *item = &c->heif_info[i];
+        if (item->item_id != c->cur_item_id)
+            continue;
+
+        item->width  = width;
+        item->height = height;
+        break;
     }
 
     return 0;
@@ -8074,10 +8197,6 @@ static int mov_read_iprp(MOVContext *c, AVIOContext *pb, MOVAtom atom)
             av_log(c->fc, AV_LOG_TRACE, "ipma: property_index %d, item_id %d, item_type %s\n",
                    index + 1, item_id, av_fourcc2str(ref->type));
 
-            // Skip properties referencing items other than the primary item until support is added
-            if (item_id != c->primary_item_id)
-                continue;
-
             c->cur_item_id = item_id;
 
             ret = mov_read_default(c, &ref->b.pub,
@@ -8205,6 +8324,9 @@ static const MOVParseTableEntry mov_default_parse_table[] = {
 { MKTAG('p','c','m','C'), mov_read_pcmc }, /* PCM configuration box */
 { MKTAG('p','i','t','m'), mov_read_pitm },
 { MKTAG('e','v','c','C'), mov_read_glbl },
+{ MKTAG('d','i','m','g'), mov_read_dimg },
+{ MKTAG('t','h','m','b'), mov_read_thmb },
+{ MKTAG('i','d','a','t'), mov_read_idat },
 { MKTAG('i','r','e','f'), mov_read_iref },
 { MKTAG('i','s','p','e'), mov_read_ispe },
 { MKTAG('i','p','r','p'), mov_read_iprp },
@@ -8713,6 +8835,7 @@ static int mov_read_close(AVFormatContext *s)
     av_freep(&mov->aes_decrypt);
     av_freep(&mov->chapter_tracks);
     av_freep(&mov->heif_info);
+    av_freep(&mov->tile_id_list);
 
     return 0;
 }
@@ -8852,6 +8975,116 @@ fail:
     return ret;
 }
 
+static int mov_parse_tiles(AVFormatContext *s)
+{
+    MOVContext *mov = s->priv_data;
+    AVStreamGroup *stg = avformat_stream_group_create(s, AV_STREAM_GROUP_PARAMS_TILE_GRID, NULL);
+    AVStreamGroupTileGrid *tile_grid;
+    int err;
+
+    if (!stg)
+        return AVERROR(ENOMEM);
+
+    tile_grid = stg->params.tile_grid;
+
+    av_assert0(mov->grid_item_id >= 0);
+    for (int i = 0; i < mov->heif_info_size; i++) {
+        HEIFItem *item = &mov->heif_info[i];
+
+        if (item->item_id != mov->grid_item_id)
+            continue;
+        err = read_image_grid(s, tile_grid, item);
+        if (err < 0)
+            return err;
+        stg->id = item->item_id;
+        break;
+    }
+
+    for (int i = 0; i < mov->nb_tiles; i++) {
+        int tile_id = mov->tile_id_list[i];
+
+        for (int j = 0; j < mov->heif_info_size; j++) {
+            HEIFItem *item = &mov->heif_info[j];
+            AVStream *st = item->st;
+            AVPacketSideData *sd;
+            MOVStreamContext *sc;
+            int64_t offset = 0;
+
+            if (item->item_id != tile_id)
+                continue;
+            if (!st) {
+                av_log(s, AV_LOG_ERROR, "HEIF tile %d doesn't reference a stream\n", tile_id);
+                return AVERROR_INVALIDDATA;
+            }
+            if (item->is_idat_relative) {
+                if (!mov->idat_offset) {
+                    av_log(s, AV_LOG_ERROR, "Missing idat box for HEIF tile %d\n", tile_id);
+                    return AVERROR_INVALIDDATA;
+                }
+                offset = mov->idat_offset;
+            }
+
+            st->codecpar->width  = item->width;
+            st->codecpar->height = item->height;
+
+            err = avformat_stream_group_add_stream(stg, st);
+            if (err == AVERROR(EEXIST))
+                return AVERROR_INVALIDDATA;
+            else if (err < 0)
+                return err;
+
+            sc = st->priv_data;
+            sc->sample_sizes[0]  = item->extent_length;
+            sc->chunk_offsets[0] = item->extent_offset + offset;
+
+            st->disposition |= AV_DISPOSITION_TILE;
+
+            mov_build_index(mov, st);
+
+            sd = av_packet_side_data_new(&st->codecpar->coded_side_data,
+                                         &st->codecpar->nb_coded_side_data,
+                                         AV_PKT_DATA_TILE_INFO,
+                                         sizeof(uint32_t) * 2, 0);
+            if (!sd)
+                return AVERROR(ENOMEM);
+
+            AV_WL32(sd->data, i);
+            AV_WL32(sd->data + 4, mov->nb_tiles);
+            break;
+        }
+    }
+
+    for (int i = 0; i < mov->heif_info_size; i++) {
+        HEIFItem *item = &mov->heif_info[i];
+        AVStream *st = item->st;
+        MOVStreamContext *sc;
+        int64_t offset = 0;
+
+        if (item->item_id != mov->thmb_item_id)
+            continue;
+
+        if (!st) {
+            av_log(s, AV_LOG_ERROR, "HEIF thumbnail doesn't reference a stream\n");
+            return AVERROR_INVALIDDATA;
+        }
+        if (item->is_idat_relative) {
+            if (!mov->idat_offset) {
+                av_log(s, AV_LOG_ERROR, "Missing idat box for HEIF thumbnail\n");
+                return AVERROR_INVALIDDATA;
+            }
+            offset = mov->idat_offset;
+        }
+
+        sc = st->priv_data;
+        sc->sample_sizes[0]  = item->extent_length;
+        sc->chunk_offsets[0] = item->extent_offset + offset;
+
+        mov_build_index(mov, st);
+    }
+
+    return 0;
+}
+
 static int mov_read_header(AVFormatContext *s)
 {
     MOVContext *mov = s->priv_data;
@@ -8868,6 +9101,9 @@ static int mov_read_header(AVFormatContext *s)
 
     mov->fc = s;
     mov->trak_index = -1;
+    mov->grid_item_id = -1;
+    mov->thmb_item_id = -1;
+    mov->primary_item_id = -1;
     /* .mov and .mp4 aren't streamable anyway (only progressive download if moov is before mdat) */
     if (pb->seekable & AVIO_SEEKABLE_NORMAL)
         atom.size = avio_size(pb);
@@ -8890,23 +9126,30 @@ static int mov_read_header(AVFormatContext *s)
     av_log(mov->fc, AV_LOG_TRACE, "on_parse_exit_offset=%"PRId64"\n", avio_tell(pb));
 
     if (mov->found_iloc) {
-        for (i = 0; i < mov->heif_info_size; i++) {
-            HEIFItem *item = &mov->heif_info[i];
-            MOVStreamContext *sc;
-            AVStream *st;
+        if (mov->nb_tiles) {
+            err = mov_parse_tiles(s);
+            if (err < 0)
+                return err;
+        } else
+            for (i = 0; i < mov->heif_info_size; i++) {
+                HEIFItem *item = &mov->heif_info[i];
+                AVStream *st = item->st;
+                MOVStreamContext *sc;
 
-            if (!item->st)
-                continue;
+                if (!st)
+                    continue;
 
-            st = item->st;
-            sc = st->priv_data;
-            st->codecpar->width  = item->width;
-            st->codecpar->height = item->height;
-            sc->sample_sizes[0]  = item->extent_length;
-            sc->chunk_offsets[0] = item->extent_offset;
+                sc = st->priv_data;
+                st->codecpar->width  = item->width;
+                st->codecpar->height = item->height;
+                sc->sample_sizes[0]  = item->extent_length;
+                sc->chunk_offsets[0] = item->extent_offset;
 
-            mov_build_index(mov, st);
-        }
+                if (item->item_id == mov->primary_item_id)
+                    st->disposition |= AV_DISPOSITION_DEFAULT;
+
+                mov_build_index(mov, st);
+            }
     }
 
     if (pb->seekable & AVIO_SEEKABLE_NORMAL) {
-- 
2.43.0

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".