[FFmpeg-devel] [PATCH v20 03/20] avcodec/subtitles: Introduce new frame-based subtitle decoding API

Soft Works softworkz at hotmail.com
Sun Dec 5 18:23:36 EET 2021


- Add avcodec_decode_subtitle3 which takes subtitle frames,
  serving as compatibility shim to legacy subtitle decoding
- Add additional methods for conversion between old and new API

Signed-off-by: softworkz <softworkz at hotmail.com>
---
 doc/APIchanges          |   7 ++
 libavcodec/avcodec.h    |   8 +-
 libavcodec/codec_desc.c |  11 +++
 libavcodec/codec_desc.h |   8 ++
 libavcodec/decode.c     |  54 ++++++++++--
 libavcodec/internal.h   |  16 ++++
 libavcodec/utils.c      | 182 ++++++++++++++++++++++++++++++++++++++++
 7 files changed, 277 insertions(+), 9 deletions(-)

diff --git a/doc/APIchanges b/doc/APIchanges
index 8c63ea0311..49f1a28f71 100644
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -14,6 +14,13 @@ libavutil:     2021-04-27
 
 API changes, most recent first:
 
+2021-12-05 - xxxxxxxxxx - lavc 60.1.100 - codec_desc.h
+  Add avcodec_descriptor_get_subtitle_format()
+
+2021-12-05 - xxxxxxxxxx - lavc 60.1.100 - avcodec.h
+  Deprecate avsubtitle_free()
+  Deprecate avcodec_decode_subtitle2(), use regular decode api now
+
 2021-12-05 - xxxxxxxxxx - lavu 58.1.100 - frame.h
   Add AVMediaType field to AVFrame
   Add Fields for carrying subtitle data to AVFrame
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index b05c12e47e..3e734d3003 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -1675,7 +1675,7 @@ typedef struct AVCodecContext {
 
     /**
      * Header containing style information for text subtitles.
-     * For SUBTITLE_ASS subtitle type, it should contain the whole ASS
+     * For AV_SUBTITLE_FMT_ASS subtitle type, it should contain the whole ASS
      * [Script Info] and [V4+ Styles] section, plus the [Events] line and
      * the Format line following. It shouldn't include any Dialogue line.
      * - encoding: Set/allocated/freed by user (before avcodec_open2())
@@ -2417,7 +2417,10 @@ int avcodec_close(AVCodecContext *avctx);
  * Free all allocated data in the given subtitle struct.
  *
  * @param sub AVSubtitle to free.
+ *
+ * @deprecated Use the regular frame based encode and decode APIs instead.
  */
+attribute_deprecated
 void avsubtitle_free(AVSubtitle *sub);
 
 /**
@@ -2510,7 +2513,10 @@ enum AVChromaLocation avcodec_chroma_pos_to_enum(int xpos, int ypos);
  *                 must be freed with avsubtitle_free if *got_sub_ptr is set.
  * @param[in,out] got_sub_ptr Zero if no subtitle could be decompressed, otherwise, it is nonzero.
  * @param[in] avpkt The input AVPacket containing the input buffer.
+ *
+ * @deprecated Use the new decode API (avcodec_send_packet, avcodec_receive_frame) instead.
  */
+attribute_deprecated
 int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
                             int *got_sub_ptr,
                             AVPacket *avpkt);
diff --git a/libavcodec/codec_desc.c b/libavcodec/codec_desc.c
index 0974ee03de..e48e4532ba 100644
--- a/libavcodec/codec_desc.c
+++ b/libavcodec/codec_desc.c
@@ -3548,3 +3548,14 @@ enum AVMediaType avcodec_get_type(enum AVCodecID codec_id)
     const AVCodecDescriptor *desc = avcodec_descriptor_get(codec_id);
     return desc ? desc->type : AVMEDIA_TYPE_UNKNOWN;
 }
+
+enum AVSubtitleType avcodec_descriptor_get_subtitle_format(const AVCodecDescriptor *codec_descriptor)
+{
+    if(codec_descriptor->props & AV_CODEC_PROP_BITMAP_SUB)
+        return AV_SUBTITLE_FMT_BITMAP;
+
+    if(codec_descriptor->props & AV_CODEC_PROP_TEXT_SUB)
+        return AV_SUBTITLE_FMT_ASS;
+
+    return AV_SUBTITLE_FMT_UNKNOWN;
+}
diff --git a/libavcodec/codec_desc.h b/libavcodec/codec_desc.h
index 126b52df47..ba68d24e0e 100644
--- a/libavcodec/codec_desc.h
+++ b/libavcodec/codec_desc.h
@@ -121,6 +121,14 @@ const AVCodecDescriptor *avcodec_descriptor_next(const AVCodecDescriptor *prev);
  */
 const AVCodecDescriptor *avcodec_descriptor_get_by_name(const char *name);
 
+/**
+ * Return subtitle format from a codec descriptor
+ *
+ * @param codec_descriptor codec descriptor
+ * @return                 the subtitle type (e.g. bitmap, text)
+ */
+enum AVSubtitleType avcodec_descriptor_get_subtitle_format(const AVCodecDescriptor *codec_descriptor);
+
 /**
  * @}
  */
diff --git a/libavcodec/decode.c b/libavcodec/decode.c
index 52bf5dcd33..ac267f0df6 100644
--- a/libavcodec/decode.c
+++ b/libavcodec/decode.c
@@ -576,6 +576,37 @@ static int decode_receive_frame_internal(AVCodecContext *avctx, AVFrame *frame)
     return ret;
 }
 
+static int decode_subtitle2_priv(AVCodecContext *avctx, AVSubtitle *sub,
+                                 int *got_sub_ptr, AVPacket *avpkt);
+
+static int decode_subtitle_shim(AVCodecContext *avctx, AVFrame *frame, AVPacket *avpkt)
+{
+    int ret, got_sub_ptr = 0;
+    AVSubtitle subtitle = { 0 };
+
+    if (frame->buf[0])
+        return AVERROR(EAGAIN);
+
+    av_frame_unref(frame);
+
+    ret = decode_subtitle2_priv(avctx, &subtitle, &got_sub_ptr, avpkt);
+
+    if (ret >= 0 && got_sub_ptr) {
+        frame->type = AVMEDIA_TYPE_SUBTITLE;
+        frame->format = subtitle.format;
+        ret = av_frame_get_buffer2(frame, 0);
+
+        if (ret >= 0)
+            ret = ff_frame_put_subtitle(frame, &subtitle);
+
+        frame->pkt_dts = avpkt->dts;
+    }
+
+    avsubtitle_free(&subtitle);
+
+    return ret;
+}
+
 int attribute_align_arg avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt)
 {
     AVCodecInternal *avci = avctx->internal;
@@ -590,6 +621,9 @@ int attribute_align_arg avcodec_send_packet(AVCodecContext *avctx, const AVPacke
     if (avpkt && !avpkt->size && avpkt->data)
         return AVERROR(EINVAL);
 
+    if (avctx->codec_type == AVMEDIA_TYPE_SUBTITLE)
+        return decode_subtitle_shim(avctx, avci->buffer_frame, avpkt);
+
     av_packet_unref(avci->buffer_pkt);
     if (avpkt && (avpkt->data || avpkt->side_data_elems)) {
         ret = av_packet_ref(avci->buffer_pkt, avpkt);
@@ -651,7 +685,9 @@ int attribute_align_arg avcodec_receive_frame(AVCodecContext *avctx, AVFrame *fr
 
     if (avci->buffer_frame->buf[0]) {
         av_frame_move_ref(frame, avci->buffer_frame);
-    } else {
+    } else if (avctx->codec_type == AVMEDIA_TYPE_SUBTITLE)
+        return AVERROR(EAGAIN);
+    else {
         ret = decode_receive_frame_internal(avctx, frame);
         if (ret < 0)
             return ret;
@@ -802,9 +838,8 @@ static int utf8_check(const uint8_t *str)
     return 1;
 }
 
-int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
-                             int *got_sub_ptr,
-                             AVPacket *avpkt)
+static int decode_subtitle2_priv(AVCodecContext *avctx, AVSubtitle *sub,
+                             int *got_sub_ptr, AVPacket *avpkt)
 {
     int ret = 0;
 
@@ -844,10 +879,7 @@ int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
                                                  avctx->pkt_timebase, ms);
         }
 
-        if (avctx->codec_descriptor->props & AV_CODEC_PROP_BITMAP_SUB)
-            sub->format = 0;
-        else if (avctx->codec_descriptor->props & AV_CODEC_PROP_TEXT_SUB)
-            sub->format = 1;
+        sub->format = avcodec_descriptor_get_subtitle_format(avctx->codec_descriptor);
 
         for (unsigned i = 0; i < sub->num_rects; i++) {
             if (avctx->sub_charenc_mode != FF_SUB_CHARENC_MODE_IGNORE &&
@@ -871,6 +903,12 @@ int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
     return ret;
 }
 
+int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
+                             int *got_sub_ptr, AVPacket *avpkt)
+{
+    return decode_subtitle2_priv(avctx, sub, got_sub_ptr, avpkt);
+}
+
 enum AVPixelFormat avcodec_default_get_format(struct AVCodecContext *avctx,
                                               const enum AVPixelFormat *fmt)
 {
diff --git a/libavcodec/internal.h b/libavcodec/internal.h
index a62f8dbd4e..10443ed2d1 100644
--- a/libavcodec/internal.h
+++ b/libavcodec/internal.h
@@ -363,6 +363,22 @@ int ff_int_from_list_or_default(void *ctx, const char * val_name, int val,
 
 void ff_dvdsub_parse_palette(uint32_t *palette, const char *p);
 
+/**
+ * Copies subtitle data from AVSubtitle to AVFrame.
+ *
+ * @deprecated This is a compatibility method for interoperability with
+ * the legacy subtitle API.
+ */
+int ff_frame_put_subtitle(AVFrame* frame, const AVSubtitle* sub);
+
+/**
+ * Copies subtitle data from AVFrame to AVSubtitle.
+ *
+ * @deprecated This is a compatibility method for interoperability with
+ * the legacy subtitle API.
+ */
+int ff_frame_get_subtitle(AVSubtitle* sub, AVFrame* frame);
+
 #if defined(_WIN32) && CONFIG_SHARED && !defined(BUILDING_avcodec)
 #    define av_export_avcodec __declspec(dllimport)
 #else
diff --git a/libavcodec/utils.c b/libavcodec/utils.c
index a91a54b0dc..613c580c19 100644
--- a/libavcodec/utils.c
+++ b/libavcodec/utils.c
@@ -824,6 +824,188 @@ int av_get_audio_frame_duration(AVCodecContext *avctx, int frame_bytes)
     return FFMAX(0, duration);
 }
 
+static int subtitle_area2rect(AVSubtitleRect *dst, const AVSubtitleArea *src)
+{
+    dst->x         =  src->x;
+    dst->y         =  src->y;
+    dst->w         =  src->w;
+    dst->h         =  src->h;
+    dst->nb_colors =  src->nb_colors;
+    dst->type      =  src->type;
+    dst->flags     =  src->flags;
+
+    switch (dst->type) {
+    case AV_SUBTITLE_FMT_BITMAP:
+
+        if (src->h > 0 && src->w > 0 && src->buf[0]) {
+            uint32_t *pal;
+            AVBufferRef *buf = src->buf[0];
+            dst->data[0] = av_mallocz(buf->size);
+            memcpy(dst->data[0], buf->data, buf->size);
+            dst->linesize[0] = src->linesize[0];
+
+            dst->data[1] = av_mallocz(256 * 4);
+            pal = (uint32_t *)dst->data[1];
+
+            for (unsigned i = 0; i < 256; i++) {
+                pal[i] = src->pal[i];
+            }
+        }
+
+        break;
+    case AV_SUBTITLE_FMT_TEXT:
+
+        if (src->text)
+            dst->text = av_strdup(src->text);
+        else
+            dst->text = av_strdup("");
+
+        if (!dst->text)
+            return AVERROR(ENOMEM);
+
+        break;
+    case AV_SUBTITLE_FMT_ASS:
+
+        if (src->ass)
+            dst->ass = av_strdup(src->ass);
+        else
+            dst->ass = av_strdup("");
+
+        if (!dst->ass)
+            return AVERROR(ENOMEM);
+
+        break;
+    default:
+
+        av_log(NULL, AV_LOG_ERROR, "Subtitle rect has invalid format: %d", dst->type);
+        return AVERROR(EINVAL);
+    }
+
+    return 0;
+}
+
+static int subtitle_rect2area(AVSubtitleArea *dst, const AVSubtitleRect *src)
+{
+    dst->x         =  src->x;
+    dst->y         =  src->y;
+    dst->w         =  src->w;
+    dst->h         =  src->h;
+    dst->nb_colors =  src->nb_colors;
+    dst->type      =  src->type;
+    dst->flags     =  src->flags;
+
+    switch (dst->type) {
+    case AV_SUBTITLE_FMT_BITMAP:
+
+        if (src->h > 0 && src->w > 0 && src->data[0]) {
+            AVBufferRef *buf = av_buffer_allocz(src->h * src->linesize[0]);
+            memcpy(buf->data, src->data[0], buf->size);
+
+            dst->buf[0] = buf;
+            dst->linesize[0] = src->linesize[0];
+        }
+
+        if (src->data[1]) {
+            uint32_t *pal = (uint32_t *)src->data[1];
+
+            for (unsigned i = 0; i < 256; i++) {
+                dst->pal[i] = pal[i];
+            }
+        }
+
+        break;
+    case AV_SUBTITLE_FMT_TEXT:
+
+        if (src->text) {
+            dst->text = av_strdup(src->text);
+            if (!dst->text)
+                return AVERROR(ENOMEM);
+        }
+
+        break;
+    case AV_SUBTITLE_FMT_ASS:
+
+        if (src->ass) {
+            dst->ass = av_strdup(src->ass);
+            if (!dst->ass)
+                return AVERROR(ENOMEM);
+        }
+
+        break;
+    default:
+
+        av_log(NULL, AV_LOG_ERROR, "Subtitle area has invalid format: %d", dst->type);
+        return AVERROR(EINVAL);
+    }
+
+    return 0;
+}
+
+/**
+ * Copies subtitle data from AVSubtitle (deprecated) to AVFrame
+ *
+ * @note This is a compatibility method for conversion to the legacy API
+ */
+int ff_frame_put_subtitle(AVFrame *frame, const AVSubtitle *sub)
+{
+    frame->format = sub->format;
+    frame->subtitle_start_time = sub->start_display_time;
+    frame->subtitle_end_time = sub->end_display_time;
+    frame->subtitle_pts = sub->pts;
+
+    if (sub->num_rects) {
+        frame->subtitle_areas = av_malloc_array(sub->num_rects, sizeof(AVSubtitleArea*));
+        if (!frame->subtitle_areas)
+            return AVERROR(ENOMEM);
+
+        for (unsigned i = 0; i < sub->num_rects; i++) {
+            int ret;
+            frame->subtitle_areas[i] = av_mallocz(sizeof(AVSubtitleArea));
+            if (!frame->subtitle_areas[i])
+                return AVERROR(ENOMEM);
+            ret = subtitle_rect2area(frame->subtitle_areas[i], sub->rects[i]);
+            if (ret < 0) {
+                frame->num_subtitle_areas = i;
+                return ret;
+            }
+        }
+    }
+
+    frame->num_subtitle_areas = sub->num_rects;
+    return 0;
+}
+
+/**
+ * Copies subtitle data from AVFrame to AVSubtitle (deprecated)
+ *
+ * @note This is a compatibility method for conversion to the legacy API
+ */
+int ff_frame_get_subtitle(AVSubtitle *sub, AVFrame *frame)
+{
+    sub->start_display_time = frame->subtitle_start_time;
+    sub->end_display_time = frame->subtitle_end_time;
+    sub->pts = frame->subtitle_pts;
+
+    if (frame->num_subtitle_areas) {
+        sub->rects = av_malloc_array(frame->num_subtitle_areas, sizeof(AVSubtitleRect*));
+        if (!sub->rects)
+            return AVERROR(ENOMEM);
+
+        for (unsigned i = 0; i < frame->num_subtitle_areas; i++) {
+            int ret;
+            sub->rects[i] = av_mallocz(sizeof(AVSubtitleRect));
+            ret = subtitle_area2rect(sub->rects[i], frame->subtitle_areas[i]);
+            if (ret < 0) {
+                sub->num_rects = i;
+                return ret;
+            }
+        }
+    }
+
+    sub->num_rects = frame->num_subtitle_areas;
+    return 0;
+}
+
 int av_get_audio_frame_duration2(AVCodecParameters *par, int frame_bytes)
 {
     int duration = get_audio_frame_duration(par->codec_id, par->sample_rate,
-- 
2.30.2.windows.1



More information about the ffmpeg-devel mailing list