diff mbox series

[FFmpeg-devel,v5,10/25] avfilter/avfilter: Handle subtitle frames

Message ID c931041103c216125e3d991be35f0e8d712b94ba.1656151077.git.ffmpegagent@gmail.com
State New
Headers show
Series Subtitle Filtering 2022 | expand

Checks

Context Check Description
yinshiyou/make_loongarch64 success Make finished
yinshiyou/make_fate_loongarch64 success Make fate finished
andriy/make_x86 success Make finished
andriy/make_fate_x86 success Make fate finished

Commit Message

softworkz June 25, 2022, 9:57 a.m. UTC
From: softworkz <softworkz@hotmail.com>

Signed-off-by: softworkz <softworkz@hotmail.com>
---
 libavfilter/avfilter.c      | 12 +++++++++---
 libavfilter/avfilter.h      | 11 +++++++++++
 libavfilter/avfiltergraph.c |  5 +++++
 libavfilter/formats.c       | 16 ++++++++++++++++
 libavfilter/formats.h       |  3 +++
 libavfilter/internal.h      | 18 +++++++++++++++---
 6 files changed, 59 insertions(+), 6 deletions(-)
diff mbox series

Patch

diff --git a/libavfilter/avfilter.c b/libavfilter/avfilter.c
index 28c5430c3e..30745f41cb 100644
--- a/libavfilter/avfilter.c
+++ b/libavfilter/avfilter.c
@@ -52,7 +52,8 @@  static void tlog_ref(void *ctx, AVFrame *ref, int end)
             ref->linesize[0], ref->linesize[1], ref->linesize[2], ref->linesize[3],
             ref->pts, ref->pkt_pos);
 
-    if (ref->width) {
+    switch(ref->type) {
+    case AVMEDIA_TYPE_VIDEO:
         ff_tlog(ctx, " a:%d/%d s:%dx%d i:%c iskey:%d type:%c",
                 ref->sample_aspect_ratio.num, ref->sample_aspect_ratio.den,
                 ref->width, ref->height,
@@ -60,12 +61,13 @@  static void tlog_ref(void *ctx, AVFrame *ref, int end)
                 ref->top_field_first ? 'T' : 'B',    /* Top / Bottom */
                 ref->key_frame,
                 av_get_picture_type_char(ref->pict_type));
-    }
-    if (ref->nb_samples) {
+        break;
+    case AVMEDIA_TYPE_AUDIO:
         ff_tlog(ctx, " cl:%"PRId64"d n:%d r:%d",
                 ref->channel_layout,
                 ref->nb_samples,
                 ref->sample_rate);
+        break;
     }
 
     ff_tlog(ctx, "]%s", end ? "\n" : "");
@@ -1013,6 +1015,10 @@  int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
             av_assert1(frame->width               == link->w);
             av_assert1(frame->height               == link->h);
         }
+    } else if (link->type == AVMEDIA_TYPE_SUBTITLE) {
+        if (frame->format != link->format) {
+            av_log(link->dst, AV_LOG_WARNING, "Subtitle format change from %d to %d\n", link->format, frame->format);
+        }
     } else {
         if (frame->format != link->format) {
             av_log(link->dst, AV_LOG_ERROR, "Format change is not supported\n");
diff --git a/libavfilter/avfilter.h b/libavfilter/avfilter.h
index 2e8197c9a6..c436f304bc 100644
--- a/libavfilter/avfilter.h
+++ b/libavfilter/avfilter.h
@@ -45,6 +45,7 @@ 
 #include "libavutil/log.h"
 #include "libavutil/samplefmt.h"
 #include "libavutil/pixfmt.h"
+#include "libavutil/subfmt.h"
 #include "libavutil/rational.h"
 
 #include "libavfilter/version_major.h"
@@ -349,6 +350,12 @@  typedef struct AVFilter {
          * and outputs use the same sample rate and channel count/layout.
          */
         const enum AVSampleFormat *samples_list;
+        /**
+         * Analogous to pixels, but delimited by AV_SUBTITLE_FMT_NONE
+         * and restricted to filters that only have AVMEDIA_TYPE_SUBTITLE
+         * inputs and outputs.
+         */
+        const enum AVSubtitleType *subs_list;
         /**
          * Equivalent to { pix_fmt, AV_PIX_FMT_NONE } as pixels_list.
          */
@@ -357,6 +364,10 @@  typedef struct AVFilter {
          * Equivalent to { sample_fmt, AV_SAMPLE_FMT_NONE } as samples_list.
          */
         enum AVSampleFormat sample_fmt;
+        /**
+         * Equivalent to { sub_fmt, AV_SUBTITLE_FMT_NONE } as subs_list.
+         */
+        enum AVSubtitleType sub_fmt;
     } formats;
 
     int priv_size;      ///< size of private data to allocate for the filter
diff --git a/libavfilter/avfiltergraph.c b/libavfilter/avfiltergraph.c
index b7dbfc063b..f37be0203b 100644
--- a/libavfilter/avfiltergraph.c
+++ b/libavfilter/avfiltergraph.c
@@ -309,6 +309,8 @@  static int filter_link_check_formats(void *log, AVFilterLink *link, AVFilterForm
             return ret;
         break;
 
+    case AVMEDIA_TYPE_SUBTITLE:
+        return 0;
     default:
         av_assert0(!"reached");
     }
@@ -439,6 +441,9 @@  static int query_formats(AVFilterGraph *graph, void *log_ctx)
             if (!link)
                 continue;
 
+            if (link->type == AVMEDIA_TYPE_SUBTITLE)
+                continue;
+
             neg = ff_filter_get_negotiation(link);
             av_assert0(neg);
             for (neg_step = 1; neg_step < neg->nb_mergers; neg_step++) {
diff --git a/libavfilter/formats.c b/libavfilter/formats.c
index e8c2888c0c..12585ed428 100644
--- a/libavfilter/formats.c
+++ b/libavfilter/formats.c
@@ -19,6 +19,7 @@ 
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
+#include "libavutil/subfmt.h"
 #include "libavutil/avassert.h"
 #include "libavutil/channel_layout.h"
 #include "libavutil/common.h"
@@ -491,6 +492,13 @@  AVFilterFormats *ff_all_formats(enum AVMediaType type)
                 return NULL;
             fmt++;
         }
+    } else if (type == AVMEDIA_TYPE_SUBTITLE) {
+        if (ff_add_format(&ret, AV_SUBTITLE_FMT_BITMAP) < 0)
+            return NULL;
+        if (ff_add_format(&ret, AV_SUBTITLE_FMT_ASS) < 0)
+            return NULL;
+        if (ff_add_format(&ret, AV_SUBTITLE_FMT_TEXT) < 0)
+            return NULL;
     }
 
     return ret;
@@ -774,6 +782,10 @@  int ff_default_query_formats(AVFilterContext *ctx)
         type    = AVMEDIA_TYPE_AUDIO;
         formats = ff_make_format_list(f->formats.samples_list);
         break;
+    case FF_FILTER_FORMATS_SUBFMTS_LIST:
+        type    = AVMEDIA_TYPE_SUBTITLE;
+        formats = ff_make_format_list(f->formats.subs_list);
+        break;
     case FF_FILTER_FORMATS_SINGLE_PIXFMT:
         type    = AVMEDIA_TYPE_VIDEO;
         formats = ff_make_formats_list_singleton(f->formats.pix_fmt);
@@ -782,6 +794,10 @@  int ff_default_query_formats(AVFilterContext *ctx)
         type    = AVMEDIA_TYPE_AUDIO;
         formats = ff_make_formats_list_singleton(f->formats.sample_fmt);
         break;
+    case FF_FILTER_FORMATS_SINGLE_SUBFMT:
+        type    = AVMEDIA_TYPE_SUBTITLE;
+        formats = ff_make_formats_list_singleton(f->formats.sub_fmt);
+        break;
     default:
         av_assert2(!"Unreachable");
     /* Intended fallthrough */
diff --git a/libavfilter/formats.h b/libavfilter/formats.h
index 22224dce2d..6cf952a059 100644
--- a/libavfilter/formats.h
+++ b/libavfilter/formats.h
@@ -183,6 +183,9 @@  av_warn_unused_result
 int ff_add_channel_layout(AVFilterChannelLayouts **l,
                           const AVChannelLayout *channel_layout);
 
+av_warn_unused_result
+int ff_add_subtitle_type(AVFilterFormats **avff, int64_t fmt);
+
 /**
  * Add *ref as a new reference to f.
  */
diff --git a/libavfilter/internal.h b/libavfilter/internal.h
index 6c8496879a..7972c5c6de 100644
--- a/libavfilter/internal.h
+++ b/libavfilter/internal.h
@@ -148,9 +148,11 @@  static av_always_inline int ff_filter_execute(AVFilterContext *ctx, avfilter_act
 
 enum FilterFormatsState {
     /**
-     * The default value meaning that this filter supports all formats
-     * and (for audio) sample rates and channel layouts/counts as long
-     * as these properties agree for all inputs and outputs.
+     * The default value meaning that this filter supports
+     * - For video:     all formats
+     * - For audio:     all sample rates and channel layouts/counts
+     * - For subtitles: all subtitle formats
+     * as long as these properties agree for all inputs and outputs.
      * This state is only allowed in case all inputs and outputs actually
      * have the same type.
      * The union is unused in this state.
@@ -161,8 +163,10 @@  enum FilterFormatsState {
     FF_FILTER_FORMATS_QUERY_FUNC,       ///< formats.query active.
     FF_FILTER_FORMATS_PIXFMT_LIST,      ///< formats.pixels_list active.
     FF_FILTER_FORMATS_SAMPLEFMTS_LIST,  ///< formats.samples_list active.
+    FF_FILTER_FORMATS_SUBFMTS_LIST,     ///< formats.subs_list active.
     FF_FILTER_FORMATS_SINGLE_PIXFMT,    ///< formats.pix_fmt active
     FF_FILTER_FORMATS_SINGLE_SAMPLEFMT, ///< formats.sample_fmt active.
+    FF_FILTER_FORMATS_SINGLE_SUBFMT,    ///< formats.sub_fmt active.
 };
 
 #define FILTER_QUERY_FUNC(func)        \
@@ -174,16 +178,24 @@  enum FilterFormatsState {
 #define FILTER_SAMPLEFMTS_ARRAY(array) \
         .formats.samples_list = array, \
         .formats_state        = FF_FILTER_FORMATS_SAMPLEFMTS_LIST
+#define FILTER_SUBFMTS_ARRAY(array) \
+        .formats.subs_list = array, \
+        .formats_state        = FF_FILTER_FORMATS_SUBFMTS_LIST
 #define FILTER_PIXFMTS(...)            \
     FILTER_PIXFMTS_ARRAY(((const enum AVPixelFormat []) { __VA_ARGS__, AV_PIX_FMT_NONE }))
 #define FILTER_SAMPLEFMTS(...)         \
     FILTER_SAMPLEFMTS_ARRAY(((const enum AVSampleFormat[]) { __VA_ARGS__, AV_SAMPLE_FMT_NONE }))
+#define FILTER_SUBFMTS(...)         \
+    FILTER_SUBFMTS_ARRAY(((const enum AVSubtitleType[]) { __VA_ARGS__, AV_SUBTITLE_FMT_NONE }))
 #define FILTER_SINGLE_PIXFMT(pix_fmt_)  \
         .formats.pix_fmt = pix_fmt_,    \
         .formats_state   = FF_FILTER_FORMATS_SINGLE_PIXFMT
 #define FILTER_SINGLE_SAMPLEFMT(sample_fmt_) \
         .formats.sample_fmt = sample_fmt_,   \
         .formats_state      = FF_FILTER_FORMATS_SINGLE_SAMPLEFMT
+#define FILTER_SINGLE_SUBFMT(sub_fmt_) \
+        .formats.sub_fmt = sub_fmt_,   \
+        .formats_state      = FF_FILTER_FORMATS_SINGLE_SUBFMT
 
 #define FILTER_INOUTPADS(inout, array) \
        .inout        = array, \