[Bf-blender-cvs] [85c08c9717b] blender-v2.93-release: VSE: Flush audio encode after finishing video export

[Sebastian Parborg]

Commit: 85c08c9717b104e1b537476604b6a0fa8e58e477
Author: Sebastian Parborg
Date:   Mon Jul 5 14:16:02 2021 +0200
Branches: blender-v2.93-release
https://developer.blender.org/rB85c08c9717b104e1b537476604b6a0fa8e58e477

VSE: Flush audio encode after finishing video export

We didn't flush audio after encoding finished which lead to audio
packets being lost.

In addition to this the audio timestamps were wrong because we
incremented the current audio time before using it.

Reviewed By: Richard Antalik

Differential Revision: http://developer.blender.org/D11916

===================================================================

M	source/blender/blenkernel/intern/writeffmpeg.c

===================================================================

diff --git a/source/blender/blenkernel/intern/writeffmpeg.c b/source/blender/blenkernel/intern/writeffmpeg.c
index 5ab9bdcbee6..2474c43c3e8 100644
--- a/source/blender/blenkernel/intern/writeffmpeg.c
+++ b/source/blender/blenkernel/intern/writeffmpeg.c
@@ -148,7 +148,6 @@ static int write_audio_frame(FFMpegContext *context)
 
   AUD_Device_read(
       context->audio_mixdown_device, context->audio_input_buffer, context->audio_input_samples);
-  context->audio_time += (double)context->audio_input_samples / (double)c->sample_rate;
 
   frame = av_frame_alloc();
   frame->pts = context->audio_time / av_q2d(c->time_base);
@@ -183,7 +182,7 @@ static int write_audio_frame(FFMpegContext *context)
                            context->audio_input_samples * c->channels * context->audio_sample_size,
                            1);
 
-  int success = 0;
+  int success = 1;
 
   int ret = avcodec_send_frame(c, frame);
   if (ret < 0) {
@@ -368,7 +367,7 @@ static int write_video_frame(FFMpegContext *context, int cfra, AVFrame *frame, R
   return success;
 }
 
-/* read and encode a frame of audio from the buffer */
+/* read and encode a frame of video from the buffer */
 static AVFrame *generate_video_frame(FFMpegContext *context, const uint8_t *pixels)
 {
   AVCodecParameters *codec = context->video_stream->codecpar;
@@ -1220,9 +1219,8 @@ fail:
  * parameter.
  * </p>
  */
-static void flush_ffmpeg(FFMpegContext *context)
+static void flush_ffmpeg(AVCodecContext *c, AVStream *stream, AVFormatContext *outfile)
 {
-  AVCodecContext *c = context->video_codec;
   AVPacket *packet = av_packet_alloc();
 
   avcodec_send_frame(c, NULL);
@@ -1241,13 +1239,13 @@ static void flush_ffmpeg(FFMpegContext *context)
       break;
     }
 
-    packet->stream_index = context->video_stream->index;
-    av_packet_rescale_ts(packet, c->time_base, context->video_stream->time_base);
+    packet->stream_index = stream->index;
+    av_packet_rescale_ts(packet, c->time_base, stream->time_base);
 #  ifdef FFMPEG_USE_DURATION_WORKAROUND
-    my_guess_pkt_duration(context->outfile, context->video_stream, packet);
+    my_guess_pkt_duration(context->outfile, stream, packet);
 #  endif
 
-    int write_ret = av_interleaved_write_frame(context->outfile, packet);
+    int write_ret = av_interleaved_write_frame(outfile, packet);
     if (write_ret != 0) {
       fprintf(stderr, "Error writing delayed frame: %s\n", av_err2str(write_ret));
       break;
@@ -1390,12 +1388,13 @@ static void end_ffmpeg_impl(FFMpegContext *context, int is_autosplit);
 #  ifdef WITH_AUDASPACE
 static void write_audio_frames(FFMpegContext *context, double to_pts)
 {
-  int finished = 0;
+  AVCodecContext *c = context->audio_codec;
 
-  while (context->audio_stream && !finished) {
-    if ((context->audio_time >= to_pts) || (write_audio_frame(context))) {
-      finished = 1;
+  while (context->audio_stream) {
+    if ((context->audio_time >= to_pts) || !write_audio_frame(context)) {
+      break;
     }
+    context->audio_time += (double)context->audio_input_samples / (double)c->sample_rate;
   }
 }
 #  endif
@@ -1416,9 +1415,6 @@ int BKE_ffmpeg_append(void *context_v,
 
   PRINT("Writing frame %i, render width=%d, render height=%d\n", frame, rectx, recty);
 
-  /* why is this done before writing the video frame and again at end_ffmpeg? */
-  //  write_audio_frames(frame / (((double)rd->frs_sec) / rd->frs_sec_base));
-
   if (context->video_stream) {
     avframe = generate_video_frame(context, (unsigned char *)pixels);
     success = (avframe && write_video_frame(context, frame - start_frame, avframe, reports));
@@ -1454,8 +1450,13 @@ static void end_ffmpeg_impl(FFMpegContext *context, int is_autosplit)
 #  endif
 
   if (context->video_stream) {
-    PRINT("Flushing delayed frames...\n");
-    flush_ffmpeg(context);
+    PRINT("Flushing delayed video frames...\n");
+    flush_ffmpeg(context->video_codec, context->video_stream, context->outfile);
+  }
+
+  if (context->audio_stream) {
+    PRINT("Flushing delayed audio frames...\n");
+    flush_ffmpeg(context->audio_codec, context->audio_stream, context->outfile);
   }
 
   if (context->outfile) {