mlt_position video_expected;
int audio_index;
int video_index;
- int first_pts;
+ int64_t first_pts;
int64_t last_position;
int seekable;
int64_t current_position;
error = get_basic_info( self, profile, filename );
// Initialize position info
- self->first_pts = -1;
+ self->first_pts = AV_NOPTS_VALUE;
self->last_position = POSITION_INITIAL;
if ( !self->audio_format )
}
static int seek_video( producer_avformat self, mlt_position position,
- int64_t req_position, int must_decode, int use_new_seek, int *ignore )
+ int64_t req_position, int preseek, int use_pts )
{
mlt_producer producer = self->parent;
int paused = 0;
double source_fps = mlt_properties_get_double( properties, "meta.media.frame_rate_num" ) /
mlt_properties_get_double( properties, "meta.media.frame_rate_den" );
- if ( self->av_frame && position + 1 == self->video_expected )
- {
- // We're paused - use last image
- paused = 1;
- }
- else if ( !self->seekable && position > self->video_expected && ( position - self->video_expected ) < 250 )
+ // find initial PTS
+ if ( self->last_position == POSITION_INITIAL )
{
- // Fast forward - seeking is inefficient for small distances - just ignore following frames
- *ignore = ( int )( ( position - self->video_expected ) / mlt_producer_get_fps( producer ) * source_fps );
- codec_context->skip_loop_filter = AVDISCARD_NONREF;
- }
- else if ( self->seekable && ( position < self->video_expected || position - self->video_expected >= 12 || self->last_position < 0 ) )
- {
- if ( use_new_seek && self->last_position == POSITION_INITIAL )
- {
- // find first key frame
- int ret = 0;
- int toscan = 100;
- AVPacket pkt;
+ int ret = 0;
+ int toscan = 100;
+ AVPacket pkt;
- while ( ret >= 0 && toscan-- > 0 )
+ while ( ret >= 0 && toscan-- > 0 )
+ {
+ ret = av_read_frame( context, &pkt );
+ if ( ret >= 0 && pkt.stream_index == self->video_index )
{
- ret = av_read_frame( context, &pkt );
- if ( ret >= 0 && ( pkt.flags & PKT_FLAG_KEY ) && pkt.stream_index == self->video_index )
- {
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "first_pts %"PRId64" dts %"PRId64" pts_dts_delta %d\n", pkt.pts, pkt.dts, (int)(pkt.pts - pkt.dts) );
- self->first_pts = pkt.pts;
+ mlt_log_debug( MLT_PRODUCER_SERVICE(producer),
+ "first_pts %"PRId64" dts %"PRId64" pts_dts_delta %d\n",
+ pkt.pts, pkt.dts, (int)(pkt.pts - pkt.dts) );
+ self->first_pts = use_pts? pkt.pts : pkt.dts;
+ if ( self->first_pts != AV_NOPTS_VALUE )
toscan = 0;
- }
- av_free_packet( &pkt );
}
- // Rewind
- av_seek_frame( context, -1, 0, AVSEEK_FLAG_BACKWARD );
+ av_free_packet( &pkt );
}
+ av_seek_frame( context, -1, 0, AVSEEK_FLAG_BACKWARD );
+ }
+ if ( self->av_frame && position + 1 == self->video_expected )
+ {
+ // We're paused - use last image
+ paused = 1;
+ }
+ else if ( self->seekable && ( position < self->video_expected || position - self->video_expected >= 12 || self->last_position < 0 ) )
+ {
// Calculate the timestamp for the requested frame
- int64_t timestamp;
- if ( use_new_seek )
- {
- timestamp = ( req_position - 0.1 / source_fps ) /
- ( av_q2d( stream->time_base ) * source_fps );
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "pos %"PRId64" pts %"PRId64"\n", req_position, timestamp );
- if ( self->first_pts > 0 )
- timestamp += self->first_pts;
- else if ( context->start_time != AV_NOPTS_VALUE )
- timestamp += context->start_time;
- if ( req_position <= 0 )
- timestamp = 0;
- }
- else
- {
- timestamp = ( int64_t )( ( double )req_position / source_fps * AV_TIME_BASE + 0.5 );
- if ( context->start_time != AV_NOPTS_VALUE )
- timestamp += context->start_time;
- }
- if ( must_decode )
- timestamp -= AV_TIME_BASE;
+ int64_t timestamp = req_position / ( av_q2d( stream->time_base ) * source_fps );
+ if ( req_position <= 0 )
+ timestamp = 0;
+ else if ( self->first_pts != AV_NOPTS_VALUE )
+ timestamp += self->first_pts;
+ else if ( context->start_time != AV_NOPTS_VALUE )
+ timestamp += context->start_time;
+ if ( preseek && av_q2d( stream->time_base ) != 0 )
+ timestamp -= 2 / av_q2d( stream->time_base );
if ( timestamp < 0 )
timestamp = 0;
mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "seeking timestamp %"PRId64" position %d expected %d last_pos %"PRId64"\n",
timestamp, position, self->video_expected, self->last_position );
// Seek to the timestamp
- if ( use_new_seek )
+ // NOTE: reopen_video is disabled at this time because it is causing trouble with A/V sync.
+ if ( 1 || use_pts || req_position > 0 || self->last_position <= 0 )
{
codec_context->skip_loop_filter = AVDISCARD_NONREF;
av_seek_frame( context, self->video_index, timestamp, AVSEEK_FLAG_BACKWARD );
- }
- else if ( req_position > 0 || self->last_position <= 0 )
- {
- av_seek_frame( context, -1, timestamp, AVSEEK_FLAG_BACKWARD );
+
+ // flush any pictures still in decode buffer
+ avcodec_flush_buffers( codec_context );
}
else
{
self->current_position = POSITION_INVALID;
self->last_position = POSITION_INVALID;
av_freep( &self->av_frame );
-
- if ( use_new_seek )
- {
- // flush any pictures still in decode buffer
- avcodec_flush_buffers( codec_context );
- }
}
}
return paused;
// Packet
AVPacket pkt;
- // Special case ffwd handling
- int ignore = 0;
-
// We may want to use the source fps if available
double source_fps = mlt_properties_get_double( properties, "meta.media.frame_rate_num" ) /
mlt_properties_get_double( properties, "meta.media.frame_rate_den" );
strcmp( codec_context->codec->name, "rawvideo" ) );
// Turn on usage of new seek API and PTS for seeking
- int use_new_seek = self->seekable &&
- codec_context->codec_id == CODEC_ID_H264 && !strcmp( context->iformat->name, "mpegts" );
- if ( mlt_properties_get( properties, "new_seek" ) )
- use_new_seek = mlt_properties_get_int( properties, "new_seek" );
+ int use_pts = self->seekable &&
+ codec_context->codec_id == CODEC_ID_H264 && codec_context->has_b_frames;
+ if ( mlt_properties_get( properties, "use_pts" ) )
+ use_pts = mlt_properties_get_int( properties, "use_pts" );
double delay = mlt_properties_get_double( properties, "video_delay" );
// Seek if necessary
const char *interp = mlt_properties_get( frame_properties, "rescale.interp" );
int preseek = must_decode
#if defined(FFUDIV) && LIBAVFORMAT_VERSION_INT >= ((53<<16)+(24<<8)+2)
- && ( !use_new_seek || ( interp && strcmp( interp, "nearest" ) ) )
+ && ( !use_pts || ( interp && strcmp( interp, "nearest" ) ) )
#endif
&& codec_context->has_b_frames;
- int paused = seek_video( self, position, req_position, preseek, use_new_seek, &ignore );
+ int paused = seek_video( self, position, req_position, preseek, use_pts );
// Seek might have reopened the file
context = self->video_format;
if ( ret >= 0 && pkt.stream_index == self->video_index && pkt.size > 0 )
{
// Determine time code of the packet
- if ( use_new_seek )
+ if ( pkt.pts == AV_NOPTS_VALUE )
{
- int64_t pts = pkt.pts;
- if ( self->first_pts > 0 )
- pts -= self->first_pts;
- else if ( context->start_time != AV_NOPTS_VALUE )
- pts -= context->start_time;
- int_position = ( int64_t )( ( av_q2d( stream->time_base ) * pts + delay ) * source_fps + 0.1 );
- if ( pkt.pts == AV_NOPTS_VALUE )
+ self->invalid_pts_counter++;
+ if ( self->invalid_pts_counter > 20 )
{
- self->invalid_pts_counter++;
- if ( self->invalid_pts_counter > 20 )
- {
- mlt_log_panic( MLT_PRODUCER_SERVICE(producer), "\ainvalid PTS; DISABLING NEW_SEEK!\n" );
- mlt_properties_set_int( properties, "new_seek", 0 );
- int_position = req_position;
- use_new_seek = 0;
- }
+ mlt_log_warning( MLT_PRODUCER_SERVICE(producer), "PTS invalid; using DTS instead\n" );
+ mlt_properties_set_int( properties, "use_pts", 0 );
+ use_pts = 0;
}
- else
- {
- self->invalid_pts_counter = 0;
- }
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "pkt.pts %"PRId64" req_pos %"PRId64" cur_pos %"PRId64" pkt_pos %"PRId64"\n",
- pkt.pts, req_position, self->current_position, int_position );
}
else
{
- if ( pkt.dts != AV_NOPTS_VALUE )
- {
- int_position = ( int64_t )( ( av_q2d( stream->time_base ) * pkt.dts + delay ) * source_fps + 0.5 );
- if ( context->start_time != AV_NOPTS_VALUE )
- int_position -= ( int64_t )( context->start_time * source_fps / AV_TIME_BASE + 0.5 );
- if ( int_position == self->last_position )
- int_position = self->last_position + 1;
- }
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "pkt.dts %"PRId64" req_pos %"PRId64" cur_pos %"PRId64" pkt_pos %"PRId64"\n",
- pkt.dts, req_position, self->current_position, int_position );
- // Make a dumb assumption on streams that contain wild timestamps
- if ( abs( req_position - int_position ) > 999 )
- {
- int_position = req_position;
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), " WILD TIMESTAMP!" );
- }
+ self->invalid_pts_counter = 0;
+ }
+ int64_t pts = ( use_pts && pkt.pts != AV_NOPTS_VALUE )? pkt.pts : pkt.dts;
+ if ( pts != AV_NOPTS_VALUE )
+ {
+ if ( !self->seekable && self->first_pts == AV_NOPTS_VALUE )
+ self->first_pts = pts;
+ if ( self->first_pts != AV_NOPTS_VALUE )
+ pts -= self->first_pts;
+ else if ( context->start_time != AV_NOPTS_VALUE )
+ pts -= context->start_time;
+ int_position = ( int64_t )( ( av_q2d( stream->time_base ) * pts + delay ) * source_fps + 0.5 );
+ if ( int_position == self->last_position )
+ int_position = self->last_position + 1;
+ }
+ mlt_log_debug( MLT_PRODUCER_SERVICE(producer),
+ "V pkt.pts %"PRId64" pkt.dts %"PRId64" req_pos %"PRId64" cur_pos %"PRId64" pkt_pos %"PRId64"\n",
+ pkt.pts, pkt.dts, req_position, self->current_position, int_position );
+
+ // Make a dumb assumption on streams that contain wild timestamps
+ if ( abs( req_position - int_position ) > 999 )
+ {
+ int_position = req_position;
+ mlt_log_warning( MLT_PRODUCER_SERVICE(producer), " WILD TIMESTAMP!\n" );
}
self->last_position = int_position;
self->vdpau->is_decoded = 0;
}
#endif
- codec_context->reordered_opaque = pkt.pts;
+ codec_context->reordered_opaque = int_position;
if ( int_position >= req_position )
codec_context->skip_loop_filter = AVDISCARD_NONE;
#if (LIBAVCODEC_VERSION_INT >= ((52<<16)+(26<<8)+0))
if ( got_picture )
{
- if ( use_new_seek )
- {
- // Determine time code of the packet
- int64_t pts = self->av_frame->reordered_opaque;
- if ( self->first_pts > 0 )
- pts -= self->first_pts;
- else if ( context->start_time != AV_NOPTS_VALUE )
- pts -= context->start_time;
- int_position = ( int64_t )( ( av_q2d( stream->time_base) * pts + delay ) * source_fps + 0.1 );
- mlt_log_debug( MLT_PRODUCER_SERVICE(producer), "got frame %"PRId64", key %d\n", int_position, self->av_frame->key_frame );
- }
- // Handle ignore
+ if ( use_pts )
+ // Get position of reordered frame
+ int_position = self->av_frame->reordered_opaque;
+
if ( int_position < req_position )
- {
- ignore = 0;
got_picture = 0;
- }
else if ( int_position >= req_position )
- {
- ignore = 0;
codec_context->skip_loop_filter = AVDISCARD_NONE;
- }
- else if ( ignore -- )
- {
- got_picture = 0;
- }
}
mlt_log_debug( MLT_PRODUCER_SERVICE(producer), " got_pic %d key %d\n", got_picture, pkt.flags & PKT_FLAG_KEY );
}
}
}
-static int seek_audio( producer_avformat self, mlt_position position, double timecode, int *ignore )
+static int seek_audio( producer_avformat self, mlt_position position, double timecode )
{
int paused = 0;
// We're paused - silence required
paused = 1;
}
- else if ( !self->seekable && position > self->audio_expected && ( position - self->audio_expected ) < 250 )
- {
- // Fast forward - seeking is inefficient for small distances - just ignore following frames
- *ignore = position - self->audio_expected;
- }
else if ( position < self->audio_expected || position - self->audio_expected >= 12 )
{
AVFormatContext *context = self->audio_format;
// Skip this on non-seekable, audio-only inputs.
if ( pkt.pts >= 0 && ( self->seekable || self->video_format ) && *ignore == 0 && audio_used > samples / 2 )
{
- double current_pts = av_q2d( context->streams[ index ]->time_base ) * pkt.pts;
+ int64_t pts = pkt.pts;
+ if ( self->first_pts != 0 )
+ pts -= self->first_pts;
+ else if ( context->start_time != AV_NOPTS_VALUE )
+ pts -= context->start_time;
+ double timebase = av_q2d( context->streams[ index ]->time_base );
+ int64_t int_position = ( int64_t )( timebase * pts * fps + 0.5 );
int64_t req_position = ( int64_t )( timecode * fps + 0.5 );
- int64_t int_position = ( int64_t )( current_pts * fps + 0.5 );
- if ( context->start_time != AV_NOPTS_VALUE )
- int_position -= ( int64_t )( fps * context->start_time / AV_TIME_BASE + 0.5 );
+
+ mlt_log_debug( MLT_PRODUCER_SERVICE(self->parent),
+ "A pkt.pts %"PRId64" pkt.dts %"PRId64" req_pos %"PRId64" cur_pos %"PRId64" pkt_pos %"PRId64"\n",
+ pkt.pts, pkt.dts, req_position, self->current_position, int_position );
if ( int_position > 0 )
{
*ignore = req_position - int_position;
else if ( self->audio_index != INT_MAX && int_position > req_position + 2 )
// We are ahead, so seek backwards some more
- seek_audio( self, req_position, timecode - 1.0, ignore );
+ seek_audio( self, req_position, timecode - 1.0 );
}
}
int ignore[ MAX_AUDIO_STREAMS ] = { 0 };
// Flag for paused (silence)
- int paused = seek_audio( self, position, real_timecode, &ignore[0] );
+ int paused = seek_audio( self, position, real_timecode );
// Initialize ignore for all streams from the seek return value
int i = MAX_AUDIO_STREAMS;