Chiebot-Mirror
/
FFmpeg
mirror of https://github.com/FFmpeg/FFmpeg.git
8
0
Fork 0
Code Issues Projects Releases Wiki Activity

A/V sync patch from Gildas Bazin.

Browse Source 
Originally committed as revision 2098 to svn://svn.ffmpeg.org/ffmpeg/trunk
pull/126/head
Roman Shaposhnik 22 years ago
parent 302898fcbe
commit e7d0374f27
1 changed files with 49 additions and 60 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 109
      ffmpeg.c

109
ffmpeg.c
Unescape View File

@ -205,7 +205,8 @@ typedef struct AVInputStream {
int64_t start; /* time when read started */
unsigned long frame; /* current frame */
AVFrac pts; /* synthetic pts for cases where pkt.pts == 0 */
AVFrac next_pts; /* synthetic pts for cases where pkt.pts == 0 */
int64_t pts; /* current pts */
} AVInputStream;
typedef struct AVInputFile {
@ -472,42 +473,33 @@ static void do_video_out(AVFormatContext *s,
but not for the general case */
if (audio_sync) {
/* compute the A-V delay and duplicate/remove frames if needed */
double adelta, vdelta, apts, vpts, av_delay;
if (audio_sync->sync_ipts != AV_NOPTS_VALUE &&
ost->sync_ipts != AV_NOPTS_VALUE) {
adelta = (double)(ost->st->pts.val - audio_sync->sync_opts) *
s->pts_num / s->pts_den;
apts = audio_sync->sync_ipts + adelta;
vdelta = (double)(ost->st->pts.val - ost->sync_opts) *
s->pts_num / s->pts_den;
vpts = ost->sync_ipts + vdelta;
av_delay = apts - vpts;
// printf("delay=%f\n", av_delay);
if (av_delay < -AV_DELAY_MAX)
nb_frames = 2;
else if (av_delay > AV_DELAY_MAX)
nb_frames = 0;
}
double adelta, vdelta, av_delay;
adelta = audio_sync->sync_ipts - ((double)audio_sync->sync_opts *
s->pts_num / s->pts_den);
vdelta = ost->sync_ipts - ((double)ost->sync_opts *
s->pts_num / s->pts_den);
av_delay = adelta - vdelta;
// printf("delay=%f\n", av_delay);
if (av_delay < -AV_DELAY_MAX)
nb_frames = 2;
else if (av_delay > AV_DELAY_MAX)
nb_frames = 0;
} else {
double vdelta;
if (ost->sync_ipts != AV_NOPTS_VALUE) {
vdelta = (double)(ost->st->pts.val) * s->pts_num / s->pts_den - (ost->sync_ipts - ost->sync_ipts_offset);
if (vdelta < 100 && vdelta > -100 && ost->sync_ipts_offset) {
if (vdelta < -AV_DELAY_MAX)
nb_frames = 2;
else if (vdelta > AV_DELAY_MAX)
nb_frames = 0;
} else {
ost->sync_ipts_offset -= vdelta;
if (!ost->sync_ipts_offset)
ost->sync_ipts_offset = 0.000001; /* one microsecond */
}
vdelta = (double)(ost->st->pts.val) * s->pts_num / s->pts_den - (ost->sync_ipts - ost->sync_ipts_offset);
if (vdelta < 100 && vdelta > -100 && ost->sync_ipts_offset) {
if (vdelta < -AV_DELAY_MAX)
nb_frames = 2;
else if (vdelta > AV_DELAY_MAX)
nb_frames = 0;
} else {
ost->sync_ipts_offset -= vdelta;
if (!ost->sync_ipts_offset)
ost->sync_ipts_offset = 0.000001; /* one microsecond */
}
}
@ -1108,13 +1100,14 @@ static int av_encode(AVFormatContext **output_files,
for(i=0;i<nb_istreams;i++) {
ist = ist_table[i];
is = input_files[ist->file_index];
ist->pts = 0;
switch (ist->st->codec.codec_type) {
case CODEC_TYPE_AUDIO:
av_frac_init(&ist->pts,
av_frac_init(&ist->next_pts,
0, 0, is->pts_num * ist->st->codec.sample_rate);
break;
case CODEC_TYPE_VIDEO:
av_frac_init(&ist->pts,
av_frac_init(&ist->next_pts,
0, 0, is->pts_num * ist->st->codec.frame_rate);
break;
default:
@ -1220,10 +1213,6 @@ static int av_encode(AVFormatContext **output_files,
len = pkt.size;
ptr = pkt.data;
while (len > 0) {
int64_t ipts;
ipts = AV_NOPTS_VALUE;
/* decode the packet if needed */
data_buf = NULL; /* fail safe */
data_size = 0;
@ -1233,8 +1222,15 @@ static int av_encode(AVFormatContext **output_files,
/* NOTE2: even if the fraction is not initialized,
av_frac_set can be used to set the integer part */
if (ist->frame_decoded) {
/* If pts is unavailable -- we have to use synthetic one */
ipts = (pkt.pts == AV_NOPTS_VALUE) ? ist->pts.val : pkt.pts;
/* If pts is unavailable -- we have to use synthetic one */
if( pkt.pts != AV_NOPTS_VALUE )
{
ist->pts = ist->next_pts.val = pkt.pts;
}
else
{
ist->pts = ist->next_pts.val;
}
ist->frame_decoded = 0;
}
@ -1255,7 +1251,7 @@ static int av_encode(AVFormatContext **output_files,
continue;
}
data_buf = (uint8_t *)samples;
av_frac_add(&ist->pts,
av_frac_add(&ist->next_pts,
is->pts_den * data_size / (2 * ist->st->codec.channels));
break;
case CODEC_TYPE_VIDEO:
@ -1280,7 +1276,7 @@ static int av_encode(AVFormatContext **output_files,
len -= ret;
continue;
}
av_frac_add(&ist->pts,
av_frac_add(&ist->next_pts,
is->pts_den * ist->st->codec.frame_rate_base);
}
break;
@ -1334,23 +1330,17 @@ static int av_encode(AVFormatContext **output_files,
if (ost->source_index == ist_index) {
os = output_files[ost->file_index];
if (ipts != AV_NOPTS_VALUE) {
#if 0
printf("%d: got pts=%f %f\n",
i, pkt.pts / 90000.0,
(ipts - ost->st->pts.val) / 90000.0);
printf("%d: got pts=%f %f\n", i, pkt.pts / 90000.0,
(ist->pts - ost->st->pts.val) / 90000.0);
#endif
/* set the input output pts pairs */
ost->sync_ipts = (double)ipts * is->pts_num /
is->pts_den;
/* XXX: take into account the various fifos,
in particular for audio */
ost->sync_opts = ost->st->pts.val;
//printf("ipts=%lld sync_ipts=%f sync_opts=%lld pts.val=%lld pkt.pts=%lld\n", ipts, ost->sync_ipts, ost->sync_opts, ost->st->pts.val, pkt.pts);
} else {
//printf("pts.val=%lld\n", ost->st->pts.val);
ost->sync_ipts = AV_NOPTS_VALUE;
}
/* set the input output pts pairs */
ost->sync_ipts = (double)ist->pts * is->pts_num /
is->pts_den;
/* XXX: take into account the various fifos,
in particular for audio */
ost->sync_opts = ost->st->pts.val;
//printf("ipts=%lld sync_ipts=%f sync_opts=%lld pts.val=%lld pkt.pts=%lld\n", ist->pts, ost->sync_ipts, ost->sync_opts, ost->st->pts.val, pkt.pts);
if (ost->encoding_needed) {
switch(ost->st->codec.codec_type) {
@ -1397,7 +1387,6 @@ static int av_encode(AVFormatContext **output_files,
}
}
av_free(buffer_to_free);
ipts = AV_NOPTS_VALUE;
}
discard_packet:
av_free_packet(&pkt);

Write Preview
Loading…
Cancel
Save