mantarimay/Aegisub
1
0
Fork 0
forked from mia/Aegisub
Code Pull requests Activity

attempt at fixing the lavc audio skew problems by making very very sure all data in each packet is decoded.

Browse source 
Originally committed to SVN as r2094.
This commit is contained in:
Karl Blomster 2008-03-21 20:30:34 +00:00
parent 6d8f862aed
commit ee04563ff6
1 changed files with 35 additions and 26 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

61
aegisub/audio_provider_lavc.cpp
View file

@ -169,37 +169,46 @@ void LAVCAudioProvider::GetAudio(void *buf, int64_t start, int64_t count)
AVPacket packet;
while (samples_to_decode > 0 && av_read_frame(lavcfile->fctx, &packet) >= 0) {
while (packet.stream_index == audStream) {
int temp_output_buffer_size = AVCODEC_MAX_AUDIO_FRAME_SIZE; /* see constructor, it malloc()'s buffer to this */
int decoded_samples;
/* we're not dealing with video packets in this here provider */
if (packet.stream_index == audStream) {
int size = packet.size;
uint8_t *data = packet.data;
if (avcodec_decode_audio2(codecContext, buffer, &temp_output_buffer_size, packet.data, packet.size) <= 0)
throw _T("Failed to decode audio");
if (temp_output_buffer_size == 0) /* gets changed to number of bytes actually output, so this is sanity checking */
break;
while (size > 0) {
int temp_output_buffer_size = AVCODEC_MAX_AUDIO_FRAME_SIZE; /* see constructor, it malloc()'s buffer to this */
int retval, decoded_samples;
retval = avcodec_decode_audio2(codecContext, buffer, &temp_output_buffer_size, data, size)
if (retval <= 0)
throw _T("Failed to decode audio");
if (temp_output_buffer_size == 0) /* sanity checking, shouldn't ever happen */
break;
decoded_samples = temp_output_buffer_size / 2;
/* do we need to resample? */
if (rsct) {
if ((int64_t)(decoded_samples * resample_ratio / codecContext->channels) > samples_to_decode)
decoded_samples = (int64_t)(samples_to_decode / resample_ratio * codecContext->channels);
decoded_samples = audio_resample(rsct, _buf, buffer, decoded_samples / codecContext->channels);
decoded_samples = temp_output_buffer_size / 2;
size -= retval;
data += retval;
/* make sure we somehow didn't end up with more samples than we wanted */
assert(decoded_samples <= samples_to_decode);
} else {
/* no resampling needed, just copy to the buffer */
/* if (decoded_samples > samples_to_decode)
decoded_samples = samples_to_decode; */
/* I do not understand the point of the above, changed to a more reasonable assertation instead -Fluff */
assert(decoded_samples <= samples_to_decode);
/* do we need to resample? */
if (rsct) {
if ((int64_t)(decoded_samples * resample_ratio / codecContext->channels) > samples_to_decode)
decoded_samples = (int64_t)(samples_to_decode / resample_ratio * codecContext->channels);
decoded_samples = audio_resample(rsct, _buf, buffer, decoded_samples / codecContext->channels);
memcpy(_buf, buffer, temp_output_buffer_size);
/* make sure we somehow didn't end up with more samples than we wanted */
assert(decoded_samples <= samples_to_decode);
} else {
/* no resampling needed, just copy to the buffer */
/* if (decoded_samples > samples_to_decode)
decoded_samples = samples_to_decode; */
/* I do not understand the point of the above, changed to a more reasonable assertation instead -Fluff */
assert(decoded_samples <= samples_to_decode);
memcpy(_buf, buffer, temp_output_buffer_size);
}
_buf += decoded_samples;
samples_to_decode -= decoded_samples;
}
_buf += decoded_samples;
samples_to_decode -= decoded_samples;
/* break; */ /* why did this loop need to be broken manually? */
}
av_free_packet(&packet);