You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Using vorbis 1.3.6, when trying to decode an example Ogg-Vorbis file from Wikipedia, it always starts reading at an offset (of 11968 samples or 0.27138321995464854s respectively).
Setting this offset to 0 using ov_pcm_seek(), it still reads from said offset, even though ov_pcm_tell() still returns 0.
When trying to read samples after seeking to 0, the tell position suddenly jumps:
Although you might not see it at first glance, but the tell function was lying to you all along.
We're reading 10 x 1024 samples, or 10240 total samples.
If you add those to the initial offset of 11968, you get (surprise) 22208.
So seek didn't do it's job and neither did tell.
The data is also read from the offset and not the beginning of the file.
Whatever I tried, I could not get vorbisfile to decode the first part of the file for some reason.
It's not broken either - I could open it with Audacity, VLC, Groove, etc.
I've also tried both ov_read and ov_read_float.
The text was updated successfully, but these errors were encountered:
Using vorbis 1.3.6, when trying to decode an example Ogg-Vorbis file from Wikipedia, it always starts reading at an offset (of
11968
samples or0.27138321995464854s
respectively).Setting this offset to
0
usingov_pcm_seek()
, it still reads from said offset, even thoughov_pcm_tell()
still returns0
.When trying to read samples after seeking to
0
, the tell position suddenly jumps:Although you might not see it at first glance, but the tell function was lying to you all along.
We're reading 10 x 1024 samples, or 10240 total samples.
If you add those to the initial offset of 11968, you get (surprise) 22208.
So seek didn't do it's job and neither did tell.
The data is also read from the offset and not the beginning of the file.
Whatever I tried, I could not get vorbisfile to decode the first part of the file for some reason.
It's not broken either - I could open it with Audacity, VLC, Groove, etc.
I've also tried both
ov_read
andov_read_float
.The text was updated successfully, but these errors were encountered: