Xing and Fraunhofer VBRI

[Jim]

Hi James.

I managed to take a look at the file you made available.  Unfortunately I wasn't 
able to correct the zero length/bitrate problem.  I was able to get TagLib to 
read the file properly.  (It was originally falsely detecting the beginning of a 
frame but that was easy to fix.)

However, that file has a bad Xing header.  Or rather, the Xing header is valid 
but the values that it contains seem impossible.  The reason you are getting a 
bad file length is because the Xing header says there is only one frame in the 
entire file.  So 1 frame at 44100 Hz sample rate would mean less than one second 
in length.  I am not sure if this is the cause for the other files you have.

Jim


On 12/10/2010 9:58 AM, James O. wrote:
> Jim,
>
> Here is an MP3 that causes the issues described (zero length). It seems odd to
> me that other tracks in the same album (I would assume the same encoding run) do
> not exhibit similar issues.
>
> http://codefromjames.com/files/linked/3weeks.mp3
>
> This band is defunct and the albums was released as a free MP3 download. Should
> have no copyright issues.
>
> Thanks for looking into this. If I'm not mistaken a number of Linux media
> players use TagLib at their core. I wonder if they have seen similar problems?
>
> James
>
>
>
> _______________________________________________
> taglib-devel mailing list
> taglib-devel at kde.org
> https://mail.kde.org/mailman/listinfo/taglib-devel