dbera.web at gmail.com
Wed Oct 11 16:02:44 CEST 2006
> > I was having some problem with the duration of a vbr mpeg file and
> > then i ran into the problem of a faulty Xing header. That lead me to
> > the source of taglib and lame (mpglib) to figure out how they compute
> > framesize. Now, I dont have much expertise in these things so I
> > compared the framesize calculation in lame and taglib. And I found
> > some differences!!!
> > Maybe someone here can tell me which is the correct logic ? (I couldnt
> > find any 'free' reliable source of MPEG standard :P). I am writing a
> > simple tool to fix faulty xing headers in some of my vbr mp3 files and
> > it will be helpful to know the correct way to compute framelength.
> There was also some other inconsistency in how the duration was
> calculated. I am attaching a patch fixing all these. Dunno if the
> calculation logic is correct; no reliable header information at hand.
> I based my calculations on lame(mpglib) source code and
Since no one replied till now, I am afraid this issue will be
forgotten. I am using the patched taglib (patched as I mentioned in
earlier emails) and tested them against all the files which were
giving incorrect duration with the original taglib. All of them seem
to work fine.
To me this looks definitely like a bug. Lest this issues gets lost
in the mailing list, I will file a bug and attach the patch there. I
hope thats ok.
Thanks for your time,
Debajyoti Bera @ http://dtecht.blogspot.com
beagle / KDE fan
Mandriva / Inspiron-1100 user
More information about the taglib-devel