On Donnerstag 29 Januar 2009, Dirk Meyer wrote: > No, it is not caused by seek and it is no infinite loop. dump_IFD tried > to parse 18761 entries in the EXIF header maker note and the first entry > itself calls s2n around 50.500 times.
Uuuuh.
> I guess it will finish in some
> hours. Anyway, I added a small check to prevent kaa.metadata from trying
> to parse too many entries. But EXIF.py has again proven to very
> unreliable. Any idea how to replace it?
I have once written (boost::)python bindings for libexif. They've been
released in their contrib-folder in the past, but seem to have been removed
again when they got outdated. There's a much more up-to-date version on my
hard disk.
OTOH, there's libkexif from KDE, which is much more lightweight AFAIK, mostly
oriented towards basic tasks (e.g. orientation and some hand-picked other
attributes).
> Can you send me a mp3 file with the problem in a private mail please?
Can you please send it to me, too? I also wrote an MP3 analyzing program and
I have already a large collection of (potentially) interesting files. (I'll
then quickly report back about the internal structure of the file, including
headers, tags, junk etc.)
BTW: I have repeatedly thought that I'd like to replace EyeD3.py with
[py]TagLib. TagLib is very stable, maintained, and blazingly fast. Since my
last freevo update, I do not have any cache since "freevo cache" always
crashes, but not before it munged my MP3 collection for sth. like 6 hours(!).
(Which makes debugging reaaally painful.)
--
Ciao, / / .o.
/--/ ..o
/ / ANS ooo
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------------ This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________ Freevo-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freevo-devel
