Robin Bowes;179738 Wrote: 
> 
> > 
> > OK, how would you differentiate between a file whose audio has been
> > modified (e.g. re-ripped) and one whose tags have changed?
> > 
> > In both cases, the source file timestamp will have changed.
> 
I see your point. For me, it is very rare I that I re-rip a file, but I
change tags (mostly genre) reasonably fequently. I guess maybe some
people re-rip more frequently than they ammend tags.

Robin Bowes;179738 Wrote: 
> Here's an idea: Use the MD5CHECKSUM from the flac file to determine
> whether or not the audio has changed.
> 
> So, the first time a file is converted, we read the MD5CHECKSUM from
> the
> flac file and write it as a tag to the mp3 file.
> 
> Next time, we check the mp3 file for a tag containing the MD5CHECKSUM.
> 
> If it's present and matches, we don't transcode the file.
> If it's present and doesn't match, we must transcode.
> If it's not present then, to be correct, we *should* transcode.
> However,
> this will cause problems when upgrading from previous versions of
> flac2mp3 as no MD5CHECKSUM tag will be present in the mp3 files. I'm
> currently thinking how best to deal with this situation.
> 
> R.

I wasn't aware of MD5CHECKSUM, but assuming this is a checksum of
purely the audio data, that sounds like an excellent (and pretty much
bullet-proof) idea. 

WRT upgrading, how about the user decide not to transcode via a
command-line options? Maybe something like *--CopyMd5IfMissing*. This
would obviously have to be made clear in the upgrade notes.

Thanks very much for the fast response & your ideas.

Richard


-- 
SadGamerGeek
------------------------------------------------------------------------
SadGamerGeek's Profile: http://forums.slimdevices.com/member.php?userid=6145
View this thread: http://forums.slimdevices.com/showthread.php?t=32641

_______________________________________________
ripping mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/ripping

Reply via email to