BUG workaround - How to add UTF8 ID3 tagged MP3 to Linuxmce

[Viking]

Hi,

At the moment there is a bug in Linuxmce causing it to destroy any spcial ID3 tags. What I mean is for example danish or german characters. 
See here : http://svn.linuxmce.org/trac.cgi/ticket/508

The workaround it to double encode theese tags. And you have to do this BEFORE you put the music somewhere where LinuxMCE will detect it !!

I have written 2 scripts for that. One saving the current state (SaveID3taginfo.sh) and a second (DoubleUTF8encodeMP3.sh) for double encoding the tags. The first is just saving the old infos (only for the files where it is needed) in a *.taginfo file so that we are able to restore the tags some day if needed You must run both scripts as the second is relying on that the frst was run!
There is no script yet that can undo the changes - maybe it will never be needed.


I have tested this on all my audio, bu can't garantee that it will work flawlessly, so please do a backup of your audio before using this!

When you are finished you should also use convmv to rename your directories and filenames to UTF-8. That way it looks nicer in samba and does not cause any problems for the UpdateMedia daemon.

Note that ALL files below the directory where you start the script is changed!!


Here is what to do.
The convmv are called twice because I had the impression that not all directories was renamed the first time if you remove the --notest then it shows you what it would do, but does noting.

Code: [Select]

cd <where your mp3 files are>
SaveID3taginfo.sh 2>&1 | tee SaveID3taginfo.log
DoubleUTF8encodeMP3.sh 2>&1 | tee DoubleUTF8encodeMP3.log

find . -type f -exec convmv --notest -f iso-8859-1 -t utf8 {} \;
find . -type f -exec convmv --notest -f iso-8859-1 -t utf8 {} \;

find . -type d -exec convmv --notest -f iso-8859-1 -t utf8 {} \;
find . -type d -exec convmv --notest -f iso-8859-1 -t utf8 {} \;

Greetings
Viking