Hi there,

this is a feature request. If there is already a way to do this I would
appreciate if someone could tell me how. I have access to a huge mp3 collection which is shared among several users over a network via NFS. There are approximately 100.000 files. The creation of the database takes amarok about 15 hours. I suppose this is mainly due to the relatively low speed of NFS, not due to the CPU or the hard disk. Creating a database of 4000 files stored locally takes about 2 minutes.

It would be fantastic to actually build the database without using the GUI, i.e. building the database directly on the server (mysql).

Is there a way to do this?


regards
dtmr

Yes, mysql is available as a database backend for amaroK, your distro should have a version of amaroK with this enabled. If you are building from source, as long as mysql is installed on the build machine, amaroK should build with mysql support.

The trick once you get the right amaroK installed, is making sure the music collection is mounted with the exact same path on all the systems that will use the database, i.e. you must mount the nfs share in the same place on every computer. For example, /mnt/music.

Also, if you want to use the watch folders feature, make sure only one computer is monitoring the nfs share for changes, so all the computers don\'t try to update the database when a song is added. Pick only one computer to manage the database and turn off watch folders on all the others.

Hope this helps.

Hi,

thanks for your help! But this is not quite what I\'m looking for.
The database should be created by the server containing the mp3 files. The server does not have a GUI. There should be a way to create the database *without* using the graphical interface of amarok, i.e. a console program which connects to the SQL server to submit all the id3 tags and other information about the files.


thanks in advance,

dtmr

Yes, but there isn\'t.

Unfortunately, the only way is the method I specified.