itdb - iTunes Database and utilities

itdb will load all the data from iTunes into a MySQL database. Once you have the data in your database you can do some fun stuff with it.

playlistlinks will make symbolic links and m3u files for your playlists. itdb2html is a super old crusty, crufty web-ui for your music

Packages

Here's what I had to do on my mac to get things working.

brew install mysql python3
pip3 install appscript argcomplete cheetah3 humanize mysqlclient python-dateutil requests tdqm

If python3 -c 'import MySQLdb' fails you might need to this bullshit:

brew install openssl@1.1
cd /usr/local/opt/mysql/lib/
ln -s /usr/local/opt/openssl/lib/libssl.1.1.dylib
ln -s /usr/local/opt/openssl/lib/libcrypto.1.1.dylib

Database Setup

Copy the config file and install mysql (I like brew).

You might need to add secure_file_priv = '' to /usr/loca/etc/my.cnf

mysql.server start

cp itdb.config.example ~/.itdb.config
# maybe you want to edit ~/.itdb.config now?
mysqladmin -u root create itdb
mysql -u root

Then run these commands (you might want to change the password (and in ~/.itdb.config too):

CREATE USER 'itdb'@'localhost' IDENTIFIED BY 'itdb';
GRANT ALL PRIVILEGES ON itdb.* TO 'itdb'@'localhost' WITH GRANT OPTION;
GRANT FILE on *.* to itdb@'localhost';
FLUSH PRIVILEGES;

Once the database is made you can then create the database using itdb.sql (since .itdb.config is in inifile format you can use --defaults-file) and then load your data:

mysql --defaults-file=~/.itdb.config < itdb.sql
# did that work?
mysql --defaults-file=~/.itdb.config <<< 'SHOW TABLES'
# load it
./itdbloader.py
# verify it works
mysql --defaults-file=~/.itdb.config -E <<< 'SELECT COUNT(*) AS num_tracks FROM tracks; SELECT COUNT(*) AS num_playlisys FROM playlists;' | fgrep -v '*****'

itdbloader.py

I found iTunes.py over at lazycat.org (no longer available there) which I've modified to also load playlist as well as track information. It uses the most excellent pulldom python API which attempts to be more memory efficient than loading the whole dom into memory.

The XML says it's UTF-8 put pulldom gets confused when it found some characters in it so I made it 'sanitize' the XML by stripping out non-ASCII characters, I know this isn't strictly a good thing, but I'll spend more time on that once I have it working more (I said that in 2006...).

options

-m show the maximum size of each column - useful for adjusting column sizes in itdb.sql -n do not clear the database and the auto generated cache files -f force the loading even if the .xml file is older than the stat file

playlistlinks.py

This utility will make a nest of symlinks of your playlists. This is very useful to make CD's of playlists, or copy them to your Android phone.

Usage

playlistlinks.py --help will tell you all the command line options, I'll take you through some use cases.

playlistlinks.py -p 'My Favorite Music' will make symlinks in ~/tmp/out for your playlist called My Favorite Music. You can then use mkisofs to make a burnable cd/dvd image (mkisofs commands are printed by playlistlinks.py).

./playlistlinks.py -f iTunes --folder '4+' will take all the playlists under a folder called 4+ in your iTunes app and make symlinks for all the files under there. It will retain the iTunes directory structure which means that files will not be duplicated. You can take this output in ~/tmp/out and rsync it to antoher machine, copy over your iTunes library files and then fire up iTunes and the iTunes on that machine will be able to play the playlists you copied over. I use this to sync the best parts of my iTunes library to another machine.

The -f or --format option allow you to specify how the symlinks are created. You can use % type formatting and reference any information from a file. e.g. -f '%(Genre)s/%(Artist)s-%(Album)s' will make a files like this Rock/ACDC-Back in Black/001-Hells Bells.mp3. Note that -f is just for the directory name. The trackname and a playlist index are added by playlistlinks.

You can also randomize the track ordering and create m3u files.

You can provide multiple playlist and folder options on the command line and it will make links for all of them. You can also tell it to make m3u files for whatever files are currently in the destination folder. This is useful to sync only part of your library, but make all your playlists (but only include files that were synced).

Here's how I use it to sync music to my Moto-X phone:

  ./playlistlinks.py -m -d /tmp/pl/phone -f iTunes \
    --folder 'Awesome' \
    --nonewmusic \
    --folder 'Shuffles' \
    --folder "MP3's CD's" \
    --folder "Eclectic Music" \
    --folder 'SFW' \
    --folder '4+' \

This makes links for all my music in the playlists under Awesome (about 8Gig of 161Gig total). Then I make m3u files for all my playlists under those other categories (without copying any more music files). Then I use adbrsync.py from android-tools to copy these to my phone: adbrsync.py /tmp/pl/phone /sdcard/Music. Now I can use RocketPlayer to play my music!

sql_to_playlist

sql_to_playlist will make a playlist from the results of an sql query. Use it like this:

./sql_to_playlist --type=TV\ Shows --playlist uppercase \
  --sql "select name, Persistent_ID FROM tracks WHERE kind = 'MPEG-4 video file' and name REGEXP BINARY '^[^a-z]+\$'"

orphaned_files

Finds files that are in your library xml but do not exist on the filesystem. Also finds files under your music root that are not in your itunes library.

itdb2html

Sorry to fizzle out at the end, itdb2html is a super old janky way to make webpages about your library. I really should re-implement it using AngularJS but I just don't have a need for it anymore.

The overview page is interesting to see which artists you like and much music of each genre you have. I also like the distribution of ratings display.

arkarkark / itdb