Internet Archive audio track to MusicBrainz release matching tool

This tool uses ia_download.py to download data from the Internet Archive for a list of items into a local postgres db. Then mb_matcher.py picks items from the DB, tries to match them and saves matches/errors back to the DB.

The err field in the match table of the DB stores a free form text string of what happened during the last time that this "match" was worked on. Possible values include:

• cannot parse ia json
• too few tracks on this release
• 0 tracks found in json
• release with missing track duration
• incomplete tracklist
• key error in ia json
• match below threshold
• no toc match

Knowing the error, the script can be improved to try and match more data or improve the parsing scripts to deal the shitty inbound data.

When a match is found, the result is written to the mbid column in the match table.