Difference between revisions of "MusicBrainz Database/Schema"

From MusicBrainz Wiki
Jump to navigationJump to search
(link to Next Generation Schema/XML)
Line 126: Line 126:
* [[Next Generation Schema/Release Editor]]
* [[Next Generation Schema/Release Editor]]
* [[Next_Generation_Schema/Edit_User_Interface]]
* [[Next_Generation_Schema/Edit_User_Interface]]
== Web Service ==
* [[Next Generation Schema/XML]]
== Roadmap ==
== Roadmap ==

Revision as of 11:09, 28 October 2009



Same as our current artist entity, but with new "country" and "gender" attributes.

Artist Credit

List of artists, variations of artist names and pieces of text to join the artist names. Examples:

Release Group

Represents an abstract "album" entity. Technically it's a group of releases, with a specified type. The type is the same as current release type. Examples:


Real-world release object you can buy in your music store. It has release date and country, list of catalog number and label pairs, packaging type and release status. Example:

  • 1984 US release of "The Wall" by "Pink Floyd", release on label "Columbia Records" with catalog number "C2K 36183" and UPC "074643618328", it's an official release and comes with two CDs in jewel case.


Piece of physical media, included in a release. Contains information about the format, position in the release and an optional title. Has attached CD TOCs. Example:

  • CD1 of the 1984 US release of "The Wall" by "Pink Floyd"
  • CD2 of the 2005 UK release of "Aerial" by "Kate Bush", named "A Sky of Honey"


Simple list of tracks. Tracklists are used by mediums.


This object is not visible to users on it's own, only in the context of a tracklist. It contains a link to a recording, title, artist credit and the position on the tracklist.


Represents unique audio data. Has title, artist credit, duration, list of PUIDs and ISRCs. Examples:

  • Album version of the track "Into the Blue" by "Moby"
  • Remix "Into the Blue (Buzz Boys Main Room Mayhem mix)" by "Moby"
  • Remix "Into the Blue (Underground mix)" by "Moby"


One layer above recordings ("song", "composition", etc.). While recording represents audio data, work represents the abstract idea behind the recording.

  • Song "Into the Blue" by "Moby" -- all the recordings listed above will be linked to this object


Same as our current label entity.

Database Schema

This diagrams shows the core database tables and relationships between them:


Because we are adding new core entities, and there are plans for adding more, ARs between them need to be optimized. Only the absolutely necessary information is in l_*_* tables, everything else is shared:


Data Dictionary

  • artist = @id + @gid + name + sort name + (type) + (begin date) + (end date) + (country) + (comment) + {artist alias}
  • artist alias = @id + name
  • artist credit = @id + 1{artist + name + (join phrase)}
  • label = @id + @gid + name + sort name + (label code) + (type) + (begin date) + (end date) + (country) + (comment) + {label alias}
  • label alias = @id + name
  • release group = @id + @gid + artist credit + name + (type) + (comment) + {release}
  • release = @id + @gid + artist credit + name + (barcode) + (date) + (country) + (status) + (packaging) + (language) + (script) + {label + catalog number} + {medium} + (comment)
  • medium = @id + position + (format) + (name) + tracklist
  • tracklist = @id + 1{track + name + artist credit + position}
  • track = @id + @gid + artist credit + name + length + (comment) + {puid} + {isrc}
  • work = @id + @gid + artist credit + name + {work alias}
  • work alias = @id + name
  • date = (year) + (month) + (day)

Data conversion

For ARs see Next Generation Schema/Relationships_Conversion

Some edge cases are compiled in Next Generation Schema/Conversion_Edge_Cases.

Code Changes




  • memcache:
    • MBID are cached in an additional layer to avoid duplication, instead of UUID -> object, there is UUID -> rowid, and then rowid -> object.
    • Used to store:
      • artist credits
      • artists
      • labels
      • links
      • (maybe release groups?)
  • Local in-process cache:
    • Very fast access, but we have no control over invalidating it in all processes, so there needs to be low TTL for all cached items.
    • Used to store:
      • release group types
      • release statuses
      • release packaging types
      • medium formats
      • artist types
      • label types
      • countries
      • scripts
      • languages

User Interface

Web Service


See Next Generation Schema/Roadmap.