SYNOPSIS

    See cpandb-cpanmeta script.

DESCRIPTION

    CPAN::SQLite::CPANMeta accompanies CPAN::SQLite by adding information
    from CPAN Meta files (META.json or META.yml) in the release files.

    CPAN::SQLite only parses the $CPAN/modules/02packages.details.txt.gz
    and $CPAN/authors/01mailrc.txt.gz. From these files these information
    can be retrieved: package/module names, the list of authors (PAUSE
    ID's, names, and emails), and the list of indexed release
    files/tarballs. Distribution names can be guessed from the release
    files, but some files do not follow convention.

    To get more information, one will need to parse the meta files (either
    from *.meta files in the authors directory, or by extracting META.yml
    or META.json from release tarballs). This module does this and put the
    resulting information to a few extra database tables. Currently the
    information extracted/recorded are: dependencies (tables: files, deps).

SEE ALSO

    Front-end for this module: cpandb-cpanmeta.

    CPAN::SQLite and its front-end cpandb.

    CPANDB::Generator CPANDB bills itself as a "unified database of CPAN
    metadata information" and is a SQLite database that contains more
    information about CPAN from various sources, e.g.: number of ratings
    from http://cpanratings.perl.org, upload date and test results from
    CPAN Testers, number of tickets from http://rt.cpan.org. However,
    generating CPANDB locally will require downloading gigabytes of data
    and take hours (generating CPAN::SQLite database +
    CPAN::SQLite::CPANMeta, on the other hand, only takes a few minutes;
    regular updating typically only takes a few seconds). An online cache
    is available and can be used via CPANDB but you cannot control the
    update frequency.