NAME
    RDF::Generator::Void - Generate VoID descriptions based on data in an RDF
    model

VERSION
    Version 0.16

SYNOPSIS
      use RDF::Generator::Void;
      use RDF::Trine::Model;
      my $mymodel   = RDF::Trine::Model->temporary_model;
      [add some data to $mymodel here]
      my $generator = RDF::Generator::Void->new(inmodel => $mymodel);
      $generator->urispace('http://example.org');
      $generator->add_endpoints('http://example.org/sparql');
      my $voidmodel = $generator->generate;

DESCRIPTION
    This module takes a RDF::Trine::Model object as input to the constructor,
    and based on the data in that model as well as data supplied by the user,
    it creates a new model with a VoID description of the data in the model.

    For a description of VoID, see <http://www.w3.org/TR/void/>.

METHODS
  new(inmodel => $mymodel, dataset_uri => URI->new($dataset_uri), level => 1);
    The constructor. It can be called with two parameters, namely, `inmodel`
    which is a model we want to describe and `dataset_uri`, which is the URI
    we want to use for the description. Users should make sure it is possible
    to get this with HTTP. If this is not possible, you may leave this field
    empty so that a simple URN can be created for you as a default.

  `inmodel`
    Read-only accessor for the model used in description creation.

  `dataset_uri`
    Read-only accessor for the URI to the dataset.

  Property Attributes
    The below attributes concern some essential properties in the VoID
    vocabulary. They are mostly arrays, and can be manipulated using array
    methods. Methods starting with `all_` will return an array of unique
    values. Methods starting with `add_` takes a list of values to add, and
    those starting with `has_no_` return a boolean value, false if the array
    is empty.

   `all_vocabularies`, `add_vocabularies`, `has_no_vocabularies`
    Methods to manipulate a list of vocabularies used in the dataset. The
    values should be a string that represents the URI of a vocabulary.

   `all_endpoints`, `add_endpoints`, `has_no_endpoints`
    Methods to manipulate a list of SPARQL endpoints that can be used to query
    the dataset. The values should be a string that represents the URI of a
    SPARQL endpoint.

   `all_titles`, `add_titles`, `has_no_titles`
    Methods to manipulate the titles of the datasets. The values should be
    RDF::Trine::Node::Literal objects, and should be set with language.
    Typically, you would have a value per language.

   `all_licenses`, `add_licenses`, `has_no_licenses`
    Methods to manipulate a list of licenses that regulates the use of the
    dataset. The values should be a string that represents the URI of a
    license.

   `urispace`, `has_urispace`
    This method is used to set the URI prefix string that will match the
    entities in your dataset. The computation of the number of entities
    depends on this being set. `has_urispace` can be used to check if it is
    set.

  Running this stuff
   `level`, `has_level`
    Set the level of detail. 0 doesn't do any statistics or heuristics, 1 has
    some statistics for the dataset as a whole, 2 will give some partition
    statistics and 3 will give subject and object counts for property
    partitions. Setting no level will give everything.

   `stats`, `clear_stats`, `has_stats`
    Method to compute a statistical summary for the data in the dataset, such
    as the number of entities, predicates, etc. `clear_stats` will clear the
    statistics and `has_stats` will return true if exists.

   generate( [ $model ] )
    Returns the VoID as an RDF::Trine::Model. You may pass a model with
    statements as argument to this method. This model may then contain
    arbitrary RDF that will be added to the RDF model. If you do not send a
    model, one will be created for you.

AUTHORS
    Kjetil Kjernsmo `<kjetilk@cpan.org>` Toby Inkster `<tobyink@cpan.org>`

TODO
    *   URI regexps support.

    *   Technical features (esp. serializations).

    *   Example resources and root resources.

    *   Data dumps.

    *   Subject classification.

    *   Method to disable heuristics.

    *   More heuristics.

    *   Linkset descriptions.

    *   Set URI space on partitions.

    *   Use CHI to cache?

    *   Use schema introspection to generate property attributes with
        MooseX::Semantic.

BUGS
    Please report any bugs you find to
    <https://github.com/kjetilk/RDF-Generator-Void/issues>

    Note that any claim that this module will generate a void in spacetime, a
    wormhole, black hole, or funny philosophy is totally bogus and without any
    scientific merit whatsoever. The lead author has made elaborate
    precautions to avoid any such issues, and expects everyone to take his
    word for it. Oh, BTW, should it just happen anyway, it won't hurt much
    <http://news.sciencemag.org/sciencenow/2012/03/scienceshot-one-black-hole-
    wont-.html>.

SUPPORT
    You can find documentation for this module with the perldoc command.

        perldoc RDF::Generator::Void

    The Perl and RDF community website is at <http://www.perlrdf.org/> where
    you can also find a mailing list to direct questions to.

    You can also look for information at:

    *   AnnoCPAN: Annotated CPAN documentation

        <http://annocpan.org/dist/RDF-Generator-Void>

    *   CPAN Ratings

        <http://cpanratings.perl.org/d/RDF-Generator-Void>

    *   MetaCPAN

        <https://metacpan.org/module/RDF::Generator::Void>

ACKNOWLEDGEMENTS
    Many thanks to Konstantin Baierer for help with
    RDF::Generator::Void::Meta::Attribute::ObjectList.

LICENSE AND COPYRIGHT
    Copyright 2012 Toby Inkster. Copyright 2012,2013,2016 Kjetil Kjernsmo.

    This program is free software; you can redistribute it and/or modify it
    under the terms of either: the GNU General Public License as published by
    the Free Software Foundation; or the Artistic License.

    See http://dev.perl.org/licenses/ for more information.