FuddiWikiGenerator

A tool to generate wiki pages from ontology

This tools reads an ontology and gathers all subjects from it, then it uses a template file to generate articles describing the contents of this ontology. Finally, it connects to a MediaWiki API and creates/updates articles.

Read more at:

• https://fuddi.eu
• https://kb.fuddi.eu

Under the hood

It uses Apache Jena to read a given ontology and Pebble Templates.

The source file and template file have to be specified by the program arguments.

Running it

There are the following arguments that have to be set in order to run the generator:

• source-file - specifies the path to an RDF file containing the ontology.
• template-directory - specifies the path to a directory with templates used to generate articles (NOTE: the path has to start with ./ or / to be read properly by Pebble Templates).
• wikimedia.url - URL to the WikiMedia API.
• wikimedia.login - login of the bot account.
• wikimedia.password - password of the bot account.

The template-directory needs to contain at least default.txt file. This tool expects files with the .txt extension for any language used in the ontology. For example, there can be en.txt file.

Example:

java -jar fuddi-wiki-generator-1.0-SNAPSHOT-jar-with-dependencies.jar \
    -source-file ontology.rdf \
    -template-directory ./templates/ \
    -wikimedia.url https://kb.fuddi.eu/api.php \
    -wikimedia.login WikiBot \
    -wikimedia.password BotPassword

Available variables

When creating a custom template, keep in mind that there are some variables that are available to use:

• subjectUri - an object of type URIRef describing a subject URI
• subjectType - an object of type URIRef describing a subject type
• all namespaces from the source ontology as Namespace objects (eg. owl, rdf or others)
• nsName - a map with keys of type Namespace and values with namespace prefixes used to lookup namespaces (eg. you can use this to find out that Namespace(http://www.w3.org/2002/07/owl#) is prefixed as owl)
• properties - a map containing all subject properties grouped by the predicate (of type URIRef)

Example usage

Retrieving full subject URI:

{{ subjectUri.uri }}

Retrieving full subject type URI and only the identifier:

URI: {{ subjectType.uri }}
Identifier: {{ subjectType.identifier }}

Subject type namespace lookup (checking prefix associated with namespace):

{{ nsName[subjectType.namespace] }}

Retrieving an URI of owl:Thing:

{{ owl.get("Thing").uri }}

Retrieving all labels associated with subject (NOTE: it returns a list of object of type SubjectProperty):

{{ properties[rdfs.get("label")] }}

Displaying subject's label in a readable form:

{{ properties[rdfs.get("label")][0].valueLiteral.value }}

Custom functions

There are 3 custom functions that anyone can use to display values:

• printLiteral(literal) - prints literal value (NOTE: literal must be of type ValueLiteral)
• wikiLink(uri) - prints URIRef as a Wiki Link
• wikiReadable(subject) - accepts SubjectProperty, ValueLiteral or URIRef and prints it in a most appropriate way

Examples:

Printing URI as a link:

{{ wikiLink(subjectUri) }}

Printing label:

{{ printLiteral(properties[rdfs.get("label")][0]) }}

Printing all properties:

{% for property in properties %}

{{ wikiReadable(property.key) }}

{% for propertyValue in property.value %}
    {{ wikiReadable(propertyValue) }}

{% endfor %}
{% endfor %}

License

All code stored here is licensed under MIT License.