ECMAScript RDFa Parser

A fairly modular, cross-platform, extensible package to extract RDF information from RDFa-enabled documents.

Features:

• Sometimes extracts RDF statements
• Passes all compliant XML tests
• Walks DOM tree to generate an RDF graph
• Extensible design for parsing supersets of RDFa and supporting new media types

Goals:

• Consume DOM nodes or SAX events
• Output a complete graph or "Statement" events
• Support custom types of nodes (e.g. variables)
• Query for DOM nodes by data they contain
• Linting of RDFa documents for:
  • check property order
  • reverse/forward incompatability
  • shadowed definitions
  • defining over registered URI schemes
  • non-normalized IRI references
  • Invalid IRIs
  • Invalid CURIEs
  • Term is used without an active vocabulary

Features

Parse a DOM tree into an RDF graph

const rdfa = require('rdfa');
const fs = require('fs');
const DOMParser = require('xmldom').DOMParser;

const filepath = './test/rdfa.github.io/test-suite/test-cases/rdfa1.1-lite/xhtml1/0021.xhtml';
const document = new DOMParser().parseFromString(fs.readFileSync(filepath, 'UTF-8'), 'text/xml');
const result = rdfa.parseDOM(rdfa.RDFaXMLParser, 'http://example.com/', document);
result.outputGraph.forEach(function(n){ console.log(n.toTurtle()); });

<http://example.com/> <http://purl.org/dc/elements/1.1/creator> "Mark Birbeck" .

Support for standards-based XML parsers

There is no included document parser, instead supports reading from a DOM (W3C or compatible), or a SAX-like event stream.

Tests are written for the following packages:

• http://npmjs.com/package/xmldom

Support for different media types

Supports plain XML, HTML, and SVG out-of-the-box, and is extendible for new host languages.

Supported host languages:

• RDFaXMLParser - for plain XML documents
• RDFaXHTMLParser - for HTML documents delivered in application/xhtml+xml
• RDFaHTMLParser - for HTML documents delivered in text/html

File index

• bin/httpd.js - an HTTP server for use with the RDFa Test Suite
• bin/rdfa-turtle.js - read a file and print extracted triples
• defaults.json - list of RDFa Initial Context vocabulary & terms
• index.js - the bulk of the logic
• test/suite.js - runs the library against a local copy of the RDFa test suite

API

parseDOM(Processor, base, document, options)

Extracts RDF statements out of a DOM document document, assuming a URI base base (always ignoring the URI in the DOM node, if any).

Returns an RDFaParser instance (see below).

• Processor: a reference to a subclass of RDFaParser
• base: the URI base for the document (where the document was downloaded from)
• document: DOM document
• options: object with additional configuration
  • forceVersion: RDFa version to use. This is normally detected.
  • defaultLanguage: default language for when no language is specified by the document (use the language specified in the Content-Language header, if any)

RDFaParser

• outputGraph - instance of RDF.Graph
• processorGraph - instance of RDF.Graph

Maintains state during processing of a document. Pass a subclass of this to parseDOM as the first argument Processor.

RDFaXMLParser

RDFaParser with the default RDF context loaded.

RDFaXHTMLParser

RDFaParser extended with the parsing rules for application/xhtml+xml documents.

RDFaHTMLParser

RDFaParser extended with the parsing rules for text/html documents.

RDFaContext

Represents an RDFa processing context that's generated during processing of an element and passed to child elements.