EasyRdf Converter

The EasyRdf Converter is a small web-based RDF serialisation converter hosted at easyrdf.org, and a companion example script bundled with the EasyRdf PHP library. EasyRdf was written and is maintained by Nicholas Humfrey (njh) as a library designed to make it easy for PHP developers to consume and produce RDF, suitable for mixed teams of experienced and inexperienced RDF users. The public converter has run on EasyRdf 1.1.1 (the last original release, December 2020); the codebase itself has been actively continued as the community fork sweetrdf/easyrdf targeting modern PHP 8.

Functionally the converter is a thin front end over the library's parse-and-reserialise pipeline. A user supplies RDF either by pasting data or by giving a URI to fetch; the input is parsed into an in-memory EasyRdf Graph (a set of subject-predicate-object triples), and that graph is then written out in a chosen target syntax. The available input and output menus are generated dynamically from whichever formats have a registered parser or serialiser. Out of the box EasyRdf handles the core RDF serialisations: RDF/XML, Turtle, N-Triples, and RDF/JSON, plus its own resource-centric RDF/PHP array representation. Optional plug-ins (the ARC2 library and the Redland rapper command) extend parsing to further syntaxes such as RDFa. Because it operates on an abstract triple graph rather than doing text-to-text rewriting, round-tripping between formats is lossless at the triple level, subject to each serialiser's conventions.

Notably for graph tooling, EasyRdf ships a GraphViz serialiser that renders a Graph as a GraphViz DOT document (and, if GraphViz is installed, on to PNG/SVG), with options to show labels instead of full URIs and to produce more concise diagrams. This places it alongside RDF-to-graph utilities such as RDF Grapher and Zazuko RDF Conv., and complements heavier RDF stacks like Jena/RIOT, RDFLib (Python), and dotNetRDF when only lightweight PHP-side conversion or visualisation is needed.

Its strengths are simplicity, zero external dependencies for the core formats, and a clean object model. Limitations follow from its scope: the built-in format set is small (no native JSON-LD, N-Quads, or TriG in the core), advanced or streaming/large-dataset workloads are better served by dedicated triplestores or converters, and the original hosted service tracks an old release, so production use is best via the maintained fork.