A reader of RDF/XML (commons 0019-mizan-integrationsFR-024, FR-025): the text of an outside ontology published as RDF/XML (the W3C's SKOS, OWL, PROV-O, SHACL, DCAT) as triples, with what the typed form cannot hold named as findings and never dropped silently. It is hand-written TypeScript, and needs no Exception (0017-two-languagesFR-003): it read every W3C RDF/XML file it was tried on to the graph rdflib reads (the decision and the evidence are in 0019FR-024).
Two layers in one file. The first is a minimal XML reader: elements, attributes, text, CDATA, comments, processing instructions, character references, the five predefined entities, the entities a <!DOCTYPE> internal subset declares (<!ENTITY rdf "http://...#">, which RDF/XML files use to shorten IRIs), namespaces and xml:lang and xml:base; it refuses what is not well formed with the line. The second is the RDF/XML grammar: node elements (rdf:Description and typed nodes, rdf:about, rdf:ID, rdf:nodeID, anonymous nodes), property elements (literals, rdf:resource, rdf:datatype, nested nodes, rdf:parseType="Resource" and "Collection"), property attributes, rdf:li counting, rdf:ID reification, and relative IRIs resolved against xml:base (RFC 3986). Blank nodes keep their structure as blank terms (labels genid-N in document order, and rdf:nodeID as written). A parseType="Literal" (an XML literal, whose canonical form this reader does not compute) and an external entity are unsupported findings, the property or entity is not read and the rest is.
RDF/XML text as triples, with the findings of what the typed form cannot hold; a document that is not well-formed XML or not RDF/XML ends the read with one syntax finding. base is the IRI the document was read from, which a relative rdf:about resolves against where the document has no xml:base.
RDF/XML text as triples, with the findings of what the typed form cannot hold; a document that is not well-formed XML or not RDF/XML ends the read with one syntax finding. base is the IRI the document was read from, which a relative rdf:about resolves against where the document has no xml:base.