Mentor RDF Parsers

Standards-compliant, fault-tolerant parsers for RDF languages, built with Chevrotain. Designed for IDEs and language tools where error recovery and concrete syntax tree (CST) access are essential.

Features

• Standards Compliance — Implements grammar productions from W3C specifications
• Fault Tolerance — Recovers from syntax errors and continues parsing
• Concrete Syntax Trees — Full token access with positions for tooling
• RDF/JS Quads — Reader classes produce RDF/JS-compliant quads
• Blank Node Tracking — Pre-assigned IDs for IDE features like "go to definition"

Supported Languages

| Language | Parser | Reader | Specification | |----------|:------:|:------:|---------------| | N-Triples | ✓ | ✓ | RDF 1.2 N-Triples | | N-Quads | ✓ | ✓ | RDF 1.2 N-Quads | | Turtle | ✓ | ✓ | RDF 1.2 Turtle | | TriG | ✓ | ✓ | RDF 1.2 TriG | | N3 | ✓ | ✓ | W3C N3 | | SPARQL 1.2 | ✓ | — | SPARQL 1.2 Query |

Installation

npm install @faubulous/mentor-rdf-parsers

Quick Start

import { TurtleLexer, TurtleParser, TurtleReader } from '@faubulous/mentor-rdf-parsers';

const input = `
  @prefix ex: <http://example.org/> .
  ex:Alice ex:knows ex:Bob .
`;

// Tokenize → Parse → Read
const lexResult = new TurtleLexer().tokenize(input);
const cst = new TurtleParser().parse(lexResult.tokens);
const quads = new TurtleReader().visit(cst);

console.log(quads[0].subject.value); // "http://example.org/Alice"

Architecture

flowchart LR
    A[Input String] --> B[Lexer]
    B -->|tokens| C[Parser]
    C -->|CST| D[Reader]
    D -->|quads| E[RDF/JS Quads]
    
    B -.->|lexing errors| F[errors]
    C -.->|parsing errors| F

| Component | Purpose | |-----------|---------| | Lexer | Tokenizes input string into a token stream | | Parser | Produces a concrete syntax tree (CST) from tokens | | Reader | Walks the CST and produces RDF/JS Quad objects |

Documentation

Examples by Language

| Language | Guide | |----------|-------| | Turtle | docs/examples/turtle.md | | N-Triples | docs/examples/n-triples.md | | N-Quads | docs/examples/n-quads.md | | TriG | docs/examples/trig.md | | N3 | docs/examples/n3.md | | SPARQL | docs/examples/sparql.md |

Advanced Topics

• Error Handling — Fault-tolerant parsing and error collection
• Blank Node IDs — Pre-assigned identifiers for IDE integration

Error Handling

Parsers collect errors instead of throwing, allowing partial results:

const parser = new TurtleParser();
const cst = parser.parse(tokens, false); // Pass false to collect errors

// Check all error sources
console.log(lexResult.errors);       // Lexer errors
console.log(parser.errors);          // Parser errors
console.log(parser.semanticErrors);  // Semantic errors (e.g., undefined prefixes)

See Error Handling Guide for details.

Token Access

Each language exports its token array for use in formatters and syntax highlighters:

import { TurtleTokens, SparqlTokens } from '@faubulous/mentor-rdf-parsers';

console.log(TurtleTokens.map(t => t.name));

Testing

npm test

The test suite includes over 1,500 tests covering all supported languages and the official W3C conformance test suites.

License

LGPL-2.1-or-later