Module algorithm 

URDNA2015 / RDNA 2015 core algorithm implementation.

This module provides a complete, spec-faithful implementation of the W3C RDF Dataset Normalization Algorithm (URDNA2015) as described in: https://www.w3.org/TR/rdf-canon/

Algorithm Overview

URDNA2015 assigns deterministic canonical blank node identifiers to every blank node in an RDF dataset. The algorithm works in five stages:

1. Collect blank nodes — enumerate every blank node appearing in any quad.
2. Hash First-Degree Quads — for each blank node, hash the set of quads it appears in using _:a for the node itself and _:z for all other blanks (unless they already have a canonical ID).
3. Issue simple IDs — blank nodes with a unique first-degree hash receive their canonical identifier immediately (_:c14n0, _:c14n1, …) in hash order.
4. Hash N-Degree Quads — for blank nodes that still share a first-degree hash, apply the full recursive N-degree hashing algorithm which walks the RDF neighbourhood to break ties.
5. Emit canonical N-Quads — replace all blank node identifiers with their canonical form, sort the resulting quads lexicographically, and join with \n.

Reference

• W3C RDF Canonicalization 1.0: https://www.w3.org/TR/rdf-canon/
• URDNA2015 test suite: https://json-ld.github.io/rdf-dataset-normalization/tests/

Structs

Canonicalizer

The main URDNA2015 canonicalizer.

IdentifierIssuer

Issues monotonically increasing canonical blank node identifiers.

Functions

canonicalize

Canonicalize an RDF dataset (slice of RdfQuads) to a canonical N-Quads string.