Crate rdf_parsers

Modules§

FatToken
Fingerprint
IncrementalBias: Role-preservation bias applied in the A* search during incremental re-parsing. When a token’s previous TermType (Subject/Predicate/Object) agrees with the current parse context, cost is reduced by strength (a discount on the otherwise free match); when it conflicts, cost is increased by strength (a penalty).
Parse
PrevParseInfo: Information from a previous parse needed for incremental re-parsing.
PrevToken: A single token from a previous parse, carrying the text, fingerprint, and bracket nesting depth needed for incremental re-parsing.
Spanned

Error
ParseMode: Controls whether the A* search performs fault-tolerant error recovery.
Step

effective_error_span: Returns the effective byte span of an error node.
parse
parse_fast: Parses text in non-fault-tolerant fast mode.
parse_incremental: Like parse_t_2 but, when prev is provided, diffs the token stream against the previous one and copies each old token’s parse-time fingerprint onto the matching new token via FatToken::set_old_kind. The A* scorer then uses bias to adjust scores for parses that agree or disagree with the previous token positions in the grammar rule stack.
spanned
tokenize