Struct ParallelParser 

pub struct ParallelParser<R: Read> { /* private fields */ }

Available on crate feature parallel only.

Parallel parser for processing RDF files using multiple threads

Implementations

impl<R: Read + Send + Sync> ParallelParser<R>

pub fn new(reader: R) -> Self

Create a new parallel parser

pub fn with_config(reader: R, config: ParallelConfig) -> Self

Create a parallel parser with custom configuration

pub fn parse_all(&mut self) -> TurtleResult<ParallelParseOutcome>

Parse the entire document in parallel

This splits the document into chunks and parses them concurrently. Chunks are cut on complete-statement boundaries (see statement_boundary) rather than raw line boundaries, so a statement that spans multiple lines (a pretty-printed predicate/object list, or a triple-quoted string containing embedded newlines) is always kept whole in a single chunk instead of being corrupted or silently dropped at a chunk boundary.

@prefix/@base/PREFIX/BASE declarations are extracted (as complete statements, so a rare multi-line prefix declaration is still captured whole) and prepended to every data chunk so prefixed names resolve correctly no matter which chunk the declaration appeared in.

In lenient mode, per-chunk parse errors are collected into the returned ParallelParseOutcome::errors instead of being silently printed to stderr.

Note: This loads the entire document into memory for splitting, since finding a statement boundary may require looking arbitrarily far ahead past a string literal.