Skip to main content

kevy_text/
lib.rs

1//! kevy-text — dictionary-free full-text core:
2//! script-aware tokenization (Latin words + CJK bigrams),
3//! per-shard inverted segments maintained synchronously with writes,
4//! BM25 ranking with shard-local statistics.
5
6#![warn(missing_docs)]
7
8mod bm25;
9pub mod cold;
10mod buckets;
11mod docblobs;
12mod docvalues;
13mod edit;
14mod fields;
15mod positions;
16mod segment;
17mod token;
18
19pub use bm25::{BM25_B, BM25_K1};
20pub use segment::sorted_order;
21pub use segment::{Clauses, parse_clauses};
22pub use segment::{
23    Bucket, CorpusStats, Distinct, Facet, FacetedMatches, Filter, QueryOpts, SegmentShape, Sort, TextMatch, TextSegment, TextStats,
24};
25pub use token::{KevyTokenizer, Tokenizer, tokenize, tokenize_spans};