dupey_core/
lib.rs

1//! dupey-core: office document families, not a generic file cleaner.
2//!
3//! Pipeline: `extract` → `exact_hash` + `near_sig` → family → ranked latest
4//! candidate. Semantic embeddings are out of scope.
5
6pub mod error;
7pub mod exact;
8pub mod extract;
9pub mod family;
10pub mod near;
11pub mod rank;
12
13pub use error::{Error, Result};
14pub use exact::{byte_hash, byte_hash_hex, exact_hash, exact_hash_hex};
15pub use extract::{extract, CanonicalText, DocMeta, Format};
16pub use family::{
17    cluster, cluster_with_config, ClusterConfig, Family, FamilyEdge, FamilyLabel, FamilyMember,
18    Relation, ScannedDoc,
19};
20pub use near::{
21    containment, containment_at_least, exact_jaccard, near_sig, overlap, overlap_at_least, score,
22    shingles, NearSignature, Overlap, DEFAULT_CONTAINS_MIN_JACCARD, DEFAULT_CONTAINS_THRESHOLD,
23    DEFAULT_NEAR_THRESHOLD, MINHASH_CANDIDATE_THRESHOLD, NUM_PERM,
24};
25pub use rank::{rank, FamilyRanking, MemberSignals, RankSignal, RankedMember};
dupey_core/lib.rs

dupey_core/
lib.rs
