Expand description
dupey-core: office document families, not a generic file cleaner.
Pipeline: extract → exact_hash + near_sig → family → ranked latest
candidate. Semantic embeddings are out of scope.
Re-exports§
pub use error::Error;pub use error::Result;pub use exact::byte_hash;pub use exact::byte_hash_hex;pub use exact::exact_hash;pub use exact::exact_hash_hex;pub use extract::extract;pub use extract::CanonicalText;pub use extract::DocMeta;pub use extract::Format;pub use family::cluster;pub use family::cluster_with_config;pub use family::ClusterConfig;pub use family::Family;pub use family::FamilyEdge;pub use family::FamilyLabel;pub use family::FamilyMember;pub use family::Relation;pub use family::ScannedDoc;pub use near::containment;pub use near::containment_at_least;pub use near::exact_jaccard;pub use near::near_sig;pub use near::overlap;pub use near::overlap_at_least;pub use near::score;pub use near::shingles;pub use near::NearSignature;pub use near::Overlap;pub use near::DEFAULT_CONTAINS_MIN_JACCARD;pub use near::DEFAULT_CONTAINS_THRESHOLD;pub use near::DEFAULT_NEAR_THRESHOLD;pub use near::MINHASH_CANDIDATE_THRESHOLD;pub use near::NUM_PERM;pub use rank::rank;pub use rank::FamilyRanking;pub use rank::MemberSignals;pub use rank::RankSignal;pub use rank::RankedMember;