Struct InvertedIndex 

pub struct InvertedIndex { /* private fields */ }

Inverted index mapping blocking keys to record IDs.

Implementations

impl InvertedIndex

pub fn new() -> Self

pub fn insert(&mut self, record_id: RecordId, keys: Vec<String>)

pub fn lookup_union(&self, keys: &[String], exclude: RecordId) -> Vec<RecordId>

Returns all record IDs sharing at least one key with the query keys, excluding exclude (the querying record itself). Result is deduplicated.

pub fn lookup_union_capped( &self, keys: &[String], exclude: RecordId, max_bucket_size: usize, ) -> Vec<RecordId>

Like Self::lookup_union but skips any bucket whose size exceeds max_bucket_size. Overfull buckets have low selectivity and produce O(n²) spurious pairs; capping them prevents unbounded memory growth.

Pass max_bucket_size = 0 to disable the cap (same as lookup_union).

pub fn bucket_size(&self, key: &str) -> usize

Returns the size of a specific bucket, or 0 if not present.

pub fn oversized_buckets(&self, max_size: usize) -> usize

Returns the number of buckets exceeding max_size.

pub fn all_pairs( &self, id_to_idx: &HashMap<RecordId, usize>, max_bucket_size: usize, ) -> Vec<(usize, usize)>

Enumerate all canonical (i < j) candidate pairs directly from bucket contents.

Pass max_bucket_size = 0 to disable the cap.

pub fn remove(&mut self, record_id: RecordId)

pub fn len(&self) -> usize

pub fn is_empty(&self) -> bool

pub fn record_count(&self) -> usize

Trait Implementations

impl BlockIndex for InvertedIndex

fn insert(&mut self, record_id: RecordId, keys: Vec<String>)

Index record_id under the given set of blocking keys.

fn lookup_union(&self, keys: &[String], exclude: RecordId) -> Vec<RecordId>

Return all record IDs sharing at least one key with keys, excluding exclude (the querying record itself). Result must be deduplicated.

fn remove(&mut self, record_id: RecordId)

Remove all index entries for record_id.

fn as_any(&self) -> &dyn Any

fn as_any_mut(&mut self) -> &mut dyn Any

impl Default for InvertedIndex

fn default() -> Self

Returns the “default value” for a type.