Struct PredicateHistogram 

pub struct PredicateHistogram {
    pub predicate_iri: String,
    pub total_count: u64,
    pub null_count: u64,
    pub distinct_count: u64,
    pub buckets: Vec<HistogramBucket>,
}

Equi-depth histogram for the object values of a specific predicate.

All buckets have approximately equal count (depth), spread across the ordered value space.

Fields

predicate_iri: String

IRI of the predicate this histogram covers.

total_count: u64

Total number of non-null values seen.

null_count: u64

Number of null / missing object values.

distinct_count: u64

Number of globally distinct object values (approximate, from build phase).

buckets: Vec<HistogramBucket>

Equi-depth buckets (sorted by lower_bound ascending).

Implementations

impl PredicateHistogram

pub fn new(predicate: &str, num_buckets: usize) -> Self

Create an empty histogram for predicate with num_buckets bucket slots.

pub fn build_from_values( predicate: &str, values: &[String], num_buckets: usize, ) -> Self

Build an equi-depth histogram from values, targeting num_buckets buckets.

The approach is:

1. Sort all (non-empty) values lexicographically.
2. Split into num_buckets equal-sized slices.
3. For each slice compute distinct count via deduplication.

Complexity: O(n log n) where n = values.len().

pub fn estimate_selectivity_eq(&self, value: &str) -> f64

Estimate the selectivity (fraction of rows) for an equality predicate object = value.

Returns a value in [0, 1]. Uses a uniform assumption within the containing bucket.

pub fn estimate_selectivity_range(&self, lo: &str, hi: &str) -> f64

Estimate the selectivity for a range predicate lo <= object <= hi.

Returns a value in [0, 1].

pub fn estimate_cardinality( &self, total_triples: u64, predicate_filter: Option<&str>, ) -> u64

Estimate the expected cardinality for a pattern that filters this predicate’s objects.

If predicate_filter is Some(value) an equality filter is applied; otherwise all triples with this predicate are returned.

total_triples is used as a fallback denominator when self.total_count is zero.

pub fn update_incremental(&mut self, new_values: &[String])

Incrementally update the histogram with additional observed values.

This is an approximate update: new values are inserted into the most appropriate existing bucket (lowest lower_bound not greater than the value), and global counts are adjusted. After many incremental updates, accuracy degrades and a full rebuild is recommended.

pub fn num_buckets(&self) -> usize

Returns the number of buckets in this histogram.

Trait Implementations

impl Clone for PredicateHistogram

fn clone(&self) -> PredicateHistogram

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for PredicateHistogram

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.