Lodestar.Text 0.6.0. This page is frozen at that release. Read the current documentation for what main says now. A link to a decision or a migration page follows main, and leaves the archive.

BkTree.WithinDistance

Every indexed item within a radius of the query, nearest first.

public IReadOnlyList<BkTreeMatch> WithinDistance(string query, int maxDistance, int? limit = null)

Parameters — query is the string to search around. maxDistance is the inclusive radius in the tree's own metric; 0 finds only an exactly equal item. limit caps the returned list at the nearest hits, never the first the traversal happened to reach — and it does not bound the search, because a nearer hit can be found at any point in it.

Returns — IReadOnlyList<BkTreeMatch>, distance ascending, ties by insertion order. Empty when nothing is within the radius, and when the tree is empty.

Example — a small dictionary and the single-edit neighbourhood of a misspelling.

using Lodestar.Text.Indexing;

BkTree tree = BkTree.OverLevenshtein();
tree.AddRange(["book", "books", "boo", "cook", "cake"]);

IReadOnlyList<BkTreeMatch> hits = tree.WithinDistance("bok", 1);
int found = hits.Count;        // => 2
string nearest = hits[0].Item; // => book

Remarks — the traversal descends into a child only when its key lies in [d - maxDistance, d + maxDistance], which is the triangle inequality and is why the metric must satisfy it. At int.MaxValue the bound is skipped rather than computed, since it would overflow and admits everything anyway.

The pruning is worth using at a radius of 1, where it runs in roughly half the wall-clock time of a length-filtered scan; past that radius it measures slower, not merely no longer faster — see BkTree for the measured table. A large radius over a large corpus is a linear scan wearing a tree.

Applies to — net10.0, netstandard2.0.

See also — BkTree.Nearest, BkTreeMatch.