@technobuddha/library/src/levenshtein-distance.ts

A large library of useful functions

import { create2dArray } from './create2d-array.ts';

/**
 * Options for the {@link levenshteinDistance} function
 * @group String
 * @category Fuzzy Match
 */
export type LevenshteinDistanceOptions = {
  /** The comparison will ignore case */
  caseInsensitive?: boolean;
};

/**
 * Compute the levenshtein distance between two strings (similarity)
 * @param input - The string
 * @param comparedTo - The string to compare to
 * @param options - see {@link LevenshteinDistanceOptions}
 * @defaultValue caseInsensitive true
 * @returns the levenshteinDistance between the two strings (0 for no similarity through 1 for equal)
 * @group String
 * @category Fuzzy Match
 */
export function levenshteinDistance(
  input: string,
  comparedTo: string,
  { caseInsensitive = true }: LevenshteinDistanceOptions = {},
): number {
  let argInput = input;
  let argComparedTo = comparedTo;

  if (argInput.length === 0 || comparedTo.length === 0) {
    return Math.max(argInput.length, comparedTo.length);
  }

  if (caseInsensitive) {
    argInput = argInput.toLocaleLowerCase();
    argComparedTo = argComparedTo.toLocaleLowerCase();
  }

  const inputLen = argInput.length;
  const comparedToLen = argComparedTo.length;
  const matrix = create2dArray(inputLen, comparedToLen, 0);

  //initialize
  for (let i = 0; i < inputLen; ++i) {
    matrix[i][0] = i;
  }
  for (let i = 0; i < comparedToLen; ++i) {
    matrix[0][i] = i;
  }

  //analyze
  for (let i = 1; i < inputLen; ++i) {
    const si = argInput.charAt(i - 1);
    for (let j = 1; j < comparedToLen; ++j) {
      const tj = argComparedTo.charAt(j - 1);
      const cost = si === tj ? 0 : 1;
      const above = matrix[i - 1][j];
      const left = matrix[i][j - 1];
      const diag = matrix[i - 1][j - 1];
      let cell = Math.min(Math.min(above + 1, left + 1), diag + cost);

      //transposition
      if (i > 1 && j > 1) {
        let trans = matrix[i - 2][j - 2] + 1;
        if (argInput.charAt(i - 2) !== argComparedTo.charAt(j - 1)) {
          trans++;
        }
        if (argInput.charAt(i - 1) !== argComparedTo.charAt(j - 2)) {
          trans++;
        }
        if (cell > trans) {
          cell = trans;
        }
      }
      matrix[i][j] = cell;
    }
  }

  return matrix[inputLen - 1][comparedToLen - 1];
}