talisman/stemmers/french/carry.js

Straightforward fuzzy matching, information retrieval and NLP building blocks for JavaScript.

'use strict';

Object.defineProperty(exports, "__esModule", {
  value: true
});
exports.default = carry;
/**
 * Talisman stemmers/french/carry
 * ===============================
 *
 * The Carry stemmer for the French language.
 *
 * [Reference]:
 * http://www.otlet-institute.org/docs/Carry.pdf
 *
 * [Article]:
 * Carry, un algorithme de désuffixation pour le français. M. Paternostre,
 * P. Francq, J. Lamoral, D. Wartel et M. Saerens. 2002
 *
 * [Note]:
 * This algorithm has been edited to handle some more cases and is thus
 * lightly different from the original paper (modifications by Guillaume
 * Plique).
 */

/**
 * Constants.
 */
var VOWELS = 'aeiouyâàëéêèïîôûùœæ',
    V = '[' + VOWELS + ']',
    C = '[^' + VOWELS + ']';

var LC = new RegExp('^' + C + '+'),
    TV = new RegExp(V + '+$'),
    M = new RegExp('(' + V + '+' + C + '+)');

/**
 * Helpers.
 */
function computeM(string) {

  // Removing leading consonants
  string = string.replace(LC, '').replace(TV, '');

  return (string.match(M) || []).length;
}

/**
 * Rules.
 */
var STEP1 = [[0, 'issaient'], [0, 'ellement', 'el'], [0, 'issement'], [0, 'alement', 'al'], [0, 'eraient'], [0, 'iraient'], [0, 'eassent'], [0, 'ussent'], [0, 'amment'], [0, 'emment'], [0, 'issant'], [0, 'issent'], [0, 'assent'], [0, 'eaient'], [0, 'issait'], [0, 'èrent'], [0, 'erent'], [0, 'irent'], [0, 'erait'], [0, 'irait'], [0, 'iront'], [0, 'eront'], [0, 'ement'], [0, 'aient'], [0, 'îrent'], [0, 'eont'], [0, 'eant'], [0, 'eait'], [0, 'ient'], [0, 'ent'], [0, 'ont'], [0, 'ant'], [0, 'eât'], [0, 'ait'], [0, 'at'], [0, 'ât'], [0, 'it'], [0, 'ît'], [0, 't'], [0, 'uction'], [1, 'ication'], [1, 'iation'], [1, 'ation'], [0, 'ition'], [0, 'tion'], [1, 'ateur'], [1, 'teur'], [0, 'eur'], [0, 'ier'], [0, 'er'], [0, 'ir'], [0, 'r'], [0, 'eassiez'], [0, 'issiez'], [0, 'assiez'], [0, 'ussiez'], [0, 'issez'], [0, 'assez'], [0, 'eriez'], [0, 'iriez'], [0, 'erez'], [0, 'irez'], [0, 'iez'], [0, 'ez'], [0, 'erai'], [0, 'irai'], [0, 'eai'], [0, 'ai'], [0, 'i'], [0, 'ira'], [0, 'era'], [0, 'ea'], [0, 'a'], [0, 'f', 'v'], [0, 'yeux', 'oeil'], [0, 'eux'], [0, 'aux', 'al'], [0, 'x'], [0, 'issante'], [1, 'atrice'], // Added
[0, 'eresse'], [0, 'eante'], [0, 'easse'], [0, 'eure'], [0, 'esse'], [0, 'asse'], [0, 'ance'], [0, 'ence'], [0, 'aise'], [0, 'euse'], [0, 'oise', 'o'], [0, 'isse'], [0, 'ante'], [0, 'ouse', 'ou'], [0, 'ière'], [0, 'ete'], [0, 'ète'], [0, 'iere'], [0, 'aire'], [1, 'ure'], [0, 'erie'], [0, 'étude'], [0, 'etude'], [0, 'itude'], [0, 'ade'], [0, 'isme'], [0, 'age'], [0, 'trice'], [0, 'cque', 'c'], [0, 'que', 'c'], [0, 'eille', 'eil'], [0, 'elle'], [0, 'able'], [0, 'iste'], [0, 'ulle', 'ul'], [0, 'gue', 'g'], [0, 'ette'], [0, 'nne', 'n'], [0, 'itée'], [0, 'ité'], [0, 'té'], [0, 'ée'], [0, 'é'], [0, 'usse'], [0, 'aise'], [0, 'ate'], [0, 'ite'], [0, 'ee'], [0, 'e'], [0, 'issements'], [0, 'issantes'], [1, 'ications'], [0, 'eassions'], [0, 'eresses'], [0, 'issions'], [0, 'assions'], [1, 'atrices'], // Added
[1, 'iations'], [0, 'issants'], [0, 'ussions'], [0, 'ements'], [0, 'eantes'], [0, 'issons'], [0, 'assons'], [0, 'easses'], [0, 'études'], [0, 'etudes'], [0, 'itudes'], [0, 'issais'], [0, 'trices'], [0, 'eilles', 'eil'], [0, 'irions'], [0, 'erions'], [1, 'ateurs'], [1, 'ations'], [0, 'usses'], [0, 'tions'], [0, 'ances'], [0, 'entes'], [1, 'teurs'], [0, 'eants'], [0, 'ables'], [0, 'irons'], [0, 'irais'], [0, 'ences'], [0, 'ients'], [0, 'ieres'], [0, 'eures'], [0, 'aires'], [0, 'erons'], [0, 'esses'], [0, 'euses'], [0, 'ulles', 'ul'], [0, 'cques', 'c'], [0, 'elles'], [0, 'ables'], [0, 'istes'], [0, 'aises'], [0, 'asses'], [0, 'isses'], [0, 'oises', 'o'], [0, 'tions'], [0, 'ouses', 'ou'], [0, 'ières'], [0, 'eries'], [0, 'antes'], [0, 'ismes'], [0, 'erais'], [0, 'eâtes'], [0, 'eâmes'], [0, 'itées'], [0, 'ettes'], [0, 'ages'], [0, 'eurs'], [0, 'ents'], [0, 'ètes'], [0, 'etes'], [0, 'ions'], [0, 'ités'], [0, 'ites'], [0, 'ates'], [0, 'âtes'], [0, 'îtes'], [0, 'eurs'], [0, 'iers'], [0, 'iras'], [0, 'eras'], [1, 'ures'], [0, 'ants'], [0, 'îmes'], [0, 'ûmes'], [0, 'âmes'], [0, 'ades'], [0, 'eais'], [0, 'eons'], [0, 'ques', 'c'], [0, 'gues', 'g'], [0, 'nnes', 'n'], [0, 'ttes'], [0, 'îtes'], [0, 'tés'], [0, 'ons'], [0, 'ais'], [0, 'ées'], [0, 'ees'], [0, 'ats'], [0, 'eas'], [0, 'ts'], [0, 'rs'], [0, 'as'], [0, 'es'], [0, 'fs', 'v'], [0, 'és'], [0, 'is'], [0, 's'], [0, 'eau'], [0, 'au']];

var STEP2 = [[1, 'ation'], [1, 'ition'], [1, 'tion'], [1, 'ent'], [1, 'el'], [0, 'i']];

var STEP3 = [[0, 'll', 'l'], [0, 'mm', 'm'], [0, 'nn', 'n'], [0, 'pp', 'p'], [0, 'tt', 't'], [0, 'ss', 's'], [0, 'y'], [0, 't'], [0, 'qu', 'c']];

/**
 * Function used to apply a set of rules to the current stem.
 *
 * @param  {string} stem - Target stem.
 * @return {string}      - The resulting stem.
 */
function applyRules(rules, stem) {

  for (var i = 0, l = rules.length; i < l; i++) {
    var _rules$i = rules[i],
        min = _rules$i[0],
        pattern = _rules$i[1],
        _rules$i$ = _rules$i[2],
        replacement = _rules$i$ === undefined ? '' : _rules$i$;


    if (stem.slice(-pattern.length) === pattern) {
      var newStem = stem.slice(0, -pattern.length) + replacement,
          m = computeM(newStem);

      if (m <= min) continue;

      return newStem;
    }
  }

  return stem;
}

/**
 * Function stemming the given world using the Carry algorithm for the French
 * language.
 *
 * @param  {string} word - The word to stem.
 * @return {string}      - The resulting stem.
 */
function carry(word) {
  var stem = word.toLowerCase();

  stem = applyRules(STEP1, stem);
  stem = applyRules(STEP2, stem);
  stem = applyRules(STEP3, stem);

  return stem;
}
module.exports = exports['default'];