@inseefr/lunatic 0.2.62 → 0.3.0-v2

package/src/utils/suggester-workers/commons-tokenizer/create-fields-tokenizer.js

@@ -0,0 +1,56 @@
+import prepareStringIndexation from './prepare-string-indexation';
+import softTokenizer from './soft-tokenizer';
+import tokenizer from 'string-tokenizer';
+import getRegExpFromPattern from './get-regexp-from-pattern';
+
+function defaultTokenizeIt(string) {
+	return [prepareStringIndexation(string)];
+}
+
+function toArray(content) {
+	return Array.isArray(content) ? content : [content];
+}
+
+function tokensToArray(tokenized) {
+	return Object.entries(tokenized).reduce(function (a, [k, values]) {
+		if (k.startsWith('pattern')) {
+			return [...a, ...toArray(values)];
+		}
+		return a;
+	}, []);
+}
+
+function createFieldsTokenizer(field, filtersStack) {
+	const {
+		name,
+		rules = [],
+		min,
+		language = 'French',
+		stemmer = true,
+		synonyms = {},
+	} = field;
+	if (rules === 'soft') {
+		return softTokenizer;
+	}
+	if (rules.length) {
+		const tokenRules = rules.reduce(function (rulesMap, pattern, index) {
+			return {
+				...rulesMap,
+				[`pattern${name}${index}`]: getRegExpFromPattern(pattern),
+			};
+		}, {});
+
+		return function (string) {
+			const what = tokenizer().input(string).tokens(tokenRules).resolve();
+			return filtersStack(tokensToArray(what), {
+				min,
+				language,
+				stemmer,
+				synonyms,
+			});
+		};
+	}
+	return defaultTokenizeIt;
+}
+
+export default createFieldsTokenizer;

package/src/utils/suggester-workers/commons-tokenizer/create-filter-stop-words.js

@@ -0,0 +1,17 @@
+import defaultStopWords from './stop-words';
+
+function createFilterStopWords(stops = defaultStopWords) {
+	const mapSW = stops.reduce(function (a, w) {
+		return { ...a, [w]: undefined };
+	}, {});
+	return function filter(tokens) {
+		return tokens.reduce(function (a, t) {
+			if (t in mapSW) {
+				return a;
+			}
+			return [...a, t];
+		}, []);
+	};
+}
+
+export default createFilterStopWords;

package/src/utils/suggester-workers/commons-tokenizer/filters/compose-filters.js

@@ -0,0 +1,10 @@
+export function composeFilters(...filters) {
+	return filters.reduce(
+		function (next, current) {
+			return (tokens, args) => next(current(tokens, args), args);
+		},
+		(t) => t
+	);
+}
+
+export default composeFilters;

package/src/utils/suggester-workers/commons-tokenizer/filters/create-filter-stop-words.js

@@ -0,0 +1,17 @@
+import defaultStopWords from './stop-words';
+
+function createFilterStopWords(stops = defaultStopWords) {
+	const mapSW = stops.reduce(function (a, w) {
+		return { ...a, [w]: undefined };
+	}, {});
+	return function filter(tokens) {
+		return tokens.reduce(function (a, t) {
+			if (t in mapSW) {
+				return a;
+			}
+			return [...a, t];
+		}, []);
+	};
+}
+
+export default createFilterStopWords;

package/src/utils/suggester-workers/commons-tokenizer/filters/create-filter-stop-words.spec.js

@@ -0,0 +1,14 @@
+import createFilterSW from './create-filter-stop-words';
+
+describe('create-filter-stop-words', function () {
+	it('only last ', function () {
+		const sw = ['un', 'de'];
+		const tokens = ['un', 'de', 'trois'];
+
+		const filter = createFilterSW(sw);
+
+		const results = filter(tokens);
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(1);
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-accents.js

@@ -0,0 +1,12 @@
+import removeAccents from 'remove-accents';
+
+function filterAccents(tokens = []) {
+	return tokens.map(function (token) {
+		if (typeof token === 'string') {
+			return removeAccents(token).toLowerCase();
+		}
+		return token;
+	});
+}
+
+export default filterAccents;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-accents.spec.js

@@ -0,0 +1,12 @@
+import filterATL from './filter-accents';
+
+describe('filter-length', function () {
+	it('maj accent', function () {
+		const tokens = ['UN', 'éà ù'];
+		const results = filterATL(tokens, { min: 3 });
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(2);
+		expect(results[0]).toBe('un');
+		expect(results[1]).toBe('ea u');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-double.js

@@ -0,0 +1,12 @@
+function filterDouble(tokens = []) {
+	const map = {};
+	return tokens.reduce(function (a, token) {
+		if (token in map) {
+			return a;
+		}
+		map[token] = true;
+		return [...a, token];
+	}, []);
+}
+
+export default filterDouble;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-double.spec.js

@@ -0,0 +1,20 @@
+import filterDouble from './filter-double';
+
+describe('filter-double', function () {
+	it('empty', function () {
+		const tokens = ['un', 'un'];
+		const results = filterDouble(tokens);
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(1);
+		expect(results[0]).toBe('un');
+	});
+
+	it('remove one', function () {
+		const tokens = ['un', 'un', 'long'];
+		const results = filterDouble(tokens, { min: 3 });
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(2);
+		expect(results[0]).toBe('un');
+		expect(results[1]).toBe('long');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-length.js

@@ -0,0 +1,7 @@
+function filterLength(tokens, { min = 2 }) {
+	return tokens.filter(function (token) {
+		return token.length >= min;
+	});
+}
+
+export default filterLength;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-length.spec.js

@@ -0,0 +1,18 @@
+import filterLength from './filter-length';
+
+describe('filter-length', function () {
+	it('empty', function () {
+		const tokens = ['un', 'no'];
+		const results = filterLength(tokens, { min: 3 });
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(0);
+	});
+
+	it('remove one', function () {
+		const tokens = ['un', 'long'];
+		const results = filterLength(tokens, { min: 3 });
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(1);
+		expect(results[0]).toBe('long');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-stemmer.js

@@ -0,0 +1,13 @@
+import getStemmer from '../get-stemmer';
+
+function filterStemmer(tokens, { language = 'French', stemmer = true }) {
+	if (stemmer) {
+		const stemming = getStemmer(language);
+		return tokens.map(function (token) {
+			return stemming(token);
+		});
+	}
+	return tokens;
+}
+
+export default filterStemmer;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-stemmer.spec.js

@@ -0,0 +1,12 @@
+import filterStemmer from './filter-stemmer';
+
+describe('filter-stemmer', function () {
+	it('commerce', function () {
+		const tokens = ['commerce', 'commercial'];
+		const results = filterStemmer(tokens, { language: 'French' });
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(2);
+		expect(results[0]).toBe('commerc');
+		expect(results[1]).toBe('commercial');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-synonyms.js

@@ -0,0 +1,36 @@
+const MAPS_FROM_ARRAY = new Map(); // Pour éviter de recalculer une map à chaque fois
+
+function filterFromObject(tokens, synonyms) {
+	return tokens.reduce(function (prec, token) {
+		if (token in synonyms) {
+			return [...prec, token, ...synonyms[token]];
+		}
+		return [...prec, token];
+	}, []);
+}
+
+function buildSynonymsMap(array) {
+	return array.reduce(function (map, { source, target }) {
+		return { ...map, [source]: target };
+	}, {});
+}
+
+function filterFromArray(tokens, synonyms) {
+	if (!MAPS_FROM_ARRAY.has(synonyms)) {
+		MAPS_FROM_ARRAY.set(synonyms, buildSynonymsMap(synonyms));
+	}
+	return filterFromObject(tokens, MAPS_FROM_ARRAY.get(synonyms));
+}
+
+function filterSynonyms(tokens, { synonyms }) {
+	if (Array.isArray(synonyms) && synonyms.length) {
+		return filterFromArray(tokens, synonyms);
+	}
+	if (typeof synonyms === 'object') {
+		return filterFromObject(tokens, synonyms);
+	}
+
+	return tokens;
+}
+
+export default filterSynonyms;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-synonyms.spec.js

@@ -0,0 +1,12 @@
+import filterSynonyms from './filter-synonyms';
+
+describe('filter-synonyms', function () {
+	it('', function () {
+		// const tokens = ['saint', 'sainte'];
+		// const results = filterStemmer(tokens, { language: 'French' });
+		// expect(Array.isArray(results)).toBe(true);
+		// expect(results.length).toBe(2);
+		// expect(results[0]).toBe('commerc');
+		// expect(results[1]).toBe('commercial');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-to-lower.js

@@ -0,0 +1,10 @@
+function filterToLower(tokens) {
+	return tokens.map(function (token) {
+		if (typeof token === 'string') {
+			return token.toLocaleLowerCase();
+		}
+		return token;
+	});
+}
+
+export default filterToLower;

package/src/utils/suggester-workers/commons-tokenizer/filters/filter-to-lower.spec.js

@@ -0,0 +1,12 @@
+import filterToLower from './filter-to-lower';
+
+describe('filter-to-lower', function () {
+	it('UN', function () {
+		const tokens = ['UN', 'no'];
+		const results = filterToLower(tokens);
+		expect(Array.isArray(results)).toBe(true);
+		expect(results.length).toBe(2);
+		expect(results[0]).toBe('un');
+		expect(results[1]).toBe('no');
+	});
+});

package/src/utils/suggester-workers/commons-tokenizer/filters/index.js

@@ -0,0 +1,2 @@
+export { default as composeFilters } from './compose-filters';
+export { default as createFilterStopWords } from './create-filter-stop-words';

package/src/utils/suggester-workers/commons-tokenizer/filters/stop-words.js

@@ -0,0 +1,118 @@
+const STOP_WORDS = [
+  'alors',
+  'au',
+  'aucuns',
+  'aussi',
+  'autre',
+  'avant',
+  'avec',
+  'avoir',
+  'bon',
+  'car',
+  'ce',
+  'cela',
+  'ces',
+  'ceux',
+  'chaque',
+  'ci',
+  'comme',
+  'comment',
+  'dans',
+  'des',
+  'du',
+  'dedans',
+  'dehors',
+  'depuis',
+  'devrait',
+  'doit',
+  'donc',
+  'dos',
+  'debut',
+  'elle',
+  'elles',
+  'en',
+  'encore',
+  'essai',
+  'est',
+  'et',
+  'eu',
+  'fait',
+  'faites',
+  'fois',
+  'font',
+  'hors',
+  'ici',
+  'il',
+  'ils',
+  'je',
+  'juste',
+  'le',
+  'les',
+  'leur',
+  'la',
+  'ma',
+  'maintenant',
+  'mais',
+  'mes',
+  'mien',
+  'moins',
+  'mon',
+  'mot',
+  'meme',
+  'ni',
+  'nommes',
+  'notre',
+  'nous',
+  'ou',
+  'par',
+  'parce',
+  'pas',
+  'peut',
+  'peu',
+  'plupart',
+  'pour',
+  'pourquoi',
+  'quand',
+  'que',
+  'quel',
+  'quelle',
+  'quelles',
+  'quels',
+  'qui',
+  'sa',
+  'sans',
+  'ses',
+  'seulement',
+  'si',
+  'sien',
+  'son',
+  'sont',
+  'sous',
+  'soyez',
+  'sois',
+  'sujet',
+  'sur',
+  'ta',
+  'tandis',
+  'tellement',
+  'tels',
+  'tes',
+  'ton',
+  'tous',
+  'tout',
+  'trop',
+  'tres',
+  'tu',
+  'voient',
+  'vont',
+  'votre',
+  'vous',
+  'vu',
+  'ca',
+  'etaient',
+  'etat',
+  'etions',
+  'ete',
+  'etre',
+];
+export default STOP_WORDS;

package/src/utils/suggester-workers/commons-tokenizer/get-regexp-from-pattern.js

@@ -0,0 +1,8 @@
+function get(pattern) {
+  if (typeof pattern === 'string') {
+    return new RegExp(pattern);
+  }
+  return pattern;
+}
+
+export default get;

package/src/utils/suggester-workers/commons-tokenizer/get-stemmer.js

@@ -0,0 +1,18 @@
+import Snowball from 'snowball';
+
+const STEMMERS = {};
+
+function getStemmer(language = 'French') {
+	if (!(language in STEMMERS)) {
+		const stemmer = new Snowball(language);
+		STEMMERS[language] = function (token) {
+			stemmer.setCurrent(token);
+			stemmer.stem();
+			return stemmer.getCurrent();
+		};
+	}
+
+	return STEMMERS[language];
+}
+
+export default getStemmer;

package/src/utils/suggester-workers/commons-tokenizer/index.js

@@ -0,0 +1,9 @@
+export { default as getStemmer } from './get-stemmer';
+export { default as createTokenizer } from './create-entity-tokenizer';
+export { default as prepareStringIndexation } from './prepare-string-indexation';
+export { default as filterStemmer } from './filters/filter-stemmer';
+export { default as filterLength } from './filters/filter-length';
+export { default as filterDouble } from './filters/filter-double';
+export { default as filterAccents } from './filters/filter-accents';
+export { default as filterToLower } from './filters/filter-to-lower';
+export { default as getRegExpFromPattern } from './get-regexp-from-pattern';

package/src/utils/suggester-workers/commons-tokenizer/prepare-string-indexation.js

@@ -0,0 +1,13 @@
+import removeAccents from 'remove-accents';
+
+function prepare(token = '', replacement = '') {
+	if (typeof token === 'string') {
+		return removeAccents(token.trim().toLowerCase()).replace(
+			/[- ']/g,
+			replacement
+		);
+	}
+	return token;
+}
+
+export default prepare;

package/src/utils/suggester-workers/commons-tokenizer/soft-tokenizer.js

@@ -0,0 +1,7 @@
+import { prepareStringIndexation } from '../commons-tokenizer';
+
+function softTokenizer(string) {
+	return [prepareStringIndexation(string, '-')];
+}
+
+export default softTokenizer;

package/src/utils/suggester-workers/create-worker.js

@@ -0,0 +1,55 @@
+// Enable cross domain workers
+
+const testSameOrigin = (url) => {
+	var loc = window.location;
+	var a = document.createElement('a');
+	a.href = url;
+	return (
+		a.hostname === loc.hostname &&
+		a.port === loc.port &&
+		a.protocol === loc.protocol
+	);
+};
+
+const createWorkerFallback = (workerUrl) => {
+	console.log('Create worker for MFE');
+	var worker = null;
+	try {
+		var blob;
+		try {
+			blob = new Blob([`importScripts('${workerUrl}');`], {
+				type: 'application/javascript',
+			});
+		} catch (e) {
+			var blobBuilder = new (window.BlobBuilder ||
+				window.WebKitBlobBuilder ||
+				window.MozBlobBuilder)();
+			blobBuilder.append(`importScripts('${workerUrl}');`);
+			blob = blobBuilder.getBlob('application/javascript');
+		}
+		var url = window.URL || window.webkitURL;
+		var blobUrl = url.createObjectURL(blob);
+		worker = new Worker(blobUrl);
+	} catch (e1) {
+		console.error(`Lunatic-worker : Failed to load web worker : ${workerUrl}`);
+	}
+	return worker;
+};
+
+export const createWorker = (workerUrl) => {
+	var worker = null;
+	try {
+		if (testSameOrigin(workerUrl)) {
+			worker = new Worker(workerUrl);
+			worker.onerror = (event) => {
+				event.preventDefault();
+				worker = createWorkerFallback(workerUrl);
+			};
+		} else {
+			worker = createWorkerFallback(workerUrl);
+		}
+	} catch (e) {
+		worker = createWorkerFallback(workerUrl);
+	}
+	return worker;
+};

package/src/utils/suggester-workers/find-best-label/find-best-label.js

@@ -0,0 +1,39 @@
+import distance from 'damerau-levenshtein';
+
+function reduceFields(map, fields, similarity) {
+	return fields.reduce(function (m, field) {
+		if (field in m) {
+			return { ...m, [field]: m[field] + similarity };
+		}
+		return { ...m, [field]: similarity };
+	}, map);
+}
+
+function reduceTokensMap(map, entries, word) {
+	return entries.reduce(function (inter, [token, { fields }]) {
+		const { similarity } = distance(token, word);
+		return reduceFields(inter, fields, similarity);
+	}, map);
+}
+
+function findBest(map) {
+	let best = -1;
+	return Object.entries(map).reduce(function (curr, [field, score]) {
+		if (score > best) {
+			best = score;
+			return field;
+		}
+		return curr;
+	}, undefined);
+}
+
+function findBestLabel(searchTokens = [], tokensMap = {}) {
+	const entries = Object.entries(tokensMap);
+	const map = searchTokens.reduce(function (map, word) {
+		return reduceTokensMap(map, entries, word);
+	}, {});
+	const best = findBest(map);
+	return best;
+}
+
+export default findBestLabel;

package/src/utils/suggester-workers/find-best-label/find-best-label.worker.js

@@ -0,0 +1,40 @@
+/* eslint-disable no-restricted-globals */
+import 'core-js/stable';
+import 'regenerator-runtime/runtime';
+import tokenize from './tokenize';
+import findBestLabel from './find-best-label';
+
+const TASKS = [];
+let RUN = false;
+
+function executeTask(task) {
+	const { option, search, idTask } = task;
+	const { tokensMap } = option;
+	const { search: st } = tokenize({ search });
+	const label = findBestLabel(st, tokensMap);
+
+	self.postMessage({ response: label, idTask });
+}
+
+function appendTask(task) {
+	TASKS.push(task);
+	if (!RUN) {
+		activate();
+	}
+}
+
+function activate() {
+	RUN = true;
+	while (TASKS.length) {
+		const task = TASKS.pop();
+		executeTask(task);
+	}
+	RUN = false;
+}
+
+self.onmessage = function (e) {
+	const { option, search, idTask } = e.data;
+	if (option && search && idTask) {
+		appendTask({ option, search, idTask });
+	}
+};

package/src/utils/suggester-workers/find-best-label/tokenize.js

@@ -0,0 +1,30 @@
+import tokenizer from 'string-tokenizer';
+
+function toArray(tokens) {
+	if (Array.isArray(tokens)) {
+		return tokens;
+	}
+	return [tokens];
+}
+
+function tokenize(object, pattern = /[\w]+/) {
+	if (typeof object === 'object') {
+		const map = {};
+		Object.entries(object).forEach(function ([key, string]) {
+			if (string.length) {
+				const { tokens } = tokenizer()
+					.input(string)
+					.tokens({ tokens: pattern })
+					.resolve();
+				map[key] = toArray(tokens);
+			} else {
+				map[key] = [];
+			}
+		});
+
+		return map;
+	}
+	return {};
+}
+
+export default tokenize;