@inseefr/lunatic 0.2.2-experimental → 0.2.3-prisme

package/src/utils/suggester-workers/append-to-index/append.js

@@ -1,25 +1,25 @@
-import { CONSTANTES } from '../../store-tools';
-import { openDb, idbBulkInsert } from '../../idb-tools';
-import MESSAGES from './store-messages';
-import prepareEntities from './prepare-entities';
-
-async function append(storeInfo, version, entities, log = () => null) {
-	try {
-		const { name, stopWords, fields } = storeInfo;
-		const prepared = prepareEntities(entities, { fields, stopWords }, log);
-		const db = await openDb(name, version);
-		log({ message: MESSAGES.startInsertBatch });
-		await idbBulkInsert(db, CONSTANTES.STORE_DATA_NAME, function (args) {
-			const { message } = args;
-			log({ message });
-		})(prepared);
-		log({ message: MESSAGES.insertBatchDone });
-		log({ message: MESSAGES.done });
-		return 'success';
-	} catch (e) {
-		log({ message: 'Errors occurred when trying to append data.' });
-		console.error(e);
-	}
-}
-
-export default append;
+import { CONSTANTES } from '../../store-tools';
+import { openDb, idbBulkInsert } from '../../idb-tools';
+import MESSAGES from './store-messages';
+import prepareEntities from './prepare-entities';
+
+async function append(storeInfo, version, entities, log = () => null) {
+	try {
+		const { name, stopWords, fields } = storeInfo;
+		const prepared = prepareEntities(entities, { fields, stopWords }, log);
+		const db = await openDb(name, version);
+		log({ message: MESSAGES.startInsertBatch });
+		await idbBulkInsert(db, CONSTANTES.STORE_DATA_NAME, function (args) {
+			const { message } = args;
+			log({ message });
+		})(prepared);
+		log({ message: MESSAGES.insertBatchDone });
+		log({ message: MESSAGES.done });
+		return 'success';
+	} catch (e) {
+		log({ message: 'Errors occurred when trying to append data.' });
+		console.error(e);
+	}
+}
+
+export default append;

package/src/utils/suggester-workers/append-to-index/append.worker.js

@@ -1,16 +1,16 @@
-/* eslint-disable no-restricted-globals */
-import 'core-js/stable';
-import 'regenerator-runtime/runtime';
-import append from './append';
-
-self.onmessage = function (e) {
-	function log(message) {
-		self.postMessage(message);
-	}
-	const { name, version, stopWords, fields, entities } = e.data;
-	append({ name, version, stopWords, fields }, version, entities, log).then(
-		function () {
-			self.postMessage('success');
-		}
-	);
-};
+/* eslint-disable no-restricted-globals */
+import 'core-js/stable';
+import 'regenerator-runtime/runtime';
+import append from './append';
+
+self.onmessage = function (e) {
+	function log(message) {
+		self.postMessage(message);
+	}
+	const { name, version, stopWords, fields, entities } = e.data;
+	append({ name, version, stopWords, fields }, version, entities, log).then(
+		function () {
+			self.postMessage('success');
+		}
+	);
+};

package/src/utils/suggester-workers/append-to-index/create-append-task.js

@@ -1,43 +1,45 @@
-const workerPath =
-	process.env.LUNATIC_LOADER_WORKER_PATH ||
-	process.env.REACT_APP_LUNATIC_LOADER_WORKER_PATH ||
-	'workers/lunatic-loader-worker-0.1.0.js';
-
-/**
- * Only with Worker
- */
-function task(info, version, log = () => null) {
-	const { name, fields, stopWords } = info;
-	const worker = new Worker(workerPath);
-	let start = false;
-	let stop = false;
-
-	function launch(entities, post = () => null) {
-		return new Promise(function (resolve) {
-			start = true;
-			worker.addEventListener('message', function (e) {
-				const { data } = e;
-				if (data === 'success') {
-					if (!stop) {
-						post();
-					}
-					resolve(data);
-				} else {
-					log(data);
-				}
-			});
-			worker.postMessage({ name, version, fields, stopWords, entities });
-		});
-	}
-
-	function terminate() {
-		if (start) {
-			stop = true;
-			worker.terminate();
-		}
-	}
-
-	return [launch, terminate];
-}
-
-export default task;
+import { createWorker } from '../create-worker';
+
+const workerPath =
+	process.env.LUNATIC_LOADER_WORKER_PATH ||
+	process.env.REACT_APP_LUNATIC_LOADER_WORKER_PATH ||
+	'workers/lunatic-loader-worker-0.1.0.js';
+
+/**
+ * Only with Worker
+ */
+function task(info, version, log = () => null) {
+	const { name, fields, stopWords } = info;
+	const worker = createWorker(workerPath);
+	let start = false;
+	let stop = false;
+
+	function launch(entities, post = () => null) {
+		return new Promise(function (resolve) {
+			start = true;
+			worker.addEventListener('message', function (e) {
+				const { data } = e;
+				if (data === 'success') {
+					if (!stop) {
+						post();
+					}
+					resolve(data);
+				} else {
+					log(data);
+				}
+			});
+			worker.postMessage({ name, version, fields, stopWords, entities });
+		});
+	}
+
+	function terminate() {
+		if (start) {
+			stop = true;
+			worker.terminate();
+		}
+	}
+
+	return [launch, terminate];
+}
+
+export default task;

package/src/utils/suggester-workers/append-to-index/index.js

@@ -1,2 +1,2 @@
-export { default } from './append';
-export { default as createAppendTask } from './create-append-task';
+export { default } from './append';
+export { default as createAppendTask } from './create-append-task';

package/src/utils/suggester-workers/append-to-index/prepare-entities.js

@@ -1,61 +1,61 @@
-import { createTokenizer } from '../commons-tokenizer';
-import MESSAGES from './store-messages';
-
-const DEFAULT_BATCH_SIZE = 1000;
-
-function appendField(fields, field) {
-	if (fields.indexOf(field) !== -1) {
-		return fields;
-	}
-	return [...fields, field];
-}
-
-function createTokensMap(tokensFields) {
-	return Object.entries(tokensFields).reduce(function (map, [field, tokens]) {
-		return tokens.reduce(function (map2, token) {
-			if (token in map2) {
-				const entry = map2[token];
-				const { fields, count } = entry;
-				return {
-					...map2,
-					[token]: {
-						count: count + 1,
-						fields: appendField(fields, field),
-					},
-				};
-			}
-			return { ...map2, [token]: { count: 1, fields: [field] } };
-		}, map);
-	}, {});
-}
-
-function prepareEntities(entities, { fields, stopWords }, log) {
-	const tokenizer = createTokenizer(fields, stopWords);
-
-	let done = 0;
-	const size = DEFAULT_BATCH_SIZE;
-	const { length: max } = entities || [];
-
-	return entities.map(function (suggestion) {
-		const { id } = suggestion;
-		if (id) {
-			const tokensFields = tokenizer(suggestion);
-			const tokensMap = createTokensMap(tokensFields);
-
-			done++;
-			if (done % size === 0 || done === max) {
-				log({
-					message: {
-						...MESSAGES.indexBatch,
-						max,
-						done,
-						percent: (done / max) * 100,
-					},
-				});
-			}
-			return { id, suggestion, tokens: Object.keys(tokensMap), tokensMap };
-		} else throw new Error(`Missing id on entity.`);
-	}, []);
-}
-
-export default prepareEntities;
+import { createTokenizer } from '../commons-tokenizer';
+import MESSAGES from './store-messages';
+
+const DEFAULT_BATCH_SIZE = 1000;
+
+function appendField(fields, field) {
+	if (fields.indexOf(field) !== -1) {
+		return fields;
+	}
+	return [...fields, field];
+}
+
+function createTokensMap(tokensFields) {
+	return Object.entries(tokensFields).reduce(function (map, [field, tokens]) {
+		return tokens.reduce(function (map2, token) {
+			if (token in map2) {
+				const entry = map2[token];
+				const { fields, count } = entry;
+				return {
+					...map2,
+					[token]: {
+						count: count + 1,
+						fields: appendField(fields, field),
+					},
+				};
+			}
+			return { ...map2, [token]: { count: 1, fields: [field] } };
+		}, map);
+	}, {});
+}
+
+function prepareEntities(entities, { fields, stopWords }, log) {
+	const tokenizer = createTokenizer(fields, stopWords);
+
+	let done = 0;
+	const size = DEFAULT_BATCH_SIZE;
+	const { length: max } = entities || [];
+
+	return entities.map(function (suggestion) {
+		const { id } = suggestion;
+		if (id) {
+			const tokensFields = tokenizer(suggestion);
+			const tokensMap = createTokensMap(tokensFields);
+
+			done++;
+			if (done % size === 0 || done === max) {
+				log({
+					message: {
+						...MESSAGES.indexBatch,
+						max,
+						done,
+						percent: (done / max) * 100,
+					},
+				});
+			}
+			return { id, suggestion, tokens: Object.keys(tokensMap), tokensMap };
+		} else throw new Error(`Missing id on entity.`);
+	}, []);
+}
+
+export default prepareEntities;

package/src/utils/suggester-workers/append-to-index/store-messages.js

@@ -1,21 +1,21 @@
-import { BULK_INSERT_MESSAGES } from '../../idb-tools';
-
-const MESSAGES = {
-	...BULK_INSERT_MESSAGES,
-	startCreateIndex: { type: 'fill-store/start-create-index' },
-	indexBatch: {
-		type: 'fill-store/index-batch',
-		max: undefined,
-		done: undefined,
-		percent: undefined,
-	},
-	createIndexDone: { type: 'fill-store/create-index-done' },
-	//
-	storeClear: { type: 'fill-store/clear-store' },
-	startInsertBatch: { type: 'fill-store/start-insert-bacth' },
-	insertBatchDone: { type: 'fill-store/insert-done' },
-	done: { type: 'fill-store/done' },
-	error: { type: 'fill-store/error' },
-};
-
-export default MESSAGES;
+import { BULK_INSERT_MESSAGES } from '../../idb-tools';
+
+const MESSAGES = {
+	...BULK_INSERT_MESSAGES,
+	startCreateIndex: { type: 'fill-store/start-create-index' },
+	indexBatch: {
+		type: 'fill-store/index-batch',
+		max: undefined,
+		done: undefined,
+		percent: undefined,
+	},
+	createIndexDone: { type: 'fill-store/create-index-done' },
+	//
+	storeClear: { type: 'fill-store/clear-store' },
+	startInsertBatch: { type: 'fill-store/start-insert-bacth' },
+	insertBatchDone: { type: 'fill-store/insert-done' },
+	done: { type: 'fill-store/done' },
+	error: { type: 'fill-store/error' },
+};
+
+export default MESSAGES;

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

@@ -0,0 +1,56 @@
+import createFieldsTokenizer from './create-fields-tokenizer';
+import { composeFilters, createFilterStopWords } from './filters';
+import filterStemmer from './filters/filter-stemmer';
+import filterLength from './filters/filter-length';
+import filterSynonyms from './filters/filter-synonyms';
+import filterAccents from './filters/filter-accents';
+import filterToLower from './filters/filter-to-lower';
+import filterDouble from './filters/filter-double';
+
+function createMapFieldsTokenizer(fields, filters) {
+	return fields.reduce(function (mapFieldTokenizers, field) {
+		const { name } = field;
+		return {
+			...mapFieldTokenizers,
+			[name]: createFieldsTokenizer(field, filters),
+		};
+	}, {});
+}
+
+function createFilterTokens(fields, stopWords) {
+	const filterStopWords = createFilterStopWords(stopWords);
+	const getFilters = composeFilters(
+		filterDouble,
+		filterAccents,
+		filterStemmer,
+		filterSynonyms,
+		filterToLower,
+		filterStopWords,
+		filterLength
+	);
+	const FIELDS_TOKENIZER_MAP = createMapFieldsTokenizer(
+		fields || [],
+		getFilters
+	);
+
+	return function (field, entity) {
+		const { name } = field;
+		const tokenizeIt = FIELDS_TOKENIZER_MAP[name];
+		if (name in entity) {
+			return tokenizeIt(`${entity[name]}`);
+		}
+		return [];
+	};
+}
+
+function createEntityTokenizer(fields, stopWords) {
+	const tokenizeAll = createFilterTokens(fields, stopWords);
+	return function (entity) {
+		return fields.reduce(function (a, field) {
+			const { name } = field;
+			return { ...a, [name]: tokenizeAll(field, entity) };
+		}, {});
+	};
+}
+
+export default createEntityTokenizer;

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

@@ -1,17 +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;
+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/{filter-double.js → filters/filter-double.js}

@@ -1,12 +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;
+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/{filter-length.js → filters/filter-length.js}

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