Adding the statcan & lein phonetic algorithms

CHANGELOG.md

@@ -3,6 +3,8 @@
 ## 0.7.0
 
 * Adding the `features/extraction/vectorizer` namespace.
+* Adding the `phonetics/lein` namespace.
+* Adding the `phonetics/statcan` namespace.
 
 ## 0.6.0
 

src/phonetics/lein.js

@@ -0,0 +1,61 @@
+/**
+ * Talisman phonetics/lein
+ * ========================
+ *
+ * The Lein name coding procedure.
+ *
+ * [Reference]:
+ * http://naldc.nal.usda.gov/download/27833/PDF
+ */
+import deburr from 'lodash/deburr';
+import {squeeze, translation} from '../helpers';
+
+/**
+ * Constants.
+ */
+const DROPPED = /[AEIOUYWH]/g;
+
+const TRANSLATION = translation('DTMNLRBFPVCJKGQSXZ', '112233444455555555');
+
+/**
+ * Helpers.
+ */
+function pad(code) {
+  return (code + '0000').slice(0, 4);
+}
+
+/**
+ * Function taking a single name and computing its lein code.
+ *
+ * @param  {string}  name - The name to process.
+ * @return {string}       - The lein code.
+ *
+ * @throws {Error} The function expects the name to be a string.
+ */
+export default function lein(name) {
+  if (typeof name !== 'string')
+    throw Error('talisman/phonetics/lein: the given name is not a string.');
+
+  let code = deburr(name)
+    .toUpperCase()
+    .replace(/[^A-Z\s]/g, '');
+
+  // 1-- Keeping the first letter
+  const first = code[0];
+  code = code.slice(1);
+
+  // 2-- Dropping vowels and Y, W & H
+  code = code.replace(DROPPED, '');
+
+  // 3-- Dropping consecutive duplicates and truncating to 4 characters
+  code = squeeze(code).slice(0, 4);
+
+  // 4-- Translations
+  const backup = code;
+  code = '';
+
+  for (let i = 0, l = backup.length; i < l; i++)
+    code += TRANSLATION[backup[i]] || backup[i];
+
+  return pad(first + code);
+}

src/phonetics/statcan.js

@@ -0,0 +1,50 @@
+/**
+ * Talisman phonetics/statcan
+ * ===========================
+ *
+ * The statistics Canada name coding technique.
+ *
+ * [Reference]:
+ * http://naldc.nal.usda.gov/download/27833/PDF
+ */
+import deburr from 'lodash/deburr';
+import {squeeze} from '../helpers';
+
+/**
+ * Constants.
+ */
+const DROPPED = /[AEIOUY]/g;
+
+/**
+ * Function taking a single name and computing its statcan code.
+ *
+ * @param  {string}  name - The name to process.
+ * @return {string}       - The statcan code.
+ *
+ * @throws {Error} The function expects the name to be a string.
+ */
+export default function statcan(name) {
+
+  if (typeof name !== 'string')
+    throw Error('talisman/phonetics/statcan: the given name is not a string.');
+
+  let code = deburr(name)
+    .toUpperCase()
+    .replace(/[^A-Z\s]/g, '');
+
+  // 1-- Keeping the first letter
+  const first = code[0];
+  code = code.slice(1);
+
+  // 2-- Dropping vowels and Y
+  code = code.replace(DROPPED, '');
+
+  // 3-- Dropping consecutive duplicates
+  code = squeeze(code);
+
+  // 4-- Dropping blanks
+  code = code.replace(/\s/g, '');
+
+  // 5-- Limiting code size to 4
+  return (first + code).slice(0, 4);
+}

test/endpoint.js

@@ -46,10 +46,12 @@ describe('phonetics', function() {
   require('./phonetics/caverphone.js');
   require('./phonetics/daitch-mokotoff.js');
   require('./phonetics/double-metaphone.js');
+  require('./phonetics/lein.js');
   require('./phonetics/metaphone.js');
   require('./phonetics/mra.js');
   require('./phonetics/nysiis.js');
   require('./phonetics/soundex.js');
+  require('./phonetics/statcan.js');
 
   describe('french', function() {
     require('./phonetics/french/phonetic.js');

test/phonetics/lein.js

@@ -0,0 +1,65 @@
+/**
+ * Talisman phonetics/lein tests
+ * ==============================
+ *
+ */
+import assert from 'assert';
+import lein from '../../src/phonetics/lein';
+
+describe('lein', function() {
+
+  it('should throw if the given word is not a string.', function() {
+    assert.throws(function() {
+      lein([]);
+    }, /string/);
+  });
+
+  it('should compute the lein code correctly.', function() {
+    const tests = [
+      ['Guillaume', 'G320'],
+      ['Dabbs', 'D450'],
+      ['Daves', 'D450'],
+      ['Davies', 'D450'],
+      ['Davis', 'D450'],
+      ['Debaca', 'D450'],
+      ['Debose', 'D450'],
+      ['Debus', 'D450'],
+      ['Defazio', 'D450'],
+      ['Defigh', 'D450'],
+      ['Deveaux', 'D450'],
+      ['Devese', 'D450'],
+      ['Devies', 'D450'],
+      ['Devos', 'D450'],
+      ['Dipiazza', 'D450'],
+      ['Divish', 'D450'],
+      ['Dobak', 'D450'],
+      ['Dobbs', 'D450'],
+      ['Dobis', 'D450'],
+      ['Dobish', 'D450'],
+      ['Dobosh', 'D450'],
+      ['Doepke', 'D450'],
+      ['Dopps', 'D450'],
+      ['Doubek', 'D450'],
+      ['Doviak', 'D450'],
+      ['Dubbs', 'D450'],
+      ['Dubke', 'D450'],
+      ['Dubois', 'D450'],
+      ['Duboise', 'D450'],
+      ['Dubose', 'D450'],
+      ['Dubs', 'D450'],
+      ['Dubukey', 'D450'],
+      ['Dubus', 'D450'],
+      ['Dufek', 'D450'],
+      ['Duffek', 'D450'],
+      ['Dupas', 'D450'],
+      ['Dupois', 'D450'],
+      ['Dupuis', 'D450'],
+      ['Arlène', 'A332'],
+      ['Lüdenscheidt', 'L125']
+    ];
+
+    tests.forEach(function([word, code]) {
+      assert.strictEqual(lein(word), code, `${word} => ${code}`);
+    });
+  });
+});

test/phonetics/statcan.js

@@ -0,0 +1,36 @@
+/**
+ * Talisman phonetics/statcan tests
+ * =================================
+ *
+ */
+import assert from 'assert';
+import statcan from '../../src/phonetics/statcan';
+
+describe('statcan', function() {
+
+  it('should throw if the given word is not a string.', function() {
+    assert.throws(function() {
+      statcan([]);
+    }, /string/);
+  });
+
+  it('should compute the statcan code correctly.', function() {
+    const tests = [
+      ['Guillaume', 'GLM'],
+      ['Daves', 'DVS'],
+      ['Davies', 'DVS'],
+      ['Davis', 'DVS'],
+      ['Devese', 'DVS'],
+      ['Devies', 'DVS'],
+      ['Devos', 'DVS'],
+      ['Dove', 'DV'],
+      ['Divish', 'DVSH'],
+      ['Arlène', 'ARLN'],
+      ['Lüdenscheidt', 'LDNS']
+    ];
+
+    tests.forEach(function([word, code]) {
+      assert.strictEqual(statcan(word), code, `${word} => ${code}`);
+    });
+  });
+});