feature: implement stemming - fix #9

CONTRIBUTING.md

@@ -3,7 +3,8 @@
 1. Fork this project and install the packages with `npm install`
 - Create a new feature/patch branch
 - write test in `test` folder
-- Code code code and amend `example.js` if needed
+- Code code code
+- amend files `example.js` and `readme.md` if needed
 - Run `npm test` to check all linting and tests are passing or setup a git pre-commit hook to run `npm test`
 - Write a [good commit message](https://github.com/angular/angular.js/blob/master/CONTRIBUTING.md#commit) in the format `<type>(<scope>): <subject>`
 - Pull request using the new feature/patch branch

example.js

@@ -57,3 +57,14 @@ console.log(overlap(sentence1, sentence2, {
   ignorePlurals: [ 'hitchhikings' ],
   ignoreCommonWords: true
 }));
+
+console.log(colors.black('\nWith options stemming:'));
+console.log('var sentence1 = \'A programming course in SmallTalk\';');
+console.log('var sentence2 = \'Have you programmed in SmallTalk?\';');
+console.log('overlap(sentence1, sentence2, { stemming: true, ignoreCommonWords: true })');
+sentence1 = 'A programming course in SmallTalk';
+sentence2 = 'Have you programmed in SmallTalk?';
+console.log(overlap(sentence1, sentence2, {
+  stemming: true,
+  ignoreCommonWords: true
+}));

index.js

@@ -12,6 +12,7 @@ module.exports = function(phrase1, phrase2, options) {
   options.common = options.common || null;
   options.depluralize = options.depluralize || false;
   options.ignorePlurals = options.ignorePlurals || [];
+  options.stemming = options.stemming || false;
 
   if (options.ignoreCase) {
     phrase1 = phrase1.toLowerCase();
@@ -38,6 +39,11 @@ module.exports = function(phrase1, phrase2, options) {
     compare2 = lib.removeCommonWords(compare2, options.common);
   }
 
+  if (options.stemming) {
+    compare1 = lib.stem(compare1);
+    compare2 = lib.stem(compare2);
+  }
+
   compare1.forEach(function(element) {
     if (compare2.indexOf(element) > -1) {
       answer.push(element);

lib/lib.js

@@ -1,6 +1,7 @@
 'use strict';
 
-var pluralize = require('pluralize');
+var pluralize = require('pluralize'),
+  natural = require('natural');
 
 exports.sanitize = function sanitize(word) {
   // matches common punctuations:
@@ -34,3 +35,9 @@ exports.depluralize = function depluralize(words, ignorePlurals) {
     return pluralize.singular(element);
   })
 }
+
+exports.stem = function stem(words) {
+  return words.map(function(element) {
+    return natural.PorterStemmer.stem(element);
+  })
+}

package.json

@@ -41,6 +41,7 @@
     }
   },
   "dependencies": {
+    "natural": "^0.1.28",
     "pluralize": "~1.0.2"
   }
 }

readme.md

@@ -99,6 +99,19 @@ overlap(sentence1, sentence2, {
 // [ 'hitchhikings', 'meetup' ]
 ```
 
+###option: stemming
+
+```js
+var sentence1 = 'A programming course in SmallTalk';
+var sentence2 = 'Have you programmed in SmallTalk?';
+
+overlap(sentence1, sentence2, {
+  stemming: true,
+  ignoreCommonWords: true
+});
+// [ 'program', 'smalltalk' ]
+```
+
 Try out the examples in file `example.js` with the command `node example.js`
 
 ##Contribute

test/indexSpec.js

@@ -167,8 +167,8 @@
           var sentence1 = 'Base and base',
             sentence2 = 'Base in bases',
             overlapList = overlap(sentence1, sentence2, {
-                depluralize: true,
-                common: [ 'base' ]
+              depluralize: true,
+              common: [ 'base' ]
             });
 
           expect(overlapList).to.include.members([ 'Base' ]);
@@ -182,6 +182,24 @@
             overlapList = overlap(sentence1, sentence2);
 
           expect(overlapList).to.have.length(0);
+          done();
+        });
+      });
+
+      describe('With option stemming', function() {
+        it('returns the word stem or root word', function(done) {
+          var sentence1 = 'A programming course in SmallTalk',
+            sentence2 = 'Have you programmed in SmallTalk?',
+            overlapList = overlap(sentence1, sentence2, {
+              stemming: true,
+              ignoreCommonWords: true
+            });
+
+          expect(overlapList).to.include.members([
+            'program',
+            'smalltalk'
+          ]);
+
           done();
         });
       })

test/lib/libSpec.js

@@ -119,6 +119,29 @@
       });
     });
 
+    describe('Stem words', function() {
+
+      it('returns an array with stem words', function(done) {
+        var words = [
+            'programming',
+            'swimming',
+            'eating',
+            'traveled'
+          ],
+          answer = lib.stem(words);
+
+        expect(answer).to.be.an('array');
+        expect(answer).to.have.members([
+          'program',
+          'swim',
+          'eat',
+          'travel'
+        ]);
+
+        done();
+      });
+    });
+
   });
 
 })();