natural-content - npm Package Compare versions

Comparing version 1.0.15 to 1.1.0

index.js

@@ -81,8 +81,8 @@ const { PorterStemmerFr, TfIdf, AggressiveTokenizerFr } = require('natural');
 /**
- * removeLineBreaks - Remove line breaks in a text
+ * removeLineBreaks - Remove line breaks & tabs in a text
  *
  * @param  {type} text the text
- * @returns {type}     the text without line breaks
+ * @returns {type}     the text without line breaks and tabs
  */
-function removeLineBreaks(text) {
+function removeLineBreakTabs(text) {
   if (!text) {
@@ -92,3 +92,7 @@ return '';
 
-  return text.replace(/(\r\n|\n|\r)/gm, '').trim();
+  return text.replace(/[\t]/g, WORD_SEPARATOR) // Remove Tabs
+             .replace(/[\n\r]/g, WORD_SEPARATOR)
+             .replace(/[\n]/g, WORD_SEPARATOR)
+             .replace(/\s+/g, WORD_SEPARATOR)
+             .trim();
 }
@@ -182,2 +186,3 @@
  * @param  {number} nbrKeywords The number of keywords to return
+ * @param  {number} language The language to use for finding the stopwords
  * @returns {Arrays}             The list of keywords
@@ -219,3 +224,3 @@ */
 
-exports.removeLineBreaks = removeLineBreaks;
+exports.removeLineBreakTabs = removeLineBreakTabs;
 
@@ -222,0 +227,0 @@ exports.getWords = getWords;

package.json

 {
   "name": "natural-content",
-  "version": "1.0.15",
-  "description": "A set of natural functions like tf.idf, extract words & n-grams, remove diacritics, ... (experimental project)",
+  "version": "1.1.0",
+  "description": "A set of natural functions like extracting words & n-grams, remove diacritics, get top keywords, ... (experimental project)",
   "main": "index.js",
@@ -6,0 +6,0 @@ "scripts": {

test/test.js

@@ -42,2 +42,8 @@ const assert = require('assert');
 
+  it('Line Break, tabs and spaces', () => {
+    const text = 'Ceci est un texte en français ! \t sans caractères spéciaux ! \n L\'autre        texte\n\r';
+    const result = natural.removeLineBreakTabs(text);
+
+    assert(result === 'Ceci est un texte en français ! sans caractères spéciaux ! L\'autre texte');
+  });
   it('apostrophe', () => {
@@ -44,0 +50,0 @@ const text = 'ceci est un texte en français. l\'été sera chaud. Les conditions d\'utilisation de l\'objet';

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

30312

increased by2.05%

Lines of code

1240

increased by0.81%

No dependency changes