swearjar-extended2

This is a modified version of swearjar-extended

Profanity detection and filtering library.

Installation

npm install --save swearjar-extended2

Usage

swearjar.setLang(text)

Sets a language to load a dictionary of words to be used as filter.

| Language | Code | | -------------------------------------------------------- | ----------------- | | English | en | | Filipino | ph | | Spanish | es | | Bahasa | id |

NOTE: A US English default list located in the config directory is included and loaded by default.

var swearjar = require('swearjar-extended2');
swearjar.setLang("en");

A dictionary is just a plain JSON file containing an object where its keys are the words to check for and the values are arrays of categories where the words fall in.

{
  "regex": {
    "\\w*fuck\\w*": [
      "category1",
      "category2"
    ],
    "word2": [
      "category1"
    ],
    "word3": [
      "category2"
    ]
  },
  "simple": {
    "word1": [
      "category1",
      "category2"
    ],
    "word2": [
      "category1"
    ],
    "word3": [
      "category2"
    ]
  },
  "emoji": {
    "1f4a9": [
      "category1",
      "category2"
    ],
    "word2": [
      "category1"
    ],
    "word3": [
      "category2"
    ]
  }
}

swearjar.profane(text)

Returns true if the given string contains profanity.

var swearjar = require('swearjar-extended2');
swearjar.profane("hello there"); // false
swearjar.profane("fuck you john doe"); // true

swearjar.censor(text)

Replaces profanity with asterisks.

var clean = swearjar.censor("fuck you john doe bitch"); // **** you john doe *****

swearjar.words(text)

Get the words alongside there categories.

swearjar.words('fuck you john doe'); // { fuck: ['sexual'] }

swearjar.detailedProfane(text)

Get the words alongside there categories, count and censor the text.

swearjar.detailedProfane('fuck you john doe')

returns:

{
  categoryCount: {
    sexual: 1
  },
  censored: '**** you john doe',
  profane: true,
  wordCount: {
    fuck: 1
  },
  words: {
    fuck: [
      'sexual'
    ]
  }
}

swearjar.scorecard(text)

Generates a report from the given text.

swearjar.scorecard("fuck you john doe bitch fuck"); // {sexual: 2, insult: 1}

swearjar.addRegex(text)

Add a regex.

swearjar.addRegex('addedword?\\b', ['detected']);

swearjar.addSimple(text)

Add a simple word.

swearjar.addSimple('addedword', ['detected']);

swearjar.addEmoji(text)

Add an emoji word.

swearjar.addEmoji('1f596', ['detected']);

Acknowledgements

swearjar-node is based on Swearjar (Ruby) and Swearjar PHP.

Sources

• filipino-badwords-list
• Shutterstock
• NLP_bahasa_resources