jzombie / ts-lorem-ipsum-word-frequencies

Prototype Streaming Interface Word Counter

Building / Running

$ npm install
$ npm run build
$ npm start

Request / Response

Example request: [POST] http://localhost:3000/word-count?n=3 (where n is the number of top-most frequent words; be sure to upload a UTF-8 encoded text file)

Example response:

{
    "frequencies": [
        {
            "word": "sed",
            "count": 12
        },
        {
            "word": "id",
            "count": 10
        },
        {
            "word": "sit",
            "count": 8
        }
    ]
}

Testing

Testing with code coverage.

$ npm install
$ npm test

Considerations and Additional Improvements

I wanted to use as little RAM as possible when processing larger files, so opted for parsing off of streaming input.

Unit tests could have had additional mock data and / or modification of mock data to perform additional test scenarios.

Didn't dive into potential error handing with using incorrect encoding types, non-integer / invalid range "n" values, etc.