projects

Stephen King
REST API

Development

API Design

A RESTful API built from the collected works of the horror master. I thought it strange that there was no pragmatically accessible data for one of the greatest living authors, so I built one. His loosely connected works fit perfectly into a relational database schema.

view site

Typescript

Typed superset of Javascript

Postgres

Relational Database and management system

Prisma

Open source Node and TS ORM

Javascript

For interaction design testing

The Goal

With 65 novels/novellas, over 200 short stories, and countless memorable characters I wanted a way to access this information programatically. Moreover, many of his works are loosely connected in a 'King Universe', with different characters and locations popping up accross multiple fictions. As a fan, those relationships have always facinated me and as developer they seemed to fit perfectly in a strucutred database.

The Data Challenge

What, Where, How

Lets start with the easiest question - How am I going to get this data? Scraping, both with Python's BeautifulSoup and Javascript's Cheerio.

What type of data should I include? obviously I needed basic bibliophic info about the works like publish date, publisher, page count etc.. But I also wanted character data - specifically books they were in. On top of that I needed a common data point shared between entities the so that I could make those relationships programmically.

Where am I going to source my data? I initially turned to Wikipedia, which provides a comprehensive list of Stephen King's entire body of work, including bibliographic data in a scraper-friendly format. However, I encountered challenges when attempting to extract character information due to inconsistent formatting across individual Wikipedia pages for King's works. Fortunately, I found an alternative source in the form of dedicated fan wikis related to Stephen King's work. Specifically, I leveraged the information available on the Fandom website, which offered a wealth of character data relevant to my project. This alternative source allowed me to successfully scrape and integrate character information into the API.