ffn-parser v0.0.8

FanFiction.net Parser

Since fanfiction.net does not have an API, information about stories hosted there can only be retrieved by scraping the webpages themselves. Unfortunately, bits of information are sometimes formatted differently or are just thrown together, making it very hard to extract the needed information.

This library aims to provide a set of functions to parse the webpages and extract the information in a machine-readable format.

⚠ This library follows semantic versioning. It is not stable yet, meaning that newer versions may break backwards-compatibility at any time.

Installation

Install this library via npm or yarn.

npm install ffn-parser

Usage

This library does not download any pages from fanfiction.net. You must download these pages yourself using any library of your choice. Please note that fanfiction.net is rate limited. Do not download too many pages at once and wait the indicated amount of time if you hit the limit.

import { parseStory } from "ffn-parser";

// the id of the story you want to analyze
const storyId = 1;

const response = await fetch(`https://www.fanfiction.net/s/${storyId}`);
const html = await response.text();
const template = document.createElement("template");
template.innerHTML = html;

// info contains all the extracted information
const info = await parseStory(template.content);

Since this library needs to analyze the DOM of the page, a way to create DOM elements is needed. You might need to install a tool like JSDOM if you want to use this library on NodeJS.

For now, three functions are provided.

• parseStory, which can be used on story pages like in the example above to extract information about a story.
• parseFollows, which can be used on the alerts and favorites list in the user settings.
• parseStoryList, which can be used on lists of stories such as when searching for crossovers or exploring the stories in a community.

The latter two functions do not provide as much information about a story as parseStory, since the HTML pages they parse do not contain any more information.

This library also provides full Typescript support. The returned types are the best place to inspect what information the functions extract - or just execute them and see what happens.