Skip to content

Generate EPUB books from HTML with a simple API in Node.js and the browser.

License

Notifications You must be signed in to change notification settings

cpiber/epub-gen-memory

 
 

Repository files navigation

epub-gen-memory --- a library to make EPUBs from HTML

Generate EPUB books from HTML with a simple API in Node.js or the browser.

See JSZip Support table for which engines are supported. In addition, browsers need to support fetch as well as DOMParser and XMLSerializer to replace node-fetch and htmlparser2 (plus ecosystem).


This EPUB library will generate the needed files, as well as download all referenced images. Note that all fonts and all images respectively are downloaded in parallel in batches, so please keep that in mind when adjusting fetchTimeout.

Note that in the browser, only images on servers with CORS enabled can be downloaded.

On the server (with Node.js), image paths can also start with file://, in which case they are read directly from disc. The part after file:// must be a valid path for your system, as read by fs.readFile.

Usage

Install the lib and add it as a dependency (recommended), run in your project directory:

npm install epub-gen-memory --save

Then put this in your code:

import epub from 'epub-gen-memory';

epub(options).then(
    content => console.log("Ebook Generated Successfully!"),
    err => console.error("Failed to generate Ebook because of ", err)
);

See JSZip documentation on how to get the zip to the user. For a nodejs example please see the tests.

In environments where SharedArrayBuffer is not available, you might want to instead import from epub-gen-memory/sabstub, which includes a non-functional stub:

import epub from 'epub-gen-memory/sabstub';

The package also includes a browserifyd bundle (UMD) as epub-gen-memory/bundle. It is possible to use the bundle if you want to build for the browser. The bundle is also available from a CDN: UNPKG (latest, latest 1.x). The bundle also includes the proper return type for the browser (Promise<Blob> instead of Promise<Buffer>).

import epub from 'epub-gen-memory/bundle';

Note: This library was written in TypeScript and thus uses ESM exports, but it was compiled to CommonJS, so you can also use the following:

const epub = require('epub-gen-memory').default;

API

import epub, { EPub, optionsDefaults, chapterDefaults } from 'epub-gen-memory';
import type { Options, Content, Chapter, Font } from 'epub-gen-memory';

epub(optionsOrTitle, content, [version | verbose][])

  • optionsOrTitle: Options | string if string, then equivalent to { title: <optionsOrTitle> } (see below)
  • content: Chapter[] (see below)
  • varargs: version: 3 | 2, verbose: boolean (in any order or not at all)
  • Returns: Promise<Buffer>
  • Browser Returns: Promise<Blob>

class EPub

  • contructor(options: Options, content: Chapter[])
  • render(): Promise<EPub>
  • genEpub(): Promise<Buffer> (Browser Promise<Blob>)
  • generateAsync(options: JSZipGeneratorOptions): Promise<...> see JSZip.generateAsync

protected:

  • generateTemplateFiles(): Promise<void>
  • downloadAllFonts(): Promise<void>
  • downloadAllImages(): Promise<void>
  • makeCover(): Promise<void>

optionsDefaults([version])

  • version: number (default 3) Epub version
  • Returns: Options defaults

chapterDefaults(index)

  • index: number of the chapter
  • Returns: Chapter defaults

 

Options

  • title: string
    Title of the book
  • author: string | string[] (optional, default ['anonymous'])
    Name of the author for the book, e.g. "Alice" or ["Alice", "Bob"]
  • publisher: string (optional, default anonymous)
    Publisher name
  • description: string (optional)
    Book description
  • cover: string | File (optional)
    Book cover image URL, e.g. "http://abc.com/book-cover.jpg" or File object containing image bytes
  • tocTitle: string (optional, default Table of Contents)
    Title of the Table of Contents
  • tocInTOC: boolean (optional, default true)
    Whether to show the TOC entry in the reader's Table of Contents; only for EPUB2
  • numberChaptersInTOC: boolean (optional, default true)
    Automatically number entries in TOC
  • prependChapterTitles: boolean (optional, default true)
    Automatically put the title of the chapter before the content
  • date: string (optional, default today)
    Publication date
  • lang: string (optional, default en)
    Language code of the book
  • css: string (optional)
    CSS string, replaces our default styles, e.g: "body{background: #000}"
  • fonts: Font[] (optional)
    Array of fonts to include, see below
  • version: number (optional, default 3)
    Version of the generated EPUB, 3 for the latest version (http://idpf.org/epub/30) or 2 for the previous version (http://idpf.org/epub/201)
  • fetchTimeout: number (optional, default 20000)
    Timeout time for requests, in milliseconds; Browsers need to support AbortController and signals for this to work
  • retryTimes: number (optional, default 3)
    How many times to retry fetching resources
  • batchSize: number (optional, default 100)
    The size of the batches to use when downloading files
  • ignoreFailedDownloads: boolean (optional, default false)
    Instead of throwing, emit a warning and write an empty file if a font or image fails to download
  • verbose: boolean | ((type, ...args) => void) (optional, default false)
    Whether to log progress messages; If a function is provided, the first argument will either be 'log' or 'warn'

Chapters

Within each chapter object:

  • title: string (optional, default Chapter [number])
    Chapter title
  • author: string | string[] (optional)
    Chapter author, generates info below chapter title
  • content: string
    HTML String of the chapter content, image sources are downloaded
  • excludeFromToc: boolean (optional, default false)
    Don't list chapter in Table of Contents
  • beforeToc: boolean (optional, default false)
    List chapter before Table of Contents heading
  • filename: string (optional)
    Custom name for chapter file
  • url: string (optional)
    External link below chapter title

Fonts

Within each font object:

  • filename: string
    Name under which the font should be downloaded, including file extension
  • url: string
    URL where to find font, for best compatibility use ttf (TrueType) fonts

You can then use the fonts as such (assuming you have a font with filename Merriweather.ttf):

@font-face {
    font-family: "Merriweather";
    font-style: normal;
    font-weight: normal;
    src : url(./fonts/Merriweather.ttf);
}

Changelog

Only includes major changes

1.1.2

  • Fixed file names
    Internal chapter files are no longer given invalid names if the chapter title includes non-alphanumeric characters

1.1.0

  • Optionally disable input validation
    In the browser, by default, validation is not bundled due to size. To force use/not use, see the files lib/util/predicates{-nop}.ts.
  • Support File for cover
    Useful in the browser to allow user-uploaded files without temporary storage

Demo Preview:

Demo Preview

From Lewis Carroll "Alice's Adventures in Wonderland", based on text at https://www.cs.cmu.edu/~rgs/alice-table.html and images from http://www.alice-in-wonderland.net/resources/pictures/alices-adventures-in-wonderland.

Please see the tests for the code used. EPUBs are generated next to the built test files.

Credits

This library is based on the work of @cyrilis.

Packages

No packages published

Languages

  • TypeScript 78.1%
  • EJS 15.4%
  • JavaScript 3.4%
  • HTML 2.6%
  • CSS 0.5%