Skip to content


Folders and files

Last commit message
Last commit date

Latest commit



93 Commits

Repository files navigation


Given an epub file, do our best to count the number of words in it.


This package is available on npm as epub-wordcount. It can be installed using common JS tools:

npm i -g epub-wordcount

Basic Usage

On the CLI:

% word-count path/to/book.epub

The Strange Case of Dr. Jekyll and Mr. Hyde
  * 26,341 words

In code:

// TS:
import { countWords } from 'epub-wordcount'

// JS:
// const { countWords } = require('epub-wordcount')

countWords('./books/some-book.epub').then((numWords) => {
  console.log(`There are ${numWords} words`)
// There are 106190 words


There's also a cli tool to quickly get the count of any epub file! Invoke it via:

word-count path/to/file.epub


word-count directory/of/books

See word-count -h for more info


  • -c, --chars - Print the character count instead of the world count
  • -r, --raw - Instead of printing the nice title, just print out a numeral
  • -t, --text - Print out the whole text of the book. Great for passing into other unix functions, like wc.
  • --ignore-drm - If the function is saying your file has DRM when you know it doesn't, you can pass this flag to force the CLI to ignore the DRM warning. Might cause weird results if the actually does have DRM.

Code API

There are a number of functions exported from this package. Each one takes either a path to a file or an already-parsed file. Mostly you'll use the path, but if the epub you're parsing is in a non-standard format, then you might use that function to ensure the file parses correctly. See here for the options available.

  • countWords(pathOrEpub, ignoreDrm?) => Promise<number>
  • countCharacters(pathOrEpub, ignoreDrm?) => Promise<number>
  • getText(pathOrEpub, ignoreDrm?) => Promise<string[]>

Each of the above can be passed the result of the following:

  • parseEpubAtPath(path, ignoreDrm?) => Promise<EPub>


There's no programmatic representation for the table of contents in epub and it's hard to skip over the reviews, copyright, etc. An effort is made to only parse the actual story text, but there's a margin of error. Probably no more than ~500 words.

Pull requests welcome.

Test Books

Unit tests are run on the following e-books:

  • Stevenson's THE STRANGE CASE OF DR. JEKYLL AND MR. HYDE, from the public domain, provided by Standard Ebooks
  • A copy of The Martian, by Andy Weir. This is my copy, purchased from Apple Books. It's DRM encumbered, so releasing it publicly should not be seen as copyright infringement.

Fake ePubs

In modern versions of macOS, dragging a book out of the Books app won't give you an actual epub- it'll give you a folder with the .epub extension. Unsurprisingly, this doesn't play well with ePub tooling.

To fix, run the following command (pulled from here) fixes them for me:

# from inside the rogue directory
zip -X -r ../fixed.epub mimetype *