Epubparse

⚠️ This library is developed for my own very narrow use cases and probably insufficient for your needs. It is published both as a Rust crate to crates.io and as a NPM package (ESM module) to npm.

The sole purpose of this project is essentially to convert Epub files to simple text-only Book structures, where a Book is a tree of Chapters that contain text and/or subchapters. I am just using this to do different kinds of text analysis on a per chapter basis.

At the moment only epub files with epub2 compatible toc (table of content) files (.ncx) are supported. Many epub3 files do contain epub2 toc files for compatibility reasons. I do intend to implement support for epub3 toc files (.xhtml) in the future.

Structure

• epubparse-rs: core Rust library that compiles to WASM (published to crates.io)
• epubparse-wasm: wrapper around Rust core that provides JS compatible data types from WASM functions (published to npm, only meant to be consumed by JS lib)
• epubparse-js: JS library with ergonomic API including Typescript definitions (published to npm, for use in Browser and Node.js)

Steps to release

Prepare

• bump version in epubparse-rs/Cargo.toml
• bump versions in epubparse-wasm/Cargo.toml and epubparse-wasm/package.json
• go to epubparse-wasm folder and run build_package.sh
• bump version and epubparse-wasm dependency verion in epubparse-js/package.json
• commit

Release

Crates.io

• cd into epubparse-rs
• run cargo publish --dry-run to verify
• run cargo publish

NPM

Wasm

• cd into epubparse-wasm
• run wasm-pack login
• run wasm-pack publish

JS

• run npm run build
• run npm publish