jdpub

jdpub annotates a source document with definitions for difficult vocabulary words to make reading easier.

jdpub currently only supports annotating documents for the Japanese language.

Warning

This software is version 0.2.0, expect configuration and usage to change. The alpha release exists to provide the tool as early as possible.

Usage

A web view for jdpub is currently in development. Until then, refer to the CLI.

Usage: jdpub [OPTIONS] --output <OUTPUT> <INPUT>

Arguments:
  <INPUT>  Input uri/file

Options:
  -o, --output <OUTPUT>  Output file
  -c, --config <CONFIG>  Configuration file
  -d, --debug            debug flag
  -v, --verbose          Verbose flag
      --cover <COVER>    Cover image uri/file
      --title <TITLE>    Title of export
      --author <AUTHOR>  Author of export
  -h, --help             Print help
  -V, --version          Print version

Example usage:

• Parsing a website: jdpub https://hachha.dev/blog/jdpub -o jdpub.epub
• Parsing a file: curl https://hachha.dev/blog/jdpub -o jdpub.html && jdpub jdpub.html -o jdpub.epub

Motivation

This software was created to support reading more advanced material on an e-reader as opposed to relying on other existing tools or services like yomitan or LingQ that require a phone or pc. Additional motivation for the initial release can be read here. jdpub supports fetching and parsing webpages with various parse rules for navigating elements.

jdpub has the following goals:

• Add dictionary tooltips for the Japanese language
• Add furigana annotations for the Japanese language
• Support custom parsing rules from the source document
• Import from HTML
• Import from EPUB
• Import from plaintext
• Export to HTML
• Export to EPUB
• Export to plaintext
• Provide a web-interface for client-side conversion (WASM)
• Provide more options for adding additional metadata to exported EPUBs
• Provide a sane default styles.css
• Image support
• Additional dictionary support
• Support multiple configuration files
• Improve configuration merging support
• Support caching fetched HTML files

jdpub explicitly has the following non-goals:

• jdpub is not a reader.
• jdpub is not designed for document conversion, use calibre or other software.

While jdpub is currently only designed to support Japanese, it could be easily extended to support additional languages. The Japanese writing system is typically more complex than western writing, so techniques to tokenize the words and lookup definitions with respect to conjugations is itself more complex. For languages like English that can be tokenized solely by whitespace, the results should be better.

Software Support

jdpub adds popup tooltips to text that are supported by various e-readers. Popup tooltips work by adding the epub:type="noteref" and epub:type="footnote" attributes to XHTML anchors and asides, respectfully. Compatibility has been tested on the following readers:

• Koreader (enable Show footnotes in popup)
• Neoreader

References

• Apple popup tooltips
• HTML vs XHTML
• Usability of footnotes
• iBook support

Installation

jdpub relies on 2 crates that currently make it impossible to build manually. durf_parser is an HTML parsing & minimizing crate that I have not yet released. jmdict-fast is an excellent crate for Japanese word lookup, but building on my own pc requires modifications to the source that I have note yet upstreamed anywhere.

Releases will be periodically provided for x64 Linux until this is resolved. I have no plans to test a Windows build at this point, but I can produce one if requested. Otherwise, I suggest waiting until the web view is available.