Generate and manage custom e-reader dictionaries for your favourite fictional worlds.

Most e-readers have built-in dictionaries, letting you define words in-text as you read. While this feature can tell you the meaning of words like 'tabard' and 'thaumaturgy', it can't help you define names, places, or magical items specific to the world of your book.

Runik leverages the crowd-sourced info of fan wikis to generate custom e-reader dictionaries, extending the volcabulary of your e-reader to include fictional references.

Binaries for Linux, macOS, and Windows can be found on the releases page.

To properly convert runik dictionaries for Kindles, you'll need to install Kindle Previewer, which bundles the kindlegen program required for dictionary compilation.

1. Download and install Kindle Previewer.
2. In Runik, navigate to the Settings tab and enter the full path to the kindlegen executable.

The default kindlegen installation locations are:

• macOS: /Applications/Kindle Previewer 3.app/Contents/lib/fc/bin/kindlegen
• Windows: C:\Users\<username>\AppData\Local\Amazon\Kindle Previewer 3\lib\fc\bin\kindlegen
• Linux: Kindle Previewer does not support linux, but old versions of the standalone kindlegen CLI program should still work.

1. Clone the project

2. Run the app in dev mode

   make dev
   

   If you're developing on linux use the make dev-linux command.

3. Build for your target OS

   make build
   

Note: To build a flatpak for linux use the make build-flatpak command

You'll need the following software installed on your system before being able to build and run runik.

• Wails CLI (install the version specified in go.mod)
• Go
• Nodejs + npm
• A C compiler

Runik is in early development and is bound to contain bugs. Feel free to report any issues you encounter.

Runik is designed to work with MediaWiki sites. There is variablility in the way wikis can be configured or the way articles are written -- runik is not guaranteed to work with all of them. I mostly tested against Fandom wikis which are pretty consistent and should be well supported.

If Runik does not yet support your device -- and you are comfortable tinkering -- you can use the Runik Builder cli tool to generate dictionaries and use one of the many other tools out there to convert it for use with your device.

Why are Kobo dictionaries prefixed with dicthtml-[r]?

Kobo expects dictionaries to be in the following format dicthtml-<language-code>, automatically detecting dicthtml-es as a spanish language dictionary and displaying it as Espanol. Without the dicthtml prefix, kobos don't know to treat this file as a dictionary, and if we tried to put the dictionary name alone after the prefix (eg. dicthtml-gameofthrones), the kobo treats the first two letters "ga" as a language code, interpreting this as a gaelic dictionary. So instead, we use characters that cannot be mistaken for a language code before the name of the dictionary, "[r]". It's not a perfect solution and I'm working on a solution to get more user-friendly kobo dictionary names.

• Wails
• DictUtil
• Svelte