This is a minor mode for GNU Emacs aiming to address a common need of Japanese language learners - getting the stroke order of a particular Kanji sign. It has a built-in collection of SVG images depicting stroke orders for all Kanji. The collection is a slightly modified and limited version of the SVGs provided by the KanjiVG project. The mode's operation is pretty simple. It retrieves the Unicode code of the character under the Emacs cursor and uses it to retrieve the proper SVG file (the files are happily named using Unicode numbers). It then displays the relevant SVG in a transient buffer.
Another useful feature is transcription (Kanji to Hiragana and Kanji to Romaji). This is achieved by leveraging Kakasi - a CLI tool for transcribing between various scripts used in Japanese.
Kanji-mode is available on MELPA to be installed via package.el. If you have MELPA configured as package repository, just run M-x package-install and enter "kanji-mode". You can also select it for installation while browsing the package list after running M-x package-list-packages.
If you have use-package enabled, you can also install and enable kanji-mode as follows:
(use-package kanji-mode
:ensure t)
To install kanji-mode manually download this repository somewhere you keep your Emacs modes (I keep it under ~/.emacs.d/). Then follow these simple steps:
- Make sure the
kanji-modedirectory is part of your Emacsload-path. You can do it by adding the following line to your.emacsconfiguration file:(add-to-list 'load-path "/path/to/your/kanji-mode") - Require
kanji-modeby adding the following to.emacs:(require 'kanji-mode).kanji-modeadds a hook totext-mode, so it should launch automatically whenever you edit a text file. If you need it in other circumstances (temporary buffers, etc.), you can always launch it manually by hitting M-x and entering "kanji-mode". WARNING In order to be able to use transcription, you need to install the Kakasi utility. Most Unix-like operating systems already have some convenient way of installing it, otherwise you can find the source here.
Once kanji-mode is active, you should be able to see the string "kanji" in the minor mode list in the buffer mode line at the bottom. Whenever you want to check the stroke order for a character under the cursor, hit M-s M-o (I tried to make this keybinding intuitively suggest "stroke order") and a new buffer will appear, containing the image of your character with its stroke order. When you're done viewing it, hit q to close the buffer and return to your text. Below are two screenshots showing an example usage:
The transcription functions operate on currently selected region or, if region is empty, on the current thing-at-point. The key bindings for them are: M-s M-h (for function kanji-mode-kanji-to-hiragana) and M-s M-r (for function kanji-mode-all-to-romaji). Both functions' behavior follows the same pattern with respect to the prefix argument:
- No prefix argument: echo the result in the minibuffer
- Prefix argument
-1(invokable through M--): place the result in the kill-buffer - Any other prefix argument: create a new buffer containing the result (which can be closed by pressing q)
When trying the Kakasi transcription feature under OpenBSD I experienced some non-obvious problems whereby I would get an empty string instead of the transcription. Some investigation led me first to realize that my LANG and LC_ALL environment were set to C, causing non-ASCII characters to be garbled when running system commands. Setting them to some UTF-8 locale (e.g. en_US.UTF-8) should solve this problem. Another thing I've noticed is that the version of Kakasi available in the Ports system - 2.3.4 for OpenBSD 6.0 and current - had some trouble producing proper characters. My suggestion is to manually install the latest version - 2.3.6. In order for the ./configure script to succeed you will need to specify some additional paths:
$ wget http://kakasi.namazu.org/stable/kakasi-2.3.6.tar.gz
$ tar xvzf kakasi-2.3.6.tar.gz
$ cd kakasi-2.3.6
$ CPPFLAGS="-I /usr/local/include" CFLAGS="-I /usr/local/include" LDFLAGS="-L /usr/local/lib" ./configure
$ make
$ make install
I'd like to thank Ulrich Apel and all contributors of the KanjiVG project for doing the meticulous work I'd never have the patience to do. I've only made small changes to their files (added a white background so it works well with dark Emacs themes and increased the size to make it more legible) and dropped variants (Kaisho, etc.) since they were unusable in my mode. I'd also like to extend my thanks to Hironobu Takahashi, author of the Kakasi utility, which does an excellent job transcribing between different alphabets.