Vim, Unicode, and Digraphs

2021-February-16  Used Pandoc to convert the source of this article to AsciiDoc, edited the AsciiDoc results of that conversion, and used Hugo and Asciidoctor to generate an updated version of this article.

2020-November-4  As of today, this article, which was originally published elsewhere, has been on the web for 7 years.🎂

Unicode, in particular UTF-8 encoding, is now the de facto standard for writing on the web. In this article, I describe how I set up Vim to make it easy to write using Unicode characters that aren’t on my keyboard (for example ☹︎, ☺︎, and ♥︎). In a nutshell, I set up Vim to use:

Whether you use terminal Vim or GUI Vim, put the following in your vimrc:

If you use terminal Vim, you also need to configure your terminal emulator to use UTF-8 encoding. The way to do that depends on your terminal emulator.

Choosing a font is a huge topic, but for this article I’ll just say that you want a font that:

The monospace fonts that I use are DejaVu Sans Mono, Monaco, and Consolas. DejaVu Sans Mono is free/gratis and free/libre. Monaco is included on Macs and Consolas is included on Windows. Neither Monaco nor Consolas are free/libre.

The way to tell Vim which font to use depends on whether you are using a GUI Vim, such as gVim or MacVim, or terminal Vim. If you use terminal Vim, specify the font in your terminal emulator’s configuration.

If you use a GUI Vim, put something like the following in your vimrc:

Replace DejaVu_Sans_Mono:h16⁠^[1] with the font and size you want to use. I recommend putting this in an if-block so you can use this vimrc for both terminal Vims and GUI Vims. If you use the same vimrc across multiple devices, you can specify a list of guifonts like this:

Vim will use the first font that’s available on the current device.

Vim has built-in digraphs, which are standard 2-character keyboard shortcuts for entering Unicode characters that are specified in RFC 1345: Character Mnemonics and Character Sets. To learn about Vim’s digraphs, type the following in Vim in command-mode.

You can also read the above built-in Vim Help files on the web at Vim documentation: digraph.

Here’s an excerpt from help digraph-table:

To insert a character using its digraph, do the following:

For example, to enter a White Smiling Face (☺︎), type the following in insert-mode:

To enter a White Heart Suit (♡), type the following in insert-mode:

The default digraphs are hard to remember and almost every time I use one I need to look it up by typing:

where string is part of the name of the character that I’m searching for. For example, to find the above characters (☺ and ♡), I search the digraph-table using /face and /heart. To make my life easier, I’ve set my own personal easy-to-remember digraphs in my vimrc. Here are some of my vimrc dig settings:

The syntax, which is described in :help dig, is:

To make it easy to remember what Unicode characters I’ve set up, I include comments in my vimrc like this:

In a vimrc, text after a double quote mark (") is a comment.

With these settings, I can type the following in insert mode to insert these Unicode characters:

If I forget the digraphs that I’ve set up in my vimrc, I can list all digraphs (default and personal) by typing the following in command-mode:

At the end of the list are the digraphs I’ve added:

To quickly get to the end of the list of digraphs, type:

In Vim’s command-mode, G means “Goto last line.”

To specify a digraph for a character, you need to know the decimal representation (codepoint) of the character. Vim (of course ☺) has commands to find out the decimal, hexadecimal, and octal representations of a character. Here’s how:

Alternatively, for step 3:

For example, with either of the above commands Vim displays the following for the White Smiling Face character:

To learn more about these commands, see either:

These two commands are equivalent and are sometimes called the “get ascii” command. You can read the Vim Help file either within Vim or at Vim documentation: various #ga.