A port of ucwidth C library to Elixir, with Emoji (basis, variations and zwj-combined-sequences) supported.

Online documentation

When developing a terminal based Elixir application, I found lack of some library handling character width correctly. Turns out there's a need to port ucwidth to Elixir. So I made this library.

I'm not professional at Unicode, just read the specifications and some references. So please report any issues you meet, thanks in advance.

This library provides a function Ucwidth.width/2 for detecting the display width of a Unicode character.

For example:

iex> Ucwidth.width("行")
2

iex> Ucwidth.width("h")
1

iex> Ucwidth.width("\x0C")
0

iex> Ucwidth.width("Hello, <<👪>>, family lovers!")
29

Note: according to Unicode standard, not every character has fixed width. There are some characters with variable width, which is called ambiguous width characters. Their width vary depends on the context. This library by default treats ambiguous charaters as narrow ones, and also provides a way to pass be context, e.g:

iex> Ucwidth.width("\u00a1")
1

iex> Ucwidth.width("\u00a1", :wide)
2

Add ucwidth to your list of dependencies in mix.exs:

def deps do
  [
    {:ucwidth, "~> 0.2"}
  ]
end

Attention to performance was paid during development. You can check the performance by running:

mix run bench/benchmark.exs

This library is open-sourced under an MIT license.