A very complete i18n lib for Lua
Switch branches/tags
Nothing to show
Pull request Compare This branch is 30 commits behind kikito:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



A very complete i18n lib for Lua


i18n = require 'i18n'

-- loading stuff
i18n.set('en.welcome', 'welcome to this program')
  en = {
    good_bye = "good-bye!",
    age_msg = "your age is %{age}."
    phone_msg = {
      one = "you have one new message.",
      other = "you have %{count} new messages."
i18n.loadFile('path/to/your/files/en.lua') -- maybe load some more stuff from that file

-- setting the translation context
i18n.setLocale('en') -- English is the default locale anyway

-- getting translations
i18n.translate('welcome') -- Welcome to this program
i18n('welcome') -- Welcome to this program
i18n('age_msg', {age = 18}) -- Your age is 18.
i18n('phone_msg', {count = 1}) -- You have one new message.
i18n('phone_msg', {count = 2}) -- You have 2 new messages.
i18n('good_bye') -- Good-bye!


You can interpolate variables in 3 different ways:

-- the most usual one
i18n.set('en.variables', 'Interpolating variables: %{name} %{age}')
i18n('variables', {name='john', 'age'=10}) -- Interpolating variables: john 10

i18n.set('en.lua', 'Traditional Lua way: %d %s')
i18n('lua', {1, 'message'}) -- Traditional Lua way: 1 message

i18n.set('en.combined', 'Combined: %<name>.q %<age>.d %<age>.o')
i18n('variables', {name='john', 'age'=10}) -- Combined: john 10 12k


This lib implements the unicode.org plural rules. Just set the locale you want to use and it will deduce the appropiate pluralization rules:

i18n = require 'i18n'

  en = {
    msg = {
      one   = "one message",
      other = "%{count} messages"
  ru = {
    msg = {
      one   = "1 сообщение",
      few   = "%{count} сообщения",
      many  = "%{count} сообщений",
      other = "%{count} сообщения"

i18n('msg', {count = 1}) -- one message
i18n('msg', {count = 5}) -- 5 сообщений

The appropiate rule is chosen by finding the 'root' of the locale used: for example if the current locale is 'fr-CA', the 'fr' rules will be applied.

If the provided functions are not enough (i.e. invented languages) it's possible to specify a custom pluralization function in the second parameter of setLocale. This function must return 'one', 'few', 'other', etc given a number.


When a value is not found, the lib has several fallback mechanisms:

  • First, it will look in the current locale's parents. For example, if the locale was set to 'en-US' and the key 'msg' was not found there, it will be looked over in 'en'.
  • Second, if the value is not found in the locale ancestry, a 'fallback locale' (by default: 'en') can be used. If the fallback locale has any parents, they will be looked over too.
  • Third, if all the locales have failed, but there is a param called 'default' on the provided data, it will be used.
  • Otherwise the translation will return nil.

The parents of a locale are found by splitting the locale by its hyphens. Other separation characters (spaces, underscores, etc) are not supported.


This project uses telescope for its specs. If you want to run the specs, you will have to install telescope first. Then just execute the following from the root inspect folder:

tsc -f spec/*