The Language library will help you with localizing strings.

There is the global language dictionary (__LANGUAGE__), which contains groups, which contain identifiers and translations.

__LANGUAGE__ = {
    "__ALL__": { #`__ALL__` is the group for all translations, useful when you have only one language
        "HelloWorld": "Hello, World!"
    }
}
assert Localize( "#HelloWorld" ) == Get( "HelloWorld" ) == "Hello, World!"

The __ALL__ group is the default group for all translations, useful when you have one language.

The Localize function is the only one which I want to talk about more than in the documentation, and there's no better way to talk about something complicated than straight up show it.

Add( "Class", "Class" )
Add( "OtherClass', "Other Class" )
print( Localize( '#Class is the parent of #OtherClass who is the parent of #Another #Class' ) ) #"Class is the parent of Other Class who is the parent of Another Class"

Now, you might be wondering - why did "Another Class" translate correctly? Well, it's quite simple - "#Another" turns to "Another", as it's a non-existent translation, and "#Class" turns into "Class".

Identifiers for translations and groups can only contain the following set of characters.

{ '_',
  'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z',
  'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
  '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' }