Skip to content
heff

Runes and strings

-- | >>> greeting
-- | "héllo, \"😀\"\n"
greeting : String
greeting = "h\u{E9}llo, \"" ^ "\u{1F600}\"\n"


-- | >>> List.map vowel ['a', 'é', 'z']
-- | [True, True, False]
vowel : Rune -> Bool
vowel 'a' = True
vowel 'é' = True
vowel _ = False

Runes

A Rune is a Unicode code point, as Go’s rune is.

  • A rune literal is one character in single quotes: 'a', 'é', '😀'.
  • Escapes are a string’s: '\n', '\t', '\r', '\u{1F600}', and \ before any other character is that character, '\'', '\\'.
  • It must be a Unicode scalar value: '\u{D800}', a surrogate, is an error.
  • A ' inside a name is a prime: x'.
  • A rune literal can be a pattern.
  • A rune is not an integer: 'a' is no Int, and 97 is no Rune.

debug 'a' is "'a'"; show 'a' is "a".

Strings

A String is bytes, as Go’s is. A literal is UTF-8 ("é" is two bytes), but a string may hold any bytes, such as what a file gave, or "\xff".

  • A string literal is in double quotes, and can be a pattern.
  • Its escapes are \n, \t, \r, \u{...} (a code point, in UTF-8), \xHH (one byte), and \ before any other character, which is that character: \", \\.

debug "hi" is "\"hi\""; show "hi" is "hi".

Operations

What is done with them is in the library:

  • strings: the prelude, such as strLength, runes, fromRunes and split; ^ joins two, and groups to the left: s ^ a ^ b is (s ^ a) ^ b, so a string that nothing else holds is grown in place; mapBytes, mapRunes and foldBytes go through a string byte by byte or rune by rune without making a list of them;
  • runes: rune, imported as import rune as Rune.

JavaScript’s own UTF-16 text is another type, JSString: see Providers.