String Literals

UTF-8 by default; ascii/utf16/utf32 prefixes select another encoding.

§ 1Literals

wcl

name  = "hello"          // utf8 (default)
label = utf8"hello"      // explicit
tag   = ascii"id-007"
wide  = utf16"hello"
quad  = utf32"hello"

§ 2Types

Type	Use
utf8	Default. Variable-width Unicode in 1-4 byte sequences.
ascii	7-bit text; one byte per char, no Unicode.
utf16	Variable-width, two bytes per BMP code unit.
utf32	Fixed-width, four bytes per code point.

Encoding is metadata

The encoding is part of the value's type: a utf8 field rejects an ascii literal; widen or convert at the host layer.

§ 3Escapes

Inside a double-quoted string the usual escape sequences apply: backslash, escaped quote, ` , , `. For backslash-heavy text — regexes, LaTeX, Windows paths — prefer a raw heredoc (below).

wcl

greeting = "Hello,\nworld!"        // newline embedded
quoted   = "She said \"hi\"."

§ 4Heredocs

<<TAG ... TAG introduces a heredoc whose body runs until a line matching the closing tag. Escape sequences are interpreted, and indentation is stripped to match the closing tag's indentation, so heredocs nest comfortably inside blocks.

wcl

note = <<END
First line.
Second line.
END

§ 5Raw heredocs

A raw heredoc uses a single-quoted opening tag and takes the body verbatim — no escapes, no interpolation, just literal bytes. Ideal for LaTeX, regexes, code samples with backslashes, or anything where you do not want WCL touching the contents. Common leading whitespace is still stripped.

wcl

regex = <<'RAW'
\d{3}-\d{4}
RAW

latex = <<'TEX'
\frac{-b \pm \sqrt{b^2 - 4ac}}{2a}
TEX

Opt a string into expression interpolation with a $ prefix — see String Interpolation.