Std.Char

The Unicode code point — Cure's character type.

A Char is a Unicode scalar, represented by its code point. It is a nominal opaque carrier rather than an alias for Bounded(0x110000), so APIs cannot accidentally accept arbitrary bounded values as characters. @builtin(:char) names the kernel rule that introduces its values: a character literal is a compact code point, so the runtime representation is one machine integer and no constructor is needed (or allowed). Character literals ('a', '\n') elaborate to Char values, and a String stores its text as a List(Char) behind its own nominal boundary. This module gives the type its visible, documented home, mirroring Std.Int/Std.Float/Std.Binary/Std.Atom.

Construction is confined to literals and the checked from_code_point boundary below, which excludes values outside Unicode scalar space.

This module is the floor of the text layer: Std.Literal describes the literal syntax that produces a Char, and Std.String stores its text as a List(Char), so both sit above it. Std.Char therefore names neither — the ExpressibleByCharacterLiteral instance lives with the interface in Std.Literal, and the String-shaped case conversions live with the type in Std.String.

Functions

# fn ascii_lowercased(c: Char) -> Char extern

ASCII-only one-to-one fold for protocols (such as non-Unicode regex caseless matching) that explicitly require ASCII semantics.
# fn ascii_value(c: Char) -> Option(Int) extern
# fn between(c: Char, first: Char, last: Char) -> Bool extern
# fn code_point(c: Char) -> Int extern

The Unicode code point of c as an Int. A Char already erases to its code point, so the runtime bridge is the identity; this names the Char -> Int coercion so Std.Comparable's Char/String instances can compare code points. Named code_point (not to_int) because Std.String also exposes a to_int and the dependent pipeline resolves globals by bare name.
# fn from_code_point(value: Int) -> Option(Char) extern

Checked inverse of code_point. JSON and other text decoders use this for escaped Unicode scalars; invalid values and UTF-16 surrogate code points are rejected rather than wrapped into Bounded(0x110000).
# fn from_unicode_name(name: List(Char)) -> Option(Char) extern

Compile-time Unicode character-name lookup for syntax macros. The name arrives as a List(Char); successful macro expansion emits the resulting Char literal, so no name table or parser reaches runtime regex code.
# fn from_valid_code_point(value: Int) -> Char extern

Construct from a code point the caller has already proved is a Unicode scalar. Parsers use this only after checking range and surrogate rules.
# fn hex_digit_value(c: Char) -> Option(Int) extern
# fn is_ascii(c: Char) -> Bool extern
# fn is_cased(c: Char) -> Bool extern
# fn is_currency_symbol(c: Char) -> Bool extern
# fn is_hex_digit(c: Char) -> Bool extern
# fn is_horizontal_space(c: Char) -> Bool extern
# fn is_letter(c: Char) -> Bool extern
# fn is_lowercase(c: Char) -> Bool extern
# fn is_math_symbol(c: Char) -> Bool extern
# fn is_newline(c: Char) -> Bool extern
# fn is_number(c: Char) -> Bool extern
# fn is_punctuation(c: Char) -> Bool extern
# fn is_symbol(c: Char) -> Bool extern
# fn is_unicode_code_point(value: Int) -> Bool extern

Named scalar-boundary predicates keep parsers from duplicating Unicode's numeric ceiling or the reserved UTF-16 surrogate interval.
# fn is_unicode_digit(c: Char) -> Bool extern

Unicode property predicates used by pure Cure text parsers. These expose immutable Unicode tables only; no regex engine or compiled handle crosses the boundary.
# fn is_unicode_scalar_code_point(value: Int) -> Bool extern
# fn is_unicode_space(c: Char) -> Bool extern
# fn is_unicode_word(c: Char) -> Bool extern
# fn is_uppercase(c: Char) -> Bool extern
# fn is_utf16_surrogate_code_point(value: Int) -> Bool extern
# fn is_vertical_space(c: Char) -> Bool extern
# fn is_whitespace(c: Char) -> Bool extern
# fn is_whole_number(c: Char) -> Bool extern
# fn less_than(a: Char, b: Char) -> Bool extern
# fn lowercased_characters(c: Char) -> List(Char) extern
# fn same(a: Char, b: Char) -> Bool extern

Scalar equality and ordering. Keeping these operations here prevents consumers from unpacking Char into implementation-level integers.
# fn unicode_category(c: Char) -> Atom extern

Unicode General_Category as its stable two-letter atom (:Lu, :Nd, …). Regex property matching consumes this immutable classification; no host regex engine or compiled pattern crosses the boundary.
# fn uppercased_characters(c: Char) -> List(Char) extern

Case conversion answers a List(Char) because Unicode mappings may expand one scalar into several (uppercasing 'ß' yields SS). The shim returns a bare code-point list and an @extern performs no marshalling, so this is the shape the boundary really has; Std.String.uppercased_character/1 and Std.String.lowercased_character/1 assemble the nominal String form, which erases to {String, chars} and belongs with the type that owns it.
# fn whole_number_value(c: Char) -> Option(Int) extern

Functions

# fn ascii_lowercased(c: Char) -> Char extern

# fn ascii_value(c: Char) -> Option(Int) extern

# fn between(c: Char, first: Char, last: Char) -> Bool extern

# fn code_point(c: Char) -> Int extern

# fn from_code_point(value: Int) -> Option(Char) extern

# fn from_unicode_name(name: List(Char)) -> Option(Char) extern

# fn from_valid_code_point(value: Int) -> Char extern

# fn hex_digit_value(c: Char) -> Option(Int) extern

# fn is_ascii(c: Char) -> Bool extern

# fn is_cased(c: Char) -> Bool extern

# fn is_currency_symbol(c: Char) -> Bool extern

# fn is_hex_digit(c: Char) -> Bool extern

# fn is_horizontal_space(c: Char) -> Bool extern

# fn is_letter(c: Char) -> Bool extern

# fn is_lowercase(c: Char) -> Bool extern

# fn is_math_symbol(c: Char) -> Bool extern

# fn is_newline(c: Char) -> Bool extern

# fn is_number(c: Char) -> Bool extern

# fn is_punctuation(c: Char) -> Bool extern

# fn is_symbol(c: Char) -> Bool extern

# fn is_unicode_code_point(value: Int) -> Bool extern

# fn is_unicode_digit(c: Char) -> Bool extern

# fn is_unicode_scalar_code_point(value: Int) -> Bool extern

# fn is_unicode_space(c: Char) -> Bool extern

# fn is_unicode_word(c: Char) -> Bool extern

# fn is_uppercase(c: Char) -> Bool extern

# fn is_utf16_surrogate_code_point(value: Int) -> Bool extern

# fn is_vertical_space(c: Char) -> Bool extern

# fn is_whitespace(c: Char) -> Bool extern

# fn is_whole_number(c: Char) -> Bool extern

# fn less_than(a: Char, b: Char) -> Bool extern

# fn lowercased_characters(c: Char) -> List(Char) extern

# fn same(a: Char, b: Char) -> Bool extern

# fn unicode_category(c: Char) -> Atom extern

# fn uppercased_characters(c: Char) -> List(Char) extern

# fn whole_number_value(c: Char) -> Option(Int) extern