Source Code Representation

Source code is Unicode text encoded in UTF-8.

Characters

unicode_char  = /* any Unicode code point except NUL (U+0000) */ .
letter        = 'A' … 'Z' | 'a' … 'z' .
digit         = '0' … '9' .
newline       = /* U+000A */ .
whitespace    = ' ' | '\t' | '\r' | newline .

Carriage return (U+000D) followed by newline is normalized to newline. A lone carriage return is treated as newline.

Source Files

source_file = { import } { declaration } .

Source files use .ori extension. Test files use .test.ori extension.

File names must:

• Start with ASCII letter or underscore
• Contain only ASCII letters, digits, underscores
• End with .ori or .test.ori

Encoding

Source files must be valid UTF-8 without byte order mark. Invalid UTF-8 or presence of BOM is an error.

Line Continuation

A newline does not terminate the logical line when preceded by:

• Binary operators: +, -, *, /, %, &&, ||, ==, !=, <, >, <=, >=, .., ..=
• Opening delimiters: (, [, {
• Comma, assignment (=), arrow (->), colon (:)

Module Mapping

Each source file defines one module. Module name derives from file path relative to source root:

See Modules.