Syntax

This page summarizes the source-level syntax that all Lutra programs use. Use it when you need to confirm how names, files, paths, and punctuation work.

Files and top-level structure¶

A Lutra source file contains a sequence of top-level definitions. In practice, these are usually const, func, type, import, and module definitions.

import std::cmp

type Status: enum {open, closed}

const greeting = "hello"

func main() -> cmp(1: int32, 2)

Comments¶

Line comments start with # and continue to the end of the line.

# This is a comment.
const main = 1: int32

Identifiers¶

Identifiers name values, functions, modules, types, fields, and enum variants.

const album_count = 10: int32
func add_one(x: int32) -> x + 1

Lutra also supports qualified paths with ::.

std::option::or_default
std::date::to_timestamp
project::pg::from_albums

Delimiters and separators¶

Lutra uses these delimiters:

( and ) for grouping, function parameters, and function calls.
{ and } for tuples, tuple types, and module bodies.
[ and ] for arrays and array types.

It uses these separators and punctuation marks:

, between items.
: for type annotations and type definitions.
-> for function bodies.
=> for match branches.
. for field access.
| for pipelines and alternative patterns.
.. for tuple spread.
; between let bindings inside a block expression.

Grouping¶

Parentheses group expressions and are commonly used for:

multiline pipelines,
if branches,
match subjects,
let blocks.

func main() -> (
  [1, 2, 3]: [int32]
  | std::map(x -> x + 1)
)

Source formatting¶

Whitespace is used for readability, but punctuation determines structure. For example, these two function definitions are equivalent:

func add_one(x: int32) -> x + 1

func add_one(x: int32) -> (
  x + 1
)