Viba

A DSL for defining types using algebraic operations — sum (|), product (*), and exponent (<-).

Language Reference

Syntax

Name[T, U, ...] := body

Operator	Syntax	Meaning
Assign	Name := body	Type definition
Sum	`A	B`
Product	A * B	Both A and B
Exponent	B <- A	Function from A to B
Generic	Name[T]	Parameterized type
Tag	$label T	Named field / variant
Void	void	Product identity (A * void = A)
Never	never	Sum identity (`A
Ellipsis	...	Open/variadic type
Tuple	(A, B, C)	Shorthand for A * B * C
Code block	{ ... }	Arbitrary text, supports nesting

Strings

• "double quoted" — standard string
• 'single quoted' — single-quoted string
• '''triple quoted''' — preserves newlines and whitespace

Comments

# to end of line

Operator Precedence (low to high)

1. | (sum) — left-associative
2. * (product) — left-associative
3. <- (exponent) — right-associative

Examples

# Standard ADTs
Option[T] := $some T | ()
Result[T, E] := $ok T | $err E

# Function types
Map[A, B] := B <- A
Curried := C <- B <- A

# Struct (product of tagged fields)
MatchContext :=
  Object
  * $match_result MatchResult
  * $target fx.GraphModule

# Open sum type
Variadic := A | B | ...

# Literals
Config := "fast" * 42 * 3.14

# Code block
Handler := {def forward(self, x): return x}

Modules

Module	Summary
parser.py	Lexer + parser (PLY) — .viba source to dict AST
type.py	Typed AST dataclasses — dict to Type objects with round-tripping
unparser.py	Code generator — Type objects back to .viba source
chain.py	Flattens nested binary trees into Sum/Product/Exponent chains and back
match.py	Keyword-argument pattern matching over Type subclasses
std_coding_style.py	Validates definitions against Class/FuncDeclare/FuncImplement styles

Installation

pip install ply
python -m viba.parser   # run test suite

Demo

# Optional value
Option[T] := $some T | ()

# Result with error
Result[T, E] := $ok T | $err E

# Linked list
List[T] := T * List[T] | ()

# Dictionary entry
Pair[K, V] := $key K * $value V

# HTTP handler
Handler := Response <- Request

# Parse pipeline
Parser := AST <- Tokens <- String

# 2D point
Point := $x float * $y float

# Color enum
Color := $red int | $green int | $blue int