Saltar al contenido principal

Introduction to Wave v0.1.2-pre-beta: Assignment Operators, Added Remainder and Generalized Indexing Support

· 3 min de lectura
LunaStev
LunaStev
Programming Language Engineer

Hello! I'm LunaStev, the developer of Wave.

We are pleased to announce Wave v0.1.2-pre-beta — This update supports indexing, supports the rest of the operators. And I added an assignment operator. And I fixed a number of bugs.

✅ Added Features​

🔍 Generalized Indexing Support​

  • Extended arr[index] syntax to support variable indexing in any context.

  • Now supports dynamic indexing like arr[i], ptr[j], get_array()[k] inside if, while, return, and assignments.

  • Internally handles pointer types (e.g., i8*) using GEP and performs automatic type casting (e.g., i8 vs i64) for comparisons and assignments.

➕ Added % (Remainder) Operator Support​

  • Wave now supports the modulo operator % for integer types.

  • Expressions like a % b, 10 % 3, and var x: i32 = a % b are now fully parsed and compiled to LLVM IR using srem (signed remainder).

  • Parser was updated to recognize % as TokenType::Remainder, and IR generation uses build_int_signed_rem for correct runtime behavior.

  • % is fully supported inside expressions, assignments, conditionals, and return statements.

⚙️ Assignment Operators (+=, -=, *=, /=, %=)​

  • Supports compound assignment operators for both integers (i32) and floating-point numbers (f32).

  • Operators implemented: +=, -=, *=, /=, %=.

  • Type-safe operation with proper distinction between integer and float IR instructions:

    • add, sub, mul, sdiv, srem for integers.

    • fadd, fsub, fmul, fdiv, frem (uses fmodf) for floats.

  • Implicit type casting during assignment (e.g., assigning an int to a float variable triggers int → float conversion).

  • Proper LLVM IR generation for all supported operations, including float remainder via external fmodf call (linked with -lm).

🐛 Bug Fixes​

🔢 Accurate Token Differentiation Between Integers and Floats​

  • Number parsing logic has been overhauled to properly distinguish between integer literals (e.g., 42) and floating-point literals (e.g., 3.14, 42.0).

  • Previously, all numeric literals were parsed as Float(f64) tokens, even when no decimal point was present. → Now, tokens like 123 are correctly parsed as Number(i64), and 123.45 as Float(f64).

  • Introduced internal flag is_float to detect presence of . during scanning. If found, the number is parsed as a float; otherwise, as an integer.

  • Implemented type-safe error handling:

    • Fallbacks to TokenType::Number(0) or TokenType::Float(0.0) on parse failure, ensuring the lexer remains stable on malformed input.

  • This fix improves downstream type checking, IR generation, and expression evaluation, especially for typed languages like Wave where i64 and f64 must be handled distinctly.

✨ Other Changes​

🧠 Library and Binary 2 Coexist​

  • Add lib.rs for easy package manager creation, development, and easy access.

Showcase​

The showcase is available at Wave-Test.

Thank you for using Wave! Stay tuned for future updates and enhancements.

Installation Guide​

  1. Download:

    • Download to Curl. 
      curl -fsSL https://wave-lang.dev/install.sh | bash -s -- --version v0.1.2-pre-beta

  2. Verify Installation:

    • Open a terminal and type: 
      wavec --version
    • If the version number displays, the installation was successful.

PR and Commits​

Contributor​

@LunaStev | 🇰🇷

Website​

Website GitHub Discord Ko-fi