[go: up one dir, main page]

depyler 3.19.4

A Python-to-Rust transpiler focusing on energy-efficient, safe code generation with progressive verification
docs.rs failed to build depyler-3.19.4
Please check the build logs for more information.
See Builds for ideas on how to fix a failed build, or Metadata for how to configure docs.rs builds.
If you believe this is docs.rs' fault, open an issue.
Visit the last successful build: depyler-2.2.2

Depyler

Crates.io Documentation CI Coverage License: MIT Apache License Rust 1.83+

A Python-to-Rust transpiler with semantic verification and memory safety analysis. Depyler translates annotated Python code into idiomatic Rust, preserving program semantics while providing compile-time safety guarantees.

Recent Updates (v3.18.2)

Emergency Bug Fix Sprint - Critical transpiler fixes:

  • ✅ Fixed async methods missing async keyword in classes
  • ✅ Fixed variable initialization in async functions (await expressions)
  • ✅ Fixed print() statements (now correctly generates println!() macro)
  • ✅ Added Assert statement support (was completely missing)
  • ✅ Fixed array literal transpilation bugs
  • ✅ Added CI validation to ensure all transpiled code compiles

Status: All P0 blocking issues resolved. Generated code quality significantly improved.

See CHANGELOG.md for complete details.

Installation

cargo install depyler

Requirements

  • Rust 1.83.0 or later
  • Python 3.8+ (for test validation)

Usage

Basic Transpilation

# Transpile a Python file to Rust
depyler transpile example.py

# Transpile with semantic verification
depyler transpile example.py --verify

# Analyze migration complexity
depyler analyze example.py

Example

Input (example.py):

def fibonacci(n: int) -> int:
    if n <= 1:
        return n
    return fibonacci(n - 1) + fibonacci(n - 2)

Output (example.rs):

fn fibonacci(n: i32) -> i32 {
    if n <= 1 {
        return n;
    }
    fibonacci(n - 1) + fibonacci(n - 2)
}

Library Usage

use depyler::{transpile_file, TranspileOptions};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let options = TranspileOptions::default()
        .with_verification(true);

    let rust_code = transpile_file("example.py", options)?;
    println!("{}", rust_code);

    Ok(())
}

Features

Core Capabilities

  • Type-directed transpilation: Uses Python type annotations to generate appropriate Rust types
  • Memory safety analysis: Infers ownership and borrowing patterns
  • Semantic verification: Property-based testing to verify behavioral equivalence
  • Multiple backends: Generate Rust or Ruchy script code

Supported Python Features

Currently Supported:

  • Functions with type annotations
  • Basic types (int, float, str, bool)
  • Collections (List, Dict, Tuple, Set)
  • Control flow (if, while, for, match)
  • List/dict/set comprehensions
  • Generator expressions (NEW in v3.13.0) ✨
  • Generator functions (yield statements)
  • Exception handling (mapped to Result<T, E>)
  • Classes and methods
  • Assert statements (NEW in v3.18.2) ✨
  • Async/await (functions and methods - FIXED in v3.18.2)
  • Context managers (with statements)
  • Iterators
  • Print statements (correctly generates println! macro)

Not Supported:

  • Dynamic features (eval, exec)
  • Runtime reflection
  • Multiple inheritance
  • Monkey patching

See documentation for complete feature list.

MCP Integration

Depyler provides an MCP (Model Context Protocol) server for integration with AI assistants like Claude Code.

Setup

Add to Claude Desktop config (~/.config/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "depyler": {
      "command": "depyler",
      "args": ["agent", "start", "--foreground", "--port", "3000"]
    }
  }
}

Available Tools

  • transpile_python - Convert Python code to Rust
  • analyze_migration_complexity - Analyze migration effort
  • verify_transpilation - Verify semantic equivalence
  • pmat_quality_check - Code quality analysis

See docs/MCP_QUICKSTART.md for detailed usage.

Architecture

Depyler uses a multi-stage compilation pipeline:

Python AST → HIR → Type Inference → Rust AST → Code Generation

Key components:

  • Parser: RustPython AST parser
  • HIR: High-level intermediate representation
  • Type System: Conservative type inference with annotation support
  • Verification: Property-based testing for semantic equivalence
  • Codegen: Rust code generation via syn/quote

Quality Standards

This project follows strict quality standards enforced by CI:

  • Test coverage: 70%+ (440+ passing tests in core, 600+ workspace-wide)
  • Max cyclomatic complexity: ≤10 (enforced via PMAT)
  • Max cognitive complexity: ≤10 (enforced via PMAT)
  • Zero clippy warnings (-D warnings - BLOCKING)
  • Zero self-admitted technical debt (SATD - BLOCKING)
  • TDG grade: A- minimum (≥85 points)
  • NEW: CI validates all transpiled code compiles (v3.18.2)

Development

Running Tests

# Run all tests
cargo test --workspace

# Run with coverage
cargo llvm-cov --html --open

# Run benchmarks
cargo bench

Quality Checks

# Lint
cargo clippy --all-targets --all-features -- -D warnings

# Format
cargo fmt --all

# Quality gates
pmat quality-gate

Documentation

License

Licensed under either of:

at your option.

Contributing

Contributions are welcome. Please follow the quality standards:

  1. Write tests first (TDD)
  2. Maintain 80%+ coverage for new code
  3. Pass all clippy checks
  4. Update documentation

See CONTRIBUTING.md for details.