Module Level Code Translator Agent Skill

Module-Level Code Translator

Translate source code between programming languages while preserving behavior, adapting idioms, and generating verification tests.

Translation Workflow

Follow this sequential process for code translation:

1. Analyze Source Code

Before translating, thoroughly understand the source code:

• Identify scope: Determine if translating functions, classes, or entire modules
• Map dependencies: List all imports, external libraries, and dependencies
• Understand behavior: Identify core logic, algorithms, and expected behavior
• Note language-specific features: Identify features that require special handling (decorators, generators, async/await, etc.)
• Extract test cases: If existing tests are available, analyze them to understand expected behavior

2. Plan Translation Strategy

Create a translation plan before writing code:

• Map standard libraries: Identify equivalent libraries in target language (see references/stdlib_mappings.md)
• Adapt idioms: Plan how to translate language-specific patterns (see references/language_mappings.md)
• Handle type systems: Plan type annotations/declarations if moving between typed/untyped languages
• Identify testing framework: Choose appropriate testing framework for target language (see references/testing_frameworks.md)
• List transformations: Document major transformations needed (e.g., class → struct, decorator → annotation)

3. Translate Code

Implement the translation following target language conventions:

Core principles:

• Preserve original behavior exactly
• Follow target language idioms and conventions
• Maintain code structure when possible (functions → functions, classes → classes)
• Adapt patterns that don't translate directly (e.g., Python decorators → Java annotations)
• Add appropriate error handling for target language
• Include necessary imports and dependencies

Common transformations:

Python → JavaScript:

• Classes: Keep class structure, adapt __init__ to constructor
• List comprehensions: Convert to .map(), .filter(), .reduce()
• Decorators: Convert to higher-order functions or use libraries
• Type hints: Convert to TypeScript or JSDoc comments

Python → Java:

• Functions: Convert to static methods or instance methods in classes
• Dynamic typing: Add explicit type declarations
• Duck typing: Use interfaces or abstract classes
• List/dict operations: Use Collections framework

Python → Go:

• Classes: Convert to structs with methods
• Exceptions: Convert to error return values
• Dynamic features: Use interfaces for polymorphism
• List comprehensions: Use explicit loops

Java → C#:

• Naming: Convert PascalCase for methods
• Properties: Use C# property syntax instead of getters/setters
• Collections: Map to .NET collections
• Annotations: Convert to C# attributes

4. Generate Tests

Create comprehensive tests to verify translation correctness:

Test generation approach:

• Use appropriate test template from assets/ directory
• Port existing tests if available in source code
• Create new tests covering:
• Basic functionality for each function/method
• Edge cases (empty inputs, null values, boundary conditions)
• Error handling and exceptions
• Integration between components

Test equivalence:

• Ensure tests verify the same behavior as original code
• Use equivalent assertions (see references/testing_frameworks.md)
• Maintain test structure and organization
• Add setup/teardown as needed for target framework

5. Create Translation Summary

Document the translation with a summary including:

Mappings:

• List of source → target function/class/module mappings
• Standard library equivalences used
• External dependencies and their target equivalents

Transformations:

• Major structural changes (e.g., decorator → annotation)
• Idiom adaptations (e.g., list comprehension → map/filter)
• Type system changes (dynamic → static typing)

Verification notes:

• Test coverage summary
• Known limitations or differences
• Manual verification steps if needed

Output Format

Provide the translation in this structure:

## Translated Code

[Target language code with appropriate file structure]

## Tests

[Test code using target language testing framework]

## Translation Summary

### Mappings
- SourceClass → TargetClass
- source_function() → targetFunction()
- source_module → target.package

### Standard Library Equivalences
- source.lib.function → target.lib.function

### Transformations
- [Description of major transformations]

### Verification
- [Test coverage and verification notes]

Language-Specific Considerations

Python → JavaScript/TypeScript

• Convert snake_case to camelCase
• Handle async/await (similar syntax)
• Map list/dict operations to array/object methods
• Consider using TypeScript for type safety

Python → Java

• Add explicit type declarations throughout
• Convert modules to packages with proper structure
• Use appropriate Java collections (ArrayList, HashMap, etc.)
• Handle checked exceptions

Python → Go

• Convert classes to structs with receiver methods
• Replace exceptions with error return values
• Use goroutines for concurrency (if applicable)
• Follow Go naming conventions (exported vs unexported)

Java → C#

• Convert naming conventions (PascalCase methods)
• Use C# properties instead of getters/setters
• Map Java collections to .NET equivalents
• Convert annotations to attributes

JavaScript → Python

• Convert camelCase to snake_case
• Handle promises/async with Python async/await
• Map array methods to list comprehensions or loops
• Add type hints if using modern Python

Resources

references/

• language_mappings.md: Common idiom mappings between languages
• stdlib_mappings.md: Standard library equivalences across languages
• testing_frameworks.md: Testing framework mappings and patterns

assets/

Test templates for common target languages:

• test_template_python.py: pytest template
• test_template_javascript.js: Jest template
• test_template_java.java: JUnit 5 template
• test_template_go.go: Go testing template

Use these templates as starting points for generated tests.

Best Practices

• Preserve behavior first: Correctness is more important than idiomatic code
• Follow target conventions: Use target language's naming, structure, and patterns
• Test thoroughly: Generate comprehensive tests to verify equivalence
• Document differences: Note any behavioral differences or limitations
• Handle edge cases: Ensure edge cases are properly translated
• Consider performance: Be aware of performance implications of translations
• Use appropriate types: Leverage target language's type system effectively
• Maintain readability: Keep code readable and maintainable in target language