awepatch 0.0.14

A Python library for runtime function patching using AST manipulation

awepatch

Awesome Patch - A Python library for runtime function patching using AST manipulation.

Overview

awepatch is a powerful Python library that allows you to dynamically patch callable objects at runtime by manipulating their Abstract Syntax Tree (AST). Unlike traditional monkey patching, awepatch modifies the actual code object of functions, providing a cleaner and more maintainable approach to runtime code modification.

Features

• 🔧 Runtime Function Patching: Modify function behavior without changing source code
• 🎯 AST-Based Manipulation: Clean and precise code modifications using AST
• 🔄 Automatic Restoration: Context manager support for temporary patches
• 🎭 Multiple Patch Modes: Insert code before, after, or replace existing statements
• 📦 Batch Patching: Apply multiple patches to a function in a single call
• 🧩 Pattern Matching: Use string, regex, or tuple patterns to locate code to patch
• 🎯 Nested Matching: Target nested code blocks with tuple pattern syntax
• 🔄 Manual Control: Apply and restore patches manually with CallablePatcher
• 🔗 Decorator Support: Works with decorated functions, class methods, and static methods
• ⚡ Type-Safe: Full type hints support with strict type checking

Installation

pip install awepatch

Or using uv:

uv pip install awepatch

Quick Start

Basic Function Patching

from awepatch import patch_callable, Patch

def greet(name: str) -> str:
    message = f"Hello, {name}!"
    return message

# Temporarily patch the function
with patch_callable(
    greet,
    Patch('message = f"Hello, {name}!"', 'message = f"Hi there, {name}!"', "replace")
):
    print(greet("World"))  # Output: Hi there, World!

# Function is automatically restored after context
print(greet("World"))  # Output: Hello, World!

Patch Modes

awepatch supports three different patching modes:

Replace Mode

Replace existing code with new code:

from awepatch import patch_callable, Patch

def calculate(x: int) -> int:
    x = x * 2
    return x

with patch_callable(calculate, Patch("x = x * 2", "x = x * 3", "replace")):
    print(calculate(5))  # Output: 15

Before Mode

Insert code before the matched statement:

from awepatch import patch_callable, Patch

def process() -> list[int]:
    items: list[int] = []
    items.append(3)
    return items

with patch_callable(process, Patch("items.append(3)", "items.append(1)", "before")):
    print(process())  # Output: [1, 3]

After Mode

Insert code after the matched statement:

from awepatch import patch_callable, Patch

def process() -> list[int]:
    items: list[int] = []
    items.append(3)
    return items

with patch_callable(process, Patch("items.append(3)", "items.append(5)", "after")):
    print(process())  # Output: [3, 5]

Patching Methods

Instance Methods

from awepatch import patch_callable, Patch

class Calculator:
    def add(self, x: int, y: int) -> int:
        result = x + y
        return result

calc = Calculator()
with patch_callable(calc.add, Patch("result = x + y", "result = x + y + 1", "replace")):
    print(calc.add(2, 3))  # Output: 6

Class Methods

from awepatch import patch_callable, Patch

class MathUtils:
    @classmethod
    def multiply(cls, x: int, y: int) -> int:
        result = x * y
        return result

with patch_callable(MathUtils.multiply, Patch("result = x * y", "result = x * y * 2", "replace")):
    print(MathUtils.multiply(3, 4))  # Output: 24

Static Methods

from awepatch import patch_callable, Patch

class Helper:
    @staticmethod
    def format_name(name: str) -> str:
        result = name.upper()
        return result

with patch_callable(Helper.format_name, Patch("result = name.upper()", "result = name.lower()", "replace")):
    print(Helper.format_name("HELLO"))  # Output: hello

Pattern Matching

You can use both string literals and regular expressions for pattern matching:

import re
from awepatch import patch_callable, Patch

def process_data(value: int) -> int:
    value = value + 10
    return value

# Using string pattern
with patch_callable(process_data, Patch("value = value + 10", "value = value + 20", "replace")):
    print(process_data(5))  # Output: 25

# Using regex pattern
with patch_callable(process_data, Patch(re.compile(r"value = value \+ \d+"), "value = value + 30", "replace")):
    print(process_data(5))  # Output: 35

Nested Pattern Matching

For complex nested structures, you can use tuple patterns to match nested AST nodes:

from awepatch import patch_callable, Patch

def nested_function(x: int) -> int:
    if x > 0:
        x = x * 2
    return x

# Match nested statement inside if block
with patch_callable(
    nested_function,
    Patch(("if x > 0:", "x = x * 2"), "x = x * 3", "replace")
):
    print(nested_function(5))  # Output: 15

Advanced Usage: AST Statements

For more complex modifications, you can provide AST statements directly:

import ast
from awepatch import patch_callable, Patch

def complex_function(x: int) -> int:
    x = x * 2
    return x

# Parse replacement code as AST
new_statements = ast.parse("x = x * 5").body

with patch_callable(complex_function, Patch("x = x * 2", new_statements, "replace")):
    print(complex_function(3))  # Output: 15

Multiple Patches

Apply multiple patches to a function in a single call:

from awepatch import patch_callable, Patch

def calculate(x: int, y: int) -> int:
    x = x + 10
    y = y * 2
    result = x + y
    return result

# Apply multiple patches at once
with patch_callable(
    calculate,
    [
        Patch("x = x + 10", "print('processing x')", "before"),
        Patch("y = y * 2", "y = y * 3", "replace"),
        Patch("result = x + y", "print(f'result: {result}')", "after"),
    ]
):
    print(calculate(5, 10))
    # Output:
    # processing x
    # result: 45
    # 45

# You can also apply multiple patches to the same line (before/after modes)
def process(x: int) -> int:
    x = x + 10
    return x

with patch_callable(
    process,
    [
        Patch("x = x + 10", "print('before')", "before"),
        Patch("x = x + 10", "print('after')", "after"),
    ]
):
    result = process(5)
    # Output:
    # before
    # after

Use Cases

• Testing: Mock function behavior without complex mocking frameworks
• Debugging: Inject logging or debugging code at runtime
• Hot-patching: Apply fixes or modifications without restarting applications
• Experimentation: Test code changes quickly without modifying source files
• Instrumentation: Add monitoring or profiling code dynamically

Limitations

• Lambda functions cannot be patched (they lack proper source code information)
• Functions must have accessible source code via inspect.getsourcelines()
• Pattern matching must uniquely identify target statement(s) in the function
• Only single function definitions are supported in the AST
• Conflicting patches (e.g., combining 'replace' with 'before'/'after' on same target) are not allowed

API Reference

Patch

class Patch(NamedTuple):
    """A single patch operation.
    
    Attributes:
        trgt: The pattern(s) to search for in the source code.
              Can be a string, regex pattern, or tuple of patterns for nested matching.
        repl: The replacement code (string) or AST statements (list).
        mode: The mode of patching (before/after/replace), defaults to "before".
    """
    trgt: str | re.Pattern[str] | tuple[str | re.Pattern[str], ...]
    repl: str | Sequence[ast.stmt]
    mode: Literal["before", "after", "replace"] = "before"

patch_callable

def patch_callable(
    func: Callable[..., Any],
    /,
    patch: Patch | list[Patch],
) -> CallablePatcher:
    """
    Create a patcher for a callable's code object using AST manipulation.
    
    Can be used as a context manager or by calling apply()/restore() manually.

    Args:
        func: The function to patch.
        patch: A single Patch or list of Patch objects for applying multiple patches.

    Returns:
        CallablePatcher: A patcher object that can be used as a context manager
                        or manually controlled with apply() and restore().

    Raises:
        TypeError: If func is not callable or is a lambda function.
        ValueError: If pattern is not found, matches multiple locations, or patches conflict.
    """

CallablePatcher

class CallablePatcher:
    """A patcher object for managing function patches."""
    
    def apply(self) -> Callable[..., Any]:
        """Apply the patches to the function and return it."""
    
    def restore(self) -> None:
        """Restore the original function."""
    
    def __enter__(self) -> Callable[..., Any]:
        """Context manager entry - applies patches."""
    
    def __exit__(self, exc_type, exc_value, traceback) -> None:
        """Context manager exit - restores original function."""

Using CallablePatcher

from awepatch import patch_callable, Patch

def my_function(x: int) -> int:
    return x + 1

# Create a patcher
patcher = patch_callable(my_function, Patch("return x + 1", "return x + 2", "replace"))

# Apply manually
patcher.apply()
print(my_function(3))  # Output: 5

# Restore manually
patcher.restore()
print(my_function(3))  # Output: 4

# Or use as context manager
with patcher:
    print(my_function(3))  # Output: 5
print(my_function(3))  # Output: 4

Development

Setup Development Environment

# Clone the repository
git clone https://github.com/fanck0605/awepatch.git
cd awepatch

# Install development dependencies
uv sync

Running Tests

# Run all tests
pytest

# Run with coverage
pytest --cov=awepatch --cov-report=html

# Run specific test file
pytest tests/test_patch_callable.py

Code Quality

# Format code
ruff format

# Lint code
ruff check

# Fix auto-fixable issues
ruff check --fix

Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Author

Chuck Fan - [e-mail]

Acknowledgments

• Inspired by the need for cleaner runtime code modification in Python
• Built with modern Python tooling and best practices

---

Note: This library modifies function code objects at runtime. Use with caution in production environments and always test thoroughly.