pyfuncmonitor 0.1.2

A Python decorator for comprehensive function monitoring with execution timing, memory usage tracking, CPU monitoring, input/output validation, and structured logging.

PyFuncMonitor

A Python decorator for comprehensive function monitoring with execution timing, memory usage tracking, CPU monitoring, input/output validation, and structured logging.

Features

• Execution Monitoring: Track function execution time, memory usage, and CPU utilization
• Input/Output Validation: Automatic validation using Pydantic models and type hints
• Structured Logging: Configurable structured logging with support for file output
• Error Handling: Comprehensive exception handling and error reporting
• Flexible Configuration: Global and per-function configuration options
• Production Ready: Designed for production use with proper error handling and performance considerations

Installation

pip install pyfuncmonitor

Quick Start

from pyfuncmonitor import monitor_function

@monitor_function()
def add_numbers(a: int, b: int) -> int:
    return a + b

result = add_numbers(5, 3)
print(result)

This will output a dictionary containing the function result, execution metrics, and monitoring data:

{
    'result': 8,
    'status': 'success',
    'errors': None,
    'execution_time': 0.0001,
    'memory_usage': {
        'before': 15728640,
        'after': 15728640,
        'peak': 15728640,
        'delta': 0
    },
    'cpu_usage': 0.0,
    'timestamp': '2024-01-15T10:30:45.123456',
    'function_name': 'add_numbers'
}

Configuration

Global Configuration

Configure the monitor globally for your application:

from pyfuncmonitor import configure_monitor

configure_monitor(
    log_to_file=True,
    log_file_path="./app_monitor.log",
    log_level=20,  # INFO level
    default_return_raw_result=True
)

Environment Variables

You can also configure using environment variables:

export PYFUNCMONITOR_LOG_LEVEL=10  # DEBUG level
export PYFUNCMONITOR_LOG_TO_FILE=true
export PYFUNCMONITOR_LOG_FILE=./debug.log

Per-Function Configuration

Override global settings for specific functions:

@monitor_function(
    validate_input=True,
    validate_output=False,
    log_level="DEBUG",
    return_raw_result=True
)
def my_function(x: int) -> str:
    return str(x * 2)

Advanced Usage

Input/Output Validation with Pydantic

from pydantic import BaseModel
from pyfuncmonitor import monitor_function

class User(BaseModel):
    name: str
    age: int
    email: str

class UserResponse(BaseModel):
    user: User
    message: str

@monitor_function(validate_input=True, validate_output=True)
def create_user(user_data: User) -> UserResponse:
    return UserResponse(
        user=user_data,
        message=f"User {user_data.name} created successfully"
    )

# Usage
user = User(name="John Doe", age=30, email="[e-mail]")
result = create_user(user)

Custom Monitor Instances

from pyfuncmonitor import FunctionMonitor

# Create custom monitor with specific settings
production_monitor = FunctionMonitor(
    validate_input=True,
    validate_output=True,
    log_execution=True,
    return_raw_result=True
)

debug_monitor = FunctionMonitor(
    validate_input=True,
    validate_output=False,
    log_level="DEBUG",
    return_raw_result=False
)

@production_monitor
def critical_function(data: dict) -> dict:
    # Process critical data
    return {"processed": True}

@debug_monitor
def experimental_function(x: int) -> int:
    # Experimental code
    return x ** 2

Error Handling

The monitor automatically captures and reports errors:

@monitor_function()
def divide_numbers(a: float, b: float) -> float:
    if b == 0:
        raise ValueError("Cannot divide by zero")
    return a / b

result = divide_numbers(10, 0)
print(result["status"])  # "error"
print(result["errors"])  # List of error messages

Memory and CPU Monitoring

Monitor resource usage:

@monitor_function(
    enable_memory_monitoring=True,
    enable_cpu_monitoring=True
)
def memory_intensive_function(size: int) -> list:
    # Create a large list
    return list(range(size))

result = memory_intensive_function(1000000)
print(result["memory_usage"])  # Memory usage statistics
print(result["cpu_usage"])     # CPU usage percentage

Configuration Options

Global Configuration Parameters

• log_level: Logging level (default: INFO)
• log_to_file: Enable file logging (default: False)
• log_file_path: Path to log file (default: "./pyfuncmonitor.log")
• log_file_max_size: Maximum log file size in bytes (default: 10MB)
• log_file_backup_count: Number of backup log files (default: 5)
• default_validate_input: Default input validation setting (default: True)
• default_validate_output: Default output validation setting (default: True)
• default_log_execution: Default execution logging setting (default: True)
• default_return_raw_result: Default return format setting (default: False)
• enable_memory_monitoring: Enable memory monitoring (default: True)
• enable_cpu_monitoring: Enable CPU monitoring (default: True)

Decorator Parameters

• validate_input: Enable input validation using type hints
• validate_output: Enable output validation using type hints
• log_execution: Enable structured logging
• log_level: Log level for the function ("DEBUG", "INFO", "WARNING", "ERROR")
• return_raw_result: Return original result on success, structured format on error
• enable_memory_monitoring: Enable memory usage monitoring
• enable_cpu_monitoring: Enable CPU usage monitoring

Logging

The monitor uses structured logging with configurable output formats:

Console Logging (Development)

configure_monitor(log_to_file=False)  # Logs to console with readable format

File Logging (Production)

configure_monitor(
    log_to_file=True,
    log_file_path="/var/log/myapp/monitor.log"
)  # Logs to file in JSON format

Custom Log Levels

@monitor_function(log_level="DEBUG")
def debug_function():
    pass

@monitor_function(log_level="ERROR")  # Only log errors
def critical_function():
    pass

Best Practices

1. Production Usage: Use return_raw_result=True in production to avoid overhead
2. Input Validation: Enable input validation for external-facing functions
3. Log File Management: Configure log rotation to prevent disk space issues
4. Selective Monitoring: Disable CPU/memory monitoring for high-frequency functions if needed
5. Error Handling: Always handle the case where monitoring might return error status

# Production configuration example
configure_monitor(
    log_to_file=True,
    log_file_path="/var/log/myapp/pyfuncmonitor.log",
    log_level=20,  # INFO
    default_return_raw_result=True,
    default_validate_input=True,
    default_validate_output=False  # Disable output validation for performance
)

@monitor_function()
def api_endpoint(request_data: RequestModel) -> ResponseModel:
    # Your business logic here
    return process_request(request_data)

# Usage with error handling
result = api_endpoint(request)
if isinstance(result, dict) and result.get("status") == "error":
    # Handle error case
    logger.error("Function failed", errors=result["errors"])
    return error_response()
else:
    # Success case - result is the actual return value
    return result

Performance Considerations

• Minimal Overhead: The monitor is designed to have minimal impact on function performance
• Selective Monitoring: Disable features you don't need for better performance
• Memory Monitoring: Uses psutil for accurate memory measurements
• CPU Monitoring: Lightweight CPU usage tracking
• Logging: Structured logging with configurable levels to reduce I/O overhead

Error Handling and Debugging

Common Issues

1. Import Errors: Ensure all dependencies are installed
2. Permission Issues: Check file permissions for log files
3. Memory Monitoring: Requires psutil package
4. Type Validation: Requires proper type hints for validation to work

Debug Mode

Enable debug mode for detailed logging:

from pyfuncmonitor import configure_monitor

configure_monitor(
    log_level=10,  # DEBUG level
    log_to_file=True,
    log_file_path="./debug.log"
)

License

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

Support

• Issues: GitHub Issues
• PyPI: PyPI Package