Performance Scanner AI

What It Does

Installation

pip install openai pygments

Usage

python performance_scanner_ai.py <path_to_python_file>

python performance_scanner_ai.py example.py

Source Code

import json
import os
from pygments import lexers
from pygments.token import Token
from openai import ChatCompletion
import argparse

def analyze_code(code: str) -> dict:
    """
    Analyze the given Python code for performance bottlenecks using AI.

    Args:
        code (str): Python code to analyze.

    Returns:
        dict: JSON-like dictionary containing analysis results.
    """
    try:
        # Prepare the prompt for AI analysis
        prompt = (
            "Analyze the following Python code for performance bottlenecks and suggest optimizations:\n" + code
        )

        # Call OpenAI API (mocked in tests)
        response = ChatCompletion.create(
            model="gpt-4",
            messages=[
                {"role": "system", "content": "You are a Python performance expert."},
                {"role": "user", "content": prompt}
            ]
        )

        suggestions = response['choices'][0]['message']['content']
        return {"code": code, "analysis": suggestions}

    except Exception as e:
        return {"error": str(e)}

def scan_code(input_path: str) -> dict:
    """
    Scan a Python file for performance bottlenecks.

    Args:
        input_path (str): Path to the Python file.

    Returns:
        dict: JSON-like dictionary containing analysis results.
    """
    if not os.path.isfile(input_path):
        return {"error": "File not found."}

    try:
        with open(input_path, "r") as file:
            code = file.read()
        return analyze_code(code)
    except Exception as e:
        return {"error": str(e)}

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Performance Scanner AI")
    parser.add_argument("input_path", type=str, help="Path to the Python file to scan.")
    args = parser.parse_args()

    result = scan_code(args.input_path)
    print(json.dumps(result, indent=4))