ai/code_fingerprint.py

# ai/code_fingerprint.py
"""
Analyzes user code to extract behavioral coding patterns (code fingerprint).
These reflect how a person logically structures programs.
"""

import ast
from typing import Dict, Any

def analyze_fingerprint(code: str) -> Dict[str, Any]:
    try:
        tree = ast.parse(code)
    except Exception:
        return {
            "abstraction": 0.0,
            "data_structure_pref": "unknown",
            "control_density": 0.0,
            "refactor_tendency": 0.5
        }

    loops, conditionals, func_defs, classes = 0, 0, 0, 0
    data_structures = {"list": 0, "dict": 0, "set": 0, "tuple": 0}

    for node in ast.walk(tree):
        if isinstance(node, (ast.For, ast.While)):
            loops += 1
        elif isinstance(node, ast.If):
            conditionals += 1
        elif isinstance(node, ast.FunctionDef):
            func_defs += 1
        elif isinstance(node, ast.ClassDef):
            classes += 1
        elif isinstance(node, ast.List):
            data_structures["list"] += 1
        elif isinstance(node, ast.Dict):
            data_structures["dict"] += 1
        elif isinstance(node, ast.Set):
            data_structures["set"] += 1
        elif isinstance(node, ast.Tuple):
            data_structures["tuple"] += 1

    total_nodes = max(1, len(list(ast.walk(tree))))
    control_density = (loops + conditionals) / total_nodes
    abstraction = (classes + func_defs) / total_nodes

    data_pref = max(data_structures, key=data_structures.get, default="unknown")

    return {
        "abstraction": round(abstraction, 2),
        "data_structure_pref": data_pref,
        "control_density": round(control_density, 2),
        "refactor_tendency": 0.5 + (abstraction * 0.5)
    }