Function format_evaluation_result

tests/leetcode_reflection_agent.py:389–421

格式化评测结果为易于模型理解的字符串

(eval_result: Dict[str, Any])

def format_evaluation_result(eval_result: Dict[str, Any]) -> str:
    """格式化评测结果为易于模型理解的字符串"""
    status = eval_result.get("status", "Unknown")
    passed = eval_result.get("passed_cases", 0)
    total = eval_result.get("total_cases", 0)
    
    result_text = f"Status: {status}\n"
    result_text += f"Passed Cases: {passed}/{total}\n"
    
    if status == "Accepted":
        runtime = eval_result.get("runtime", 0)
        memory = eval_result.get("memory_usage", 0)
        runtime_beats = eval_result.get("runtime_beats", 0)
        memory_beats = eval_result.get("memory_beats", 0)
        result_text += f"Runtime: {runtime}ms (beats {runtime_beats}%)\n"
        result_text += f"Memory: {memory}MB (beats {memory_beats}%)\n"
    elif status == "Wrong Answer":
        result_text += "The solution produced incorrect output for some test cases.\n"
    elif status == "Time Limit Exceeded":
        result_text += "The solution is too slow. Consider optimizing the algorithm.\n"
    elif status == "Runtime Error":
        result_text += "The solution crashed during execution. Check for index errors, null pointers, etc.\n"
    elif status == "Memory Limit Exceeded":
        result_text += "The solution uses too much memory. Consider using more memory-efficient data structures.\n"
    elif status == "Compile Error":
        result_text += "The code failed to compile. Check for syntax errors.\n"
    
    if "details_text" in eval_result:
        # 只取前200字符的details
        details = eval_result["details_text"][:200]
        result_text += f"Details: {details}...\n"
    
    return result_text


async def generate_initial_solution(