testcase-data/Ours/update_correct_code.py

import json
import re
from collections import defaultdict

def clean_key(original_key):
    tcb_id = original_key.replace('/', ' ')
    tcb_id = re.sub(r'^[^.]*\.', '', tcb_id)
    i = tcb_id.find('」')
    if i != -1:
        tcb_id = tcb_id[i+1:]

    tcb_id = tcb_id.strip()
    return tcb_id

import datetime
def write_log(message: str, log_file: str = "log-predo.txt"):
    """
    Append a timestamped log message to a log file.

    Args:
        message (str): The message to log.
        log_file (str): The path to the log file (default is 'log.txt').

    Returns:
        None
    """
    timestamp = datetime.datetime.now().strftime("[%Y-%m-%d %H:%M:%S]")
    with open(log_file, "a", encoding="utf-8") as f:
        f.write(f"{timestamp} {message}\n")


to_update = {
    "取石子游戏": [2],
    "火星探险": [2],
    "数列区间最大值": [0, 1],
    "简单算术": [0],
    "方格取数": [0],
    "路径交点": [1],
    "JOJO": [0],
    "Fibonacci 第 n 项": [0],
    "最小瓶颈路": [1],
    "智力测验 Intelligence Test": [0],
    "Fibonacci 前 n 项和": [0],
    "炸弹攻击": [0],
    "Cowntact Tracing": [2],
    "矩阵游戏": [2],
    "密码锁": [0],
    "最小瓶颈路（加强版）": [2],
    "最小生成树": [1],
    "数列分块入门 4": [2],
    "斐波那契": [2],
    "反色游戏": [2],
    "矩阵乘法": [2],
    "混合果汁": [2],
    "骑士精神": [0],
    "旅行": [1],
    "小 Z 的房间": [2],
    "积木 Blocks": [1],
    "驾驶员 Pilots": [0],
    "Circus": [0],
    "世界树": [1],
    "Push a Box": [2],
    "取石子游戏 2": [2],
    "臭虫集成电路公司": [0],
    "PATH": [1],
    "奇怪的计算器": [1],
    "工会 Guilds": [0],
    "如果奇迹有颜色": [0],
    "棋盘": [0],
    "有意义的字符串": [1],
    "EntropyIncreaser 与金字塔": [0],
    "小 Y 和二叉树": [0],
    "简单的函数": [0],
    "Spaceship": [2],
    "一棵树": [2],
    "微信步数": [2],
    "影魔": [0],
    "火星人": [2],
    "最长树链": [0],
    "Odd": [0],
    "题": [0],
    "Sleeping in Class": [1],
    "办公楼 Offices": [0],
    "分手是祝愿": [1],
    "拉格朗日插值": [1],
    "树链剖分": [0],
    "守卫": [2],
    "吃": [0],
    "这是一道简单的数学题": [0],
    "高精度进制转换": [0, 1],
    "树上操作": [1]
}



log_file = "/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/log-update_correct.txt"
tcb_bench = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v12-update.json", "r", encoding="utf-8"))
# tcb_bench = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/ADD_data_final.json", "r", encoding="utf-8"))

correct_code_base = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/correct_code-v2.json", "r", encoding="utf-8"))

list_no_double = []
correct_code_dict = {}
double_id = []

## 添加 无重复的题面
# for k, v in correct_code.items():
#     tcb_id = clean_key(k)
#     if tcb_id in double_id:
#         continue

#     if tcb_id in list_no_double:
#         correct_code_dict.pop(tcb_id)
#         double_id.append(tcb_id)
#         write_log(f"{tcb_id} double_id", log_file)
#         continue
    
#     correct_code_dict[tcb_id] = v
#     list_no_double.append(tcb_id)


print(double_id)

## 现在更新Correct Code
for item in tcb_bench:
    tcb_id = item['wrong_code'][0]['problem']

    if tcb_id not in correct_code_base.keys():
        write_log(f"{tcb_id} no_update", log_file)
        continue
    correct_codes = correct_code_base[tcb_id]
    if len(correct_codes) >= 3:
        item['solutions'] = correct_codes[:3]
    else:
        for i, code in enumerate(correct_codes):
            item['solutions'][i] = code

json.dump(tcb_bench, open(f"/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v13.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False)

# json.dump(tcb_bench, open(f"/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-batch8.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False)

# tcb_bench_v11 = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v11-temp.json", "r", encoding="utf-8"))
# tcb_bench_v12 = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v12.json", "r", encoding="gbk"))

# ans_dict = {}
# for item in tcb_bench_v11:
#     ans_dict[item['tcb_id']] = item

# for item in tcb_bench_v12:
#     tcb_id = item['tcb_id']
#     if tcb_id not in ans_dict.keys():
#         print(tcb_id)
#         continue
#     item['solutions'] = ans_dict[item['tcb_id']]['solutions']

# json.dump(tcb_bench_v12, open(f"/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v12-update.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False)