testcase-data/Ours/util3.py

# import json

titles = [
    "屠龙勇士",
    "钥匙",
    "众数",
    "种花",
    "路径交点",
    "策略游戏",
    "滚榜",
    "遗失的赋值",
    "天天爱打卡",
    "词典",
    "序列变换",
    "动物园",
    "括号序列",
    "双序列拓展",
    "假期计划",
    "对称二叉树",
    "摆渡车",
    "回家路线",
    "炸弹攻击",
    "种树",
    "三值逻辑",
    "脑洞治疗仪",
    "消消乐",
    "机器人游戏",
    "移除石子",
    "函数调用",
    "结构体",
    "格雷码",
    "数据传输",
    "密码锁",
    "Secret Message 秘密信息",
    "儒略日",
    "L 语言",
    "填树",
    "轻重边",
    "喵了个喵",
    "Gem Island 2",
    "合并书本",
    "桂花树",
    "生成树求和 加强版",
    "冒泡排序",
    "方格染色",
    "多折较差验证",
    "迷失的地球",
    "Old Problem",
    "染色游戏",
    "火锅盛宴",
    "复读机",
    "贪吃蛇",
    "集合",
    "Emiya 家今天的饭",
    "星战",
    "微信步数",
    "影魔",
    "三叉神经树",
    "货币系统",
    "贸易",
    "诸神眷顾的幻想乡",
    "生成随机数",
    "奇怪的背包",
    "季风",
    "Sleeping Cows",
    "蚯蚓",
    "回文子串",
    "预处理器",
    "毒瘤题加强版再加强版",
    "计算几何瞎暴力",
    "迷宫守卫",
    "Set",
    "Paired Up",
    "最长待机",
    "庆典",
    "玄武密码",
    "补退选",
    "字符串匹配",
    "毒瘤题加强版",
    "卡牌"
]


# def remove_freopen_lines(code):
#     lines = code.splitlines()
#     cleaned = [line for line in lines if 'freopen' not in line]
#     return '\n'.join(cleaned)

# # 读取 JSON 文件（文件内容是数组）
# with open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v3.json", "r", encoding="utf-8") as f:
#     data = json.load(f)

# # 遍历每个对象的 tcb_id 和 solutions
# for item in data:
#     if item['tcb_id'] in titles and isinstance(item['solutions'], list):
#         cleaned_solutions = []
#         for code in item['solutions']:
#             if isinstance(code, str):
#                 cleaned_code = remove_freopen_lines(code)
#                 cleaned_solutions.append(cleaned_code)
#             else:
#                 cleaned_solutions.append(code)  # 保留非字符串项原样
#         item['solutions'] = cleaned_solutions

# # 可选：保存到新文件
# with open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v4.json", "w", encoding="utf-8") as f:
#     json.dump(data, f, ensure_ascii=False, indent=2)

# print("处理完成：已移除含有 freopen 的行。")

import json


def load_json(file_path):
    with open(file_path, "r", encoding="utf-8") as f:
        return json.load(f)

def compare_dicts(tcb_id, dict1, dict2):
    keys = set(dict1.keys()) | set(dict2.keys())
    differences = []
    for key in keys:
        val1 = dict1.get(key)
        val2 = dict2.get(key)
        if val1 != val2:
            differences.append((key, val1, val2))
    return differences

# 加载两个文件
data1 = load_json("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v4.json")
data2 = load_json("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench.json")

# 构建 tcb_id -> 对象 映射
map1 = {item['tcb_id']: item for item in data1}
map2 = {item['tcb_id']: item for item in data2}

# 所有 tcb_id
all_tcb_ids = set(map1.keys()) | set(map2.keys())

# 开始对比
for tcb_id in sorted(all_tcb_ids):
    item1 = map1.get(tcb_id)
    item2 = map2.get(tcb_id)

    if item1 and not item2:
        print(f"🟦 tcb_id {tcb_id} 仅在 file1.json 中存在")
    elif item2 and not item1:
        print(f"🟥 tcb_id {tcb_id} 仅在 file2.json 中存在")
    else:
        diffs = compare_dicts(tcb_id, item1, item2)
        if diffs:
            print(f"\n⚠️ tcb_id {tcb_id} 存在键值差异:")
            # for key, v1, v2 in diffs:
            #     print(f"  🔸 字段 '{key}' 不同：")
                # print(f"     file1: {repr(v1)}")
                # print(f"     file2: {repr(v2)}")