testcase-data/Ours/get_worry_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

# def process_and_dedup_json(filepath):
#     with open(filepath, 'r', encoding='utf-8') as f:
#         data = json.load(f)

#     deduped_data = {}
#     key_count = defaultdict(int)
#     repeated_keys = set()

#     for key, value in data.items():
#         cleaned_key = clean_key(key)
#         key_count[cleaned_key] += 1
#         if cleaned_key not in deduped_data:
#             deduped_data[cleaned_key] = value
#         else:
#             repeated_keys.add(cleaned_key)

#     return deduped_data, sorted(repeated_keys)

# # 示例使用
# input_path = '/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/final_v1.0.json'
# output_path = '/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/final_v2.0.json'

# deduped_json, repeated_ids = process_and_dedup_json(input_path)

# # 写入去重后的 JSON 文件
# with open(output_path, 'w', encoding='utf-8') as f:
#     json.dump(deduped_json, f, ensure_ascii=False, indent=2)


# add_keys = ['wrong_code', 'timeLimit', 'memoryLimit', 'rank', 'n_rows', 'balance_var', 'balance_column_sum']

# testcase_bench = []
# with open('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench.json', 'r', encoding='utf-8') as f:
#         testcase_bench = json.load(f)

# for item in testcase_bench:
#     tcb_id = item['tcb_id']
#     if tcb_id in deduped_json:
#         for key in add_keys:
#             item[key] = deduped_json[tcb_id][key]
#     else:
#         print(tcb_id)

# with open('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v2.json', 'w', encoding='utf-8') as f:
#     json.dump(testcase_bench, f, ensure_ascii=False, indent=4)

# import json

# def remove_aa_bb_keys(input_file, output_file):
#     with open(input_file, 'r', encoding='utf-8') as f:
#         data = json.load(f)

#     # 如果 data 是列表，处理每个元素
#         #         "timeLimit": 1500,
#         # "memoryLimit": 512,
#     if isinstance(data, list):
#         for item in data:
#             item.pop('timeLimit', None)
#             item.pop('memoryLimit', None)
#     else:
#         raise ValueError("JSON 须为数组形式，每个元素是一个对象")

#     # 保存结果
#     print(f"共处理 {len(data)} 条记录。")
#     return data

    

# 示例用法
# data = remove_aa_bb_keys('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v2.json', '/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v3.json')

# for item in data:
#     for w_code_item in item['wrong_code']:
#         w_code_item['problem'] = item['tcb_id']

# with open('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v3.json', 'w', encoding='utf-8') as f:
#     json.dump(data, f, ensure_ascii=False, indent=4)


# with open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/balance_v3.0.dev_with_code.json", 'r', encoding='utf-8') as f:
#     worry_code = json.load(f)

# with open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v4.json", 'r', encoding='utf-8') as f:
#     tcb_bench = json.load(f)

# worry_code_dict = {}
# for k, v in worry_code.items():
#     tcb_id = clean_key(k)
#     worry_code_dict[tcb_id] = v

# for item in tcb_bench:
#     if "地底蔷薇" in item['tcb_id']:
#         worry_codes = worry_code_dict[""]
#     else:
#         worry_codes = worry_code_dict[item['tcb_id']]
#     item['wrong_code'] = worry_codes

# with open('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v5.json', 'w', encoding='utf-8') as f:
#     json.dump(tcb_bench, f, ensure_ascii=False, indent=6)

with open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v5.json", 'r', encoding='utf-8') as f:
    tcb_bench = json.load(f)

def remove_freopen_lines(code_str: str) -> str:
    """
    移除包含 'freopen' 的代码行
    :param code_str: 原始代码字符串
    :return: 处理后的代码字符串
    """
    lines = code_str.splitlines()
    filtered_lines = [line for line in lines if 'freopen' not in line]
    return '\n'.join(filtered_lines)

for item in tcb_bench:
    for code_item in item["wrong_code"]["wrong_code"]:
        code_item['code'] = remove_freopen_lines(code_item['code'])

with open('/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v6.json', 'w', encoding='utf-8') as f:
    json.dump(tcb_bench, f, ensure_ascii=False, indent=6)