testcase-data/Ours/update_match_wrong_code.py

# import json
# import re
# from collections import defaultdict

# def get_transformed_id(problem_id):
#     if "地底蔷薇" in problem_id:
#         return "地底蔷薇"
#     tcb_id = problem_id.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()
#     if tcb_id == "":
#         print(problem_id)
#         return problem_id
#     return tcb_id

# new_wrong_code_base = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/balance_v1.0.dev_with_code.json", "r", encoding="utf-8"))
# tcb_bench = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v9.json", "r", encoding="utf-8"))

# new_wrong_code_base_dict = {}
# update_list = {}
# for k, v in new_wrong_code_base.items():
#     tcb_id = get_transformed_id(k)
#     if tcb_id not in new_wrong_code_base_dict.keys():
#         update_list[tcb_id] = [
#             {k:v}
#         ]
#     else:
#         update_list[tcb_id].append({k:v})
    
#     new_wrong_code_base_dict[tcb_id] = v

# multiple_list = {}
# for k, v in update_list.items():
#     if len(v) == 1:
#         continue
#     for item in v:
#         multiple_list.update(item)

# json.dump(multiple_list, open(f"/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/to_update.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-v10.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False)

wrong_match = {
    "重建": "#6075. 「2017 山东一轮集训 Day6」重建",
    "第一题": "#6065. 「2017 山东一轮集训 Day3」第一题",
    "选课": "#3331. 「WC2020」选课",
    "字符串": "#6031. 「雅礼集训 2017 Day1」字符串",
    "Censoring": "#10059. 「一本通 2.4 练习 2」Censoring"
}

import json
new_wrong_code_base = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/balance_v1.0.dev_with_code.json", "r", encoding="utf-8"))
wrong_code_base = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/balance_v2.0_with_code.json", "r", encoding="utf-8"))
tcb_bench = json.load(open("/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v10.json", "r", encoding="utf-8"))

for item in tcb_bench:
    if item['tcb_id'] in wrong_match.keys():
        item['wrong_code'] = new_wrong_code_base[wrong_match[item['tcb_id']]]['cowrong_codes']


json.dump(tcb_bench, open(f"/home/i-luoxianzhen/data/TestCase-Gen/data/Ours/TestcaseBench-v11.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False)