import json
import re


def extract_and_format_from_paths(json_paths, includes, excludes):
    """
    从多个 JSON 文件路径读取数据，提取包含特定关键词的内容，并按照格式要求合并。

    参数:
    json_paths (list): 包含多个 JSON 文件路径的列表。
    includes (list): 包含要检查的关键词的列表。
    excludes (list): 包含要排除的关键词的列表。

    返回:
    list: 包含所有文件中满足条件的格式化字符串列表。
    """
    all_formatted_results = []

    # 遍历每个文件路径
    for path in json_paths:
        try:
            with open(path, 'r', encoding='utf-8') as file:
                # 加载 JSON 数据
                json_data = json.load(file)
                formatted_results = []

                # 遍历 JSON 数据的每个键值对
                for key, value in json_data.items():
                    if isinstance(value, dict):
                        # 如果值是字典，检查嵌套字典的每个键值对
                        for sub_key, sub_value in value.items():
                            if any(include in sub_key for include in includes):
                                # 如果子值包含关键词，格式化并添加到结果列表
                                formatted_results.append(f"{sub_value}")
                    elif isinstance(value, str):  # clause
                        # 检查是否包含任何 include 关键词
                        for include in includes:
                            if include in value:
                                # 找到 include 之前的内容
                                prefix = value.split(include)[0]
                                # 检查 prefix 是否不包含任何 exclude 关键词
                                if not any(exclude in prefix for exclude in excludes):
                                    # 如果不包含任何 exclude 关键词，添加整个 value 到结果列表
                                    if '\n' in value:
                                        value = value.split('\n', 1)[-1]
                                    formatted_results.append(value)
                                    break  # 找到一个符合条件的就跳出循环

                # 将当前文件的结果添加到总结果列表
                all_formatted_results.extend(formatted_results)
        except FileNotFoundError:
            print(f"Error: The file '{path}' does not exist.")
        except json.JSONDecodeError:
            print(f"Error: The file '{path}' contains invalid JSON.")

    return all_formatted_results


def extract_unique_items_from_texts(texts):
    # 更新正则表达式以包括更广泛的序号类型，包括中文序号
    pattern = re.compile(r'(?:\d+\.|\（\d+\）|\(\d+\)|\d+\)|\①|\②|\③|\④|\⑤|\⑥|\⑦|\⑧|\⑨|\⑩|\⑪|\⑫)\s*')
    intro_pattern = re.compile(r'^.*?[:：]')
    punctuation_pattern = re.compile(r'[；。，、．.,:;!?！？]+$')
    url_pattern = re.compile(r'http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]|[!*\\(\\),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+')
    content_check_pattern = re.compile(r'[\u4e00-\u9fa5a-zA-Z0-9]{2,}')  # 检查是否含有至少2个连续的字母、数字或汉字

    all_results = []
    seen = set()

    for text in texts:
        # 去除文本中的制表符和换行符
        text = text.replace('\t', '').replace('\n', '')

        # 删除引导性的文本（直到冒号，但保留冒号后的内容）
        text = intro_pattern.sub('', text)

        # 替换URL为占位符，并保存URL以便后续还原
        urls = []
        def url_replacer(match):
            urls.append(match.group(0))
            return f"{{URL{len(urls)}}}"
        text = url_pattern.sub(url_replacer, text)

        # 使用数字和括号的模式分割文本
        items = pattern.split(text)

        for item in items:
            cleaned_item = item.strip()
            if cleaned_item:
                # 进一步清理每个条目
                cleaned_item = pattern.sub('', cleaned_item)
                cleaned_item = punctuation_pattern.sub('', cleaned_item)
                cleaned_item = cleaned_item.strip()

                # 还原URL
                for i, url in enumerate(urls, 1):
                    cleaned_item = cleaned_item.replace(f"{{URL{i}}}", url)

                # 添加未见过的独特条目，确保它包含足够的实质内容并长度大于3个字符
                if cleaned_item and cleaned_item not in seen and len(cleaned_item) > 3 and content_check_pattern.search(cleaned_item):
                    seen.add(cleaned_item)
                    all_results.append(cleaned_item)

    return all_results
# 使用上面定义的函数
truncate_json_path = "C:\\Users\\Administrator\\Desktop\\fsdownload\\796f7bb3-2f7a-4332-b044-9d817a07861e\\truncate_output.json"
clause_path = "C:\\Users\\Administrator\\Desktop\\fsdownload\\796f7bb3-2f7a-4332-b044-9d817a07861e\\clause1.json"
json_paths = [truncate_json_path,clause_path]  # 根据实际存放的路径来填写
includes = ["不得存在","不得与","禁止投标","对投标人的纪律"]
excludes=["招标","评标","定标"]
# 调用函数
results = extract_and_format_from_paths(json_paths, includes,excludes)
print(results)
res=extract_unique_items_from_texts(results)
print(res)