import re
import os
import time
from PyPDF2 import PdfReader, PdfWriter
from flask_app.general.clean_pdf import clean_page_content, extract_common_header
from flask_app.general.merge_pdfs import merge_selected_pdfs_for_engineering
import concurrent.futures
import logging


def get_global_logger(unique_id):
    if unique_id is None:
        return logging.getLogger()  # 获取默认的日志器
    logger = logging.getLogger(unique_id)
    return logger


logger = None
def save_pages_to_new_pdf(pdf_path, output_folder, output_suffix, start_page, end_page, common_header):
    try:
        # 获取文件基本名称
        base_file_name = os.path.splitext(os.path.basename(pdf_path))[0]

        # 构建主要输出文件路径
        output_pdf_path = os.path.join(output_folder, f"{base_file_name}_{output_suffix}.pdf")
        # 读取PDF文件
        pdf_document = PdfReader(pdf_path)
        total_pages = len(pdf_document.pages)

        # 检查起始和结束页码是否有效
        if start_page < 0 or end_page >= total_pages or start_page > end_page:
            print(f"无效的页面范围: {start_page} 到 {end_page}")
            return ""

        # 如果 output_suffix 是 'notice'，保存 start_page 之前的页面(若遇到'目录'，则截取到'目录')
        if output_suffix == 'notice' and start_page > 0:
            before_pdf_path = os.path.join(output_folder, f"{base_file_name}_before.pdf")
            before_doc = PdfWriter()
            toc_page = -1
            # 查找目录页
            for page_num in range(min(start_page, total_pages)):
                page_text = pdf_document.pages[page_num].extract_text()
                cleaned_text = clean_page_content(page_text, common_header)
                if re.search(r'目\s*录', cleaned_text, re.MULTILINE):
                    toc_page = page_num
                    break

            # 确定截取的页数
            pages_to_extract = toc_page + 1 if toc_page != -1 else start_page
            # 提取页面
            for page_num in range(pages_to_extract):
                before_doc.add_page(pdf_document.pages[page_num])
            print(before_pdf_path)
            with open(before_pdf_path, 'wb') as f_before:
                before_doc.write(f_before)

            # print(f"已保存页面从 0 到 {pages_to_extract} 为 {before_pdf_path}")

        # 提取指定范围的页面
        output_doc = PdfWriter()
        for page_num in range(start_page, end_page + 1):
            output_doc.add_page(pdf_document.pages[page_num])

        # 保存新的PDF文件
        with open(output_pdf_path, 'wb') as f_output:
            output_doc.write(f_output)
        print(f"{output_suffix} 已截取并保存页面从 {start_page + 1} 到 {end_page + 1} 为 {output_pdf_path}")
        return output_pdf_path

    except Exception as e:
        print(f"Error in save_pages_to_new_pdf: {e}")
        return ""  # 返回空字符串


def extract_pages_tobidders_notice(pdf_path, output_folder, begin_pattern, begin_page, common_header,
                                   is_secondary_match):
    pdf_document = PdfReader(pdf_path)
    exclusion_pattern = re.compile(r'文件的构成|文件的组成|须对应|需对应|须按照|需按照|须根据|需根据|文件组成|文件构成')

    def run_extraction():
        start_page = None
        mid_page = None
        end_page = None
        chapter_type = None  # 用于存储“章”或“部分”

        for i, page in enumerate(pdf_document.pages):
            text = page.extract_text() or ""
            cleaned_text = clean_page_content(text, common_header)

            if exclusion_pattern and re.search(exclusion_pattern, cleaned_text) and mid_page is not None:
                continue

            if start_page is None:
                match = re.search(begin_pattern, cleaned_text)
                if match and i > begin_page:
                    start_page = i
                    matched_text = match.group(0)  # 获取整个匹配的文本
                    if '章' in matched_text:
                        chapter_type = '章'
                    elif '部分' in matched_text:
                        chapter_type = '部分'
                    else:
                        chapter_type = None  # 未匹配到“章”或“部分”

                    if chapter_type:
                        # 根据 chapter_type 动态生成 end_pattern
                        if not is_secondary_match:
                            end_pattern = re.compile(
                                rf'^第[一二三四五六七八九十百千]+?(?:{chapter_type})\s*[\u4e00-\u9fff]+|'
                                r'^评标办法前附表|'
                                r'^附录(?:一)?[：:]|'
                                r'^附件(?:一)?[：:]|'
                                r'^附表(?:一)?[：:]',
                                re.MULTILINE
                            )
                        else:
                            end_pattern = re.compile(
                                rf'第[一二三四五六七八九十百千]+?(?:{chapter_type})\s*[\u4e00 -\u9fff]+\s*$'
                            )
                        # 根据 chapter_type 动态生成 additional_mid_pattern
                        if chapter_type == '章':
                            additional_mid_pattern = r'^第[一二三四五六七八九十百千]+?(?:部分)'
                        elif chapter_type == '部分':
                            additional_mid_pattern = r'^第[一二三四五六七八九十百千]+?(?:章)'
                        else:
                            additional_mid_pattern = ''

                        # 定义基础的 mid_pattern
                        base_mid_pattern = (
                            r'^\s*(?:[（(]\s*[一1]?\s*[)）]\s*[、．.]*|[一1][、．.]+|[、．.]+)\s*(说\s*明|总\s*则)'
                            r'|^((投标人?|磋商|供应商|谈判供应商|磋商供应商)\s*须知正文部分)'
                        )
                        # 合并基础模式和额外模式
                        if additional_mid_pattern:
                            combined_mid_pattern = re.compile(
                                rf'(?:{base_mid_pattern})|(?:{additional_mid_pattern})',
                                re.MULTILINE
                            )
                        else:
                            combined_mid_pattern = re.compile(
                                rf'{base_mid_pattern}',
                                re.MULTILINE
                            )
                        # print(f"生成的 combined_mid_pattern: {combined_mid_pattern.pattern}")  # 打印 combined_mid_pattern
                    else:
                        # 如果未匹配到“章”或“部分”，使用默认的 end_pattern 和 mid_pattern
                        if not is_secondary_match:
                            end_pattern = re.compile(
                                r'^第[一二三四五六七八九十百千]+(?:章|部分)\s*[\u4e00-\u9fff]+|'
                                r'^评标办法前附表|'  # Match "评标办法前附表" at the beginning of a line
                                r'^附录(?:一)?[：:]|'  # Match "附录一" with optional "一" and colon
                                r'^附件(?:一)?[：:]|'  # Match "附件一" with optional "一" and colon
                                r'^附表(?:一)?[：:]',  # Match "附表一" with optional "一" and colon
                                re.MULTILINE
                            )
                        else:
                            end_pattern = re.compile(
                                rf'第[一二三四五六七八九十百千]+(?:章|部分)\s*[\u4e00-\u9fff]+\s*$'
                            )
                        # print(f"使用默认的 end_pattern: {end_pattern.pattern}")  # 打印默认的 end_pattern
                        # 定义基础的 mid_pattern
                        base_mid_pattern = r'^\s*(?:[（(]\s*[一1]?\s*[)）]\s*[、．.]*|[一1][、．.]+|[、．.]+)\s*(说\s*明|总\s*则)'
                        combined_mid_pattern = re.compile(
                            rf'{base_mid_pattern}',
                            re.MULTILINE
                        )
                    continue

            if start_page is not None and mid_page is None and combined_mid_pattern:
                if re.search(combined_mid_pattern, cleaned_text):
                    mid_page = i
            if start_page is not None and mid_page is not None and chapter_type:
                if re.search(end_pattern, cleaned_text):
                    if i > mid_page:
                        end_page = i
                        break

        return start_page, mid_page, end_page

    # 运行提取
    start_page, mid_page, end_page = run_extraction()
    if start_page is None or end_page is None or mid_page is None:
        print(f"first: tobidders_notice 未找到起始或结束页在文件 {pdf_path} 中！尝试备用提取策略。")
        return "", ""

    path1 = save_pages_to_new_pdf(pdf_path, output_folder, "tobidders_notice_table", start_page, mid_page,
                                  common_header)
    path2 = save_pages_to_new_pdf(pdf_path, output_folder, "tobidders_notice", mid_page, end_page, common_header)
    return path1, path2


def extract_pages(pdf_path, output_folder, begin_pattern, begin_page, end_pattern, output_suffix, common_header,
                  is_secondary_match=False):
    # 打开PDF文件
    pdf_document = PdfReader(pdf_path)
    start_page = None
    end_page = None
    exclusion_pattern = re.compile(r'文件的构成|文件的组成|须对应|需对应|须按照|需按照|须根据|需根据|文件组成|文件构成')
    # 遍历文档的每一页，查找开始和结束短语的位置
    for i in range(len(pdf_document.pages)):
        page = pdf_document.pages[i]
        text = page.extract_text()
        if text:
            cleaned_text = clean_page_content(text, common_header)
            # if is_secondary_match and re.search(exclusion_pattern, cleaned_text):  # 跳过投标人须知正文中的"投标文件的组成"
            if re.search(exclusion_pattern, cleaned_text):
                continue
            if re.search(begin_pattern, cleaned_text) and i >= begin_page:
                if start_page and (output_suffix == "notice" or output_suffix == "invalid"):
                    pass
                else:
                    start_page = i
            if start_page is not None and re.search(end_pattern, cleaned_text):
                condition = i > start_page
                if condition:
                    is_invalid_condition = output_suffix == "invalid" and i > 30  # 这边默认无效投标至少有30页
                    if is_invalid_condition or output_suffix != "invalid":
                        end_page = i
                        break  # 找到结束页后退出循环

    # 移除对 extract_pages_twice 的调用
    if start_page is None or end_page is None:
        print(f"{output_suffix} first: 未找到起始或结束页在文件 {pdf_path} 中！")
        return [""]
    else:
        return [save_pages_to_new_pdf(pdf_path, output_folder, output_suffix, start_page, end_page, common_header)]


def extract_pages_twice(pdf_path, output_folder, output_suffix, common_header, last_begin_index):
    """
    处理 PDF 文件，作为 truncate_pdf_main 的后备方法。
    此函数将在所有模式对都失败后调用。
    """
    try:
        pdf_document = PdfReader(pdf_path)
        if output_suffix == "qualification":
            print("twice:qualificaiton!")
            # 动态设置 include_keys
            include_keys = ["资格审查", "资质审查", "符合性审查", "资格性检查", "符合性检查","资格检查","能力","信誉"]
            # 定义起始匹配模式，仅匹配“附录”、“附件”或“附表”
            begin_pattern = r'^(?:附录(?:[一1])?[：:]|附件(?:[一1])?[：:]|附表(?:[一1])?[：:])'
            # 定义结束匹配模式 - 附录、附件、附表等（移除负向前瞻）
            end_pattern_attachment = re.compile(
                r'^(?:附录[一二三四五六七八九1-9]*[：:]|附件[一二三四五六七八九1-9]*[：:]|附表[一二三四五六七八九1-9]*[：:]).*$',
                re.MULTILINE
            )
            # 定义结束匹配模式 - 章节标题、评标附表、投标人须知等
            end_pattern_chapter = re.compile(
                r'第[一二三四五六七八九]+(?:章|部分)\s*[\u4e00-\u9fff、()（）]+\s*$|'
                r'\s*评标(?:办法|方法)前附表\s*$|'
                r'投标人须知',
                re.MULTILINE
            )
            start_page = None
            end_page = None
            # 从指定的开始页开始遍历
            for i, page in enumerate(pdf_document.pages[last_begin_index:], start=last_begin_index):
                text = page.extract_text()
                if text:
                    cleaned_text = clean_page_content(text, common_header)
                    # 确定起始页，需在last_begin_index之后
                    if any(key in cleaned_text for key in include_keys):
                        if re.search(begin_pattern, cleaned_text, re.MULTILINE):
                            if start_page is None:
                                start_page = i  # 确保起始页不小于章节的开始页码
                                continue
                                # print(f"匹配到附录/附件/附表，设置起始页为: {start_page}")
                    # 确定结束页 - 附录、附件、附表等
                    if start_page is not None and end_pattern_attachment.search(cleaned_text):
                        # 额外检查当前页面是否不包含任何 include_keys
                        if not any(key in cleaned_text for key in include_keys):
                            if i > start_page:
                                end_page = i
                                # print(f"找到结束页 (附件类): {end_page}")
                                break  # 找到结束页后退出循环
                        else:
                            print(f"当前页面匹配附件结束模式但包含 include_keys，继续查找。页面: {i}")
                    # 确定结束页 - 章节标题、评标附表、投标人须知等
                    elif start_page is not None and end_pattern_chapter.search(cleaned_text):
                        if i > start_page:
                            end_page = i
                            # print(f"找到结束页 (章节标题等): {end_page}")
                            break  # 找到结束页后退出循环
            if start_page is None or end_page is None:
                print(f"{output_suffix} twice: 未找到起始或结束页在文件 {pdf_path} 中！")
                return []
            else:
                return [save_pages_to_new_pdf(pdf_path, output_folder, output_suffix, start_page, end_page, common_header)]
        elif output_suffix == "invalid":
            pdf_document = PdfReader(pdf_path)
            total_pages = len(pdf_document.pages)
            # 计算总页数的三分之二
            total = int(total_pages * 2 / 3)
            start_page = last_begin_index
            end_page = min(90, total)
            return [save_pages_to_new_pdf(pdf_path, output_folder, output_suffix, start_page, end_page, common_header)]
        else:
            print(f"{output_suffix} twice: 未定义的输出后缀。")
            return []
    except Exception as e:
        print(f"Error in extract_pages_twice: {e}")
        return []


def get_start_and_common_header(input_path):
    common_header = extract_common_header(input_path)
    last_begin_index = 0
    begin_pattern = re.compile(
        r'.*(?:招标公告|投标邀请书|投标邀请函|投标邀请)[\)）]?\s*$',
        re.MULTILINE
    )
    pdf_document = PdfReader(input_path)
    for i, page in enumerate(pdf_document.pages):
        if i > 25:
            return common_header, 0  # 如果页码大于25，直接返回last_begin_index=0
        text = page.extract_text()
        if text:
            cleaned_text = clean_page_content(text, common_header)
            if begin_pattern.search(cleaned_text):
                last_begin_index = i  # 更新第一个匹配的索引，页码从0开始
                return common_header, last_begin_index
    return common_header, last_begin_index


def truncate_pdf_main(input_path, output_folder, selection):
    if os.path.isdir(input_path):
        generated_files = []
        for file_name in os.listdir(input_path):
            if file_name.endswith('.pdf'):
                file_path = os.path.join(input_path, file_name)
                files = truncate_pdf_main(file_path, output_folder, selection)
                if files:
                    generated_files.extend(files)
        return generated_files
    elif os.path.isfile(input_path) and input_path.endswith('.pdf'):
        # base_file_name = os.path.splitext(os.path.basename(input_path))[0]
        common_header, last_begin_index = get_start_and_common_header(input_path)
        # print(last_begin_index)
        if selection == 1:
            # Selection 1: 投标人须知前附表
            pattern_pairs = [
                (
                    re.compile(
                        r'(?:第[一二三四五六七八九十百千]+(?:章|部分)\s*(?:投标人?|磋商|供应商|谈判供应商|磋商供应商)须知+|(?:一\s*、\s*)?(?:投标人?|磋商|供应商)须知前附表)'),
                    re.compile(
                        r'第[一二三四五六七八九十]+(?:章|部分)\s*[\u4e00-\u9fff]+|^评标办法前附表|^附录(?:一)?[：:]|^附件(?:一)?[：:]|^附表(?:一)?[：:]',
                        re.MULTILINE)
                ),
                (
                    re.compile(
                        r'.*(?:投标人?|磋商|供应商|谈判供应商|磋商供应商)须知\s*$|(?:一\s*、\s*)?(?:投标人?|磋商|供应商)须知前附表',
                        re.MULTILINE),
                    re.compile(
                        r'第[一二三四五六七八九十]+(?:章|部分)\s*[\u4e00-\u9fff]+|^评标办法前附表|^附录(?:一)?[：:]|^附件(?:一)?[：:]|^附表(?:一)?[：:]',
                        re.MULTILINE)
                )
            ]
            output_suffix = "tobidders_notice"
        elif selection == 2:
            # Selection 2: 评标办法
            pattern_pairs = [
                (
                    re.compile(r'^第[一二三四五六七八九十]+(?:章|部分)\s*(?=.*(?:磋商|谈判|评标|评定|评审))(?=.*(?:办法|方法))'),
                    # Alternative begin pattern
                    re.compile(r'评标办法正文|评标办法|^第[一二三四五六七八九1-9]+(?:章|部分)\s*[\u4e00-\u9fff]+')
                # Alternative end pattern
                ),
                (
                    re.compile(
                        r'(?<!见)'  # 确保前面不是“见”
                        r'第[一二三四五六七八九十1-9]+(?:章|部分)\s*'  # 匹配“第X章”或“第X部分”
                        r'[\u4e00-\u9fff、()（）]*?'  # 匹配允许的字符（中文、顿号、括号）
                        r'(?=.*(?:磋商|谈判|评标|评定|评审))'  # 确保包含“磋商”、“谈判”、“评标”、“评定”或“评审”
                        r'(?=.*(?:办法|方法))'  # 确保包含“办法”或“方法”
                        r'[\u4e00-\u9fff、()（）]*\s*$'  # 继续匹配允许的字符直到行尾
                        r'|\s*评标(办法|方法)前附表\s*$',  # 或匹配“评标办法前附表”或“评标方法前附表”
                        re.MULTILINE
                    ),
                    re.compile(r'第[一二三四五六七八九十1-9]+(?:章|部分)\s*[\u4e00-\u9fff、()（）]+\s*$', re.MULTILINE)
                )
            ]
            output_suffix = "evaluation_method"
        # TODO:exclusion
        elif selection == 3:
            # Selection 3: 资格审查条件
            pattern_pairs = [
                # (
                #     re.compile(r'^(?:附录(?:[一1])?[：:]|附件(?:[一1])?[：:]|附表(?:[一1])?[：:]).*(?:资质|能力|信誉).*$|^第[一二三四五六七八九十百千]+(?:章|部分).*?(资格).*',
                #                re.MULTILINE),
                #     re.compile(
                #         r'^(?:附录[一二三四五六七八九1-9]*[：:]|附件[一二三四五六七八九1-9]*[：:]|附表[一二三四五六七八九1-9]*[：:])(?!.*(?:资质|能力|信誉)).*$'
                #         r'^第[一二三四五六七八九1-9]+(?:章|部分)\s*[\u4e00-\u9fff]+', re.MULTILINE)
                # ),
                (
                    re.compile(
                        r'^(?:附录(?:[一1])?[：:]|附件(?:[一1])?[：:]|附表(?:[一1])?[：:]).*(?:资质|能力|信誉).*$|第[一二三四五六七八九1-9]+(?:章|部分)\s*[\u4e00-\u9fff、()（）]*资格[\u4e00-\u9fff、()（）]*\s*$',
                        re.MULTILINE),
                    re.compile(
                        r'^(?:附录[一二三四五六七八九1-9]*[：:]|附件[一二三四五六七八九1-9]*[：:]|附表[一二三四五六七八九1-9]*[：:])(?!.*(?:资质|能力|信誉)).*|'
                        r'第[一二三四五六七八九1-9]+(?:章|部分)\s*[\u4e00-\u9fff、()（）]+\s*$', re.MULTILINE)
                )
            ]
            output_suffix = "qualification"
        elif selection == 4:
            # Selection 4: 招标公告
            pattern_pairs = [
                (
                    re.compile(
                        r'^第[一二三四五六七八九十百千]+(?:章|部分).*?(?:公告|邀请书|邀请函|邀请).*|^第一卷|^投标邀请书|^投标邀请函|^投标邀请'),
                    re.compile(r'^第[一二三四五六七八九十百千]+(?:章|部分)\s*[\u4e00-\u9fff]+', re.MULTILINE)
                ),
                (
                    re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函|投标邀请)[\)）]?\s*$', re.MULTILINE),
                    re.compile(r".*(?:投标人须知|投标人须知前附表)\s*$", re.MULTILINE)
                )
            ]
            output_suffix = "notice"

        elif selection == 5:
            # Selection 5: 无效标
            pattern_pairs = [
                (
                    re.compile(
                        r'^第[一二三四五六七八九十百千]+(?:章|部分).*?(?:公告|邀请书|邀请函|邀请).*|^第一卷|^投标邀请书|^投标邀请函|^投标邀请'),
                    re.compile(r'第[一二三四五六七八九十]+章\s*合同|[：:]清标报告|^第二卷', re.MULTILINE)
                ),
                (
                    re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函|投标邀请)[\)）]?\s*$', re.MULTILINE),
                    re.compile(r'第[一二三四五六七八九十]+章\s*合同|[：:]清标报告|^第二卷', re.MULTILINE)
                )
            ]
            output_suffix = "invalid"

        else:
            print("无效的选择:请选择1-5")
            return [""]

        for idx, (begin_pattern, end_pattern) in enumerate(pattern_pairs, start=1):
            is_secondary_match = (idx == 2)
            begin_page = last_begin_index if last_begin_index != 0 else {
                1: 3,  # 前附表
                2: 10,  # 评标
                3: 5,  # 资格
                4: 0,  # 公告
                5: 0  # 无效标
            }.get(selection, 0)
            if selection == 1:  # 投标人须知
                output_paths = list(
                    extract_pages_tobidders_notice(input_path, output_folder, begin_pattern, begin_page, common_header,
                                                   is_secondary_match))
                if output_paths and any(os.path.isfile(f) for f in output_paths):
                    return output_paths
                else:
                    # print(f"Selection {selection}: 使用组合提取函数失败。尝试下一个模式对。")
                    continue
            output_paths = extract_pages(
                input_path,
                output_folder,
                begin_pattern,
                begin_page=begin_page,
                end_pattern=end_pattern,
                output_suffix=output_suffix,
                common_header=common_header,
                is_secondary_match=is_secondary_match
            )
            if output_paths and any(os.path.isfile(f) for f in output_paths):
                # print(f"Selection {selection}: 使用模式对 {idx} 成功提取。")
                return output_paths  # Return immediately upon successful extraction
            else:
                # print(f"Selection {selection}: 使用模式对 {idx} 失败。尝试下一个模式对。")
                continue

        # If all pattern pairs failed, attempt a fallback
        print(f"Selection {selection}: 所有模式对都未匹配成功。尝试执行 extract_pages_twice 或返回空。")
        fallback_result = extract_pages_twice(input_path, output_folder, output_suffix, common_header, last_begin_index)
        if fallback_result and any(os.path.isfile(f) for f in fallback_result):
            return fallback_result
        else:
            if selection == 1:
                return ["", ""]
            else:
                return ['']
    else:
        print("Invalid input path. It is neither a PDF file nor a directory containing PDF files.")
        return ['']


def truncate_pdf_multiple(input_path, output_folder, unique_id="123"):
    global logger
    logger = get_global_logger(unique_id)
    base_file_name = os.path.splitext(os.path.basename(input_path))[0]  # 纯文件名
    truncate_files = []
    selections = range(1, 6)  # 选择 1 到 5

    # 使用 ThreadPoolExecutor 进行多线程处理
    with concurrent.futures.ThreadPoolExecutor(max_workers=len(selections)) as executor:
        # 提交所有任务并保持 selection 顺序
        future_to_selection = {selection: executor.submit(truncate_pdf_main, input_path, output_folder, selection) for
                               selection in selections}

        # 按 selection 顺序收集结果
        for selection in selections:
            future = future_to_selection.get(selection)
            try:
                files = future.result()
                if files and any(f for f in files):
                    # 过滤空字符串
                    valid_files = [f for f in files if f]
                    truncate_files.extend(valid_files)
                else:
                    logger.error(f"Selection {selection}: 截取失败，已添加空字符串。")
                    # Append two empty strings for selection=1; otherwise, one empty string
                    truncate_files.extend(["", ""] if selection == 1 else [""])
            except Exception as e:
                logger.error(f"Selection {selection} 生成了一个异常: {e}")
                # Append two empty strings for selection=1; otherwise, one empty string
                truncate_files.extend(["", ""] if selection == 1 else [""])

    # Proceed with merging logic as before...
    if any(f for f in truncate_files if f):  # 检查是否有有效的文件路径
        merged_output_path = os.path.join(output_folder, f"{base_file_name}_merged_baseinfo.pdf")
        merged_result = merge_selected_pdfs_for_engineering(output_folder, truncate_files, merged_output_path,
                                                            base_file_name)
        if merged_result:
            truncate_files.append(merged_result)
            # logger.info(f"merged_baseinfo: 已生成合并文件: {merged_output_path}")
        else:
            truncate_files.append("")  # 如果 merged_result 未生成，添加空字符串
            logger.warning("merged_baseinfo: 未生成合并文件，因为没有找到需要合并的 PDF 文件。")
    else:
        truncate_files.append("")  # 如果没有文件需要合并，也添加空字符串
        logger.warning(f"merged_baseinfo: 没有文件需要合并 for {input_path}")

    return truncate_files


def truncate_pdf_specific_engineering(pdf_path, output_folder, selections, unique_id="123"):
    try:
        global logger
        logger = get_global_logger(unique_id)
        base_file_name = os.path.splitext(os.path.basename(pdf_path))[0]
        truncate_files = []

        # 使用 ThreadPoolExecutor 进行多线程处理
        with concurrent.futures.ThreadPoolExecutor(max_workers=len(selections)) as executor:
            # 提交所有任务并保持 selection 顺序
            future_to_selection = {selection: executor.submit(truncate_pdf_main, pdf_path, output_folder, selection) for
                                   selection in selections}

            # 按 selection 顺序收集结果
            for selection in selections:
                future = future_to_selection.get(selection)
                try:
                    files = future.result()
                    if files and any(f for f in files):
                        # 过滤空字符串
                        valid_files = [f for f in files if f]
                        truncate_files.extend(valid_files)
                    else:
                        logger.error(f"Selection {selection}: 截取失败，已添加空字符串。")
                        # Append two empty strings for selection=1; otherwise, one empty string
                        truncate_files.extend(["", ""] if selection == 1 else [""])
                except Exception as e:
                    logger.error(f"Selection {selection} 生成了一个异常: {e}")
                    # Append two empty strings for selection=1; otherwise, one empty string
                    truncate_files.extend(["", ""] if selection == 1 else [""])

        # Proceed with merging logic as before...
        if any(f for f in truncate_files if f):  # 检查是否有有效的文件路径
            merged_output_path = os.path.join(output_folder, f"{base_file_name}_merged_specific.pdf")
            merged_result = merge_selected_pdfs_for_engineering(output_folder, truncate_files, merged_output_path,
                                                                base_file_name)
            if merged_result:
                truncate_files.append(merged_result)
                logger.info(f"merged_specific: 已生成合并文件: {merged_output_path}")
            else:
                truncate_files.append("")  # 如果 merged_result 未生成，添加空字符串
                logger.warning("merged_specific: 未生成合并文件，因为没有找到需要合并的 PDF 文件。")
        else:
            truncate_files.append("")  # 如果没有文件需要合并，也添加空字符串
            logger.warning(f"merged_specific: 没有文件需要合并 for {pdf_path}")

        return truncate_files

    except Exception as e:
        logger.error(f"Error in truncate_pdf_specific_engineering: {e}")
        return [""] * len(selections)  # 返回与 selections 数量相同的空字符串列表

if __name__ == "__main__":
    start_time = time.time()
    # input_path = "C:\\Users\\Administrator\\Desktop\\new招标文件\\工程标"
    # input_path="C:\\Users\\Administrator\\Desktop\\fsdownload\\0b1861e6-c7f6-4541-9182-b1384ba84f3b\\ztbfile.pdf"
    # input_path = "C:\\Users\\Administrator\\Desktop\\货物标\\zbfiles\\2-招标文件.pdf"
    input_path=r"C:\Users\Administrator\Desktop\fsdownload\854fb19f-96d3-4b3e-b2ba-1a095344fd92\ztbfile.pdf"
    output_folder = "C:\\Users\\Administrator\\Desktop\\new招标文件\\output3"
    files=truncate_pdf_multiple(input_path,output_folder)
    # selections = [4, 1]  # 仅处理 selection 4、1
    # files=truncate_pdf_specific_engineering(input_path,output_folder,selections)
    for i in files:
        print(type(i))
        print(i)
    # selection = 1  # 例如：1 - 投标人须知前附表+正文, 2 - 评标办法, 3 -资格审查条件 4-招标公告 5-无效标
    # generated_files = truncate_pdf_main(input_path, output_folder, selection)
    # print(generated_files)
    # print("生成的文件:", generated_files)
    end_time = time.time()
    print("耗时：" + str(end_time - start_time))