前几天有群友提出能否通过自动化实现提取PDF书签对应页面的功能，例如一个文件有100页，其中30页有书签且不规律地分布在整个文件中，如何自动化提取这30页，并且保留原有的书签结构：
小格元旦假期时间用Python写了一个小程序分享给大家，小程序界面非常简洁，可以处理单个PDF，也可以批量处理指定文件夹及其子文件夹下所有PDF，提取后的PDF文件和原文件在同一目录下，文件名添加后缀“_提取书签对应页面”：
https://pan.baidu.com/s/1Ti0AgHTieWvRnTukUq_nJA?pwd=iReg
同时把Python源代码分享如下，欢迎大家提出宝贵意见，以便不断改进：
import tkinterimport tkinter.filedialogimport tkinter.messageboximport osimport PyPDF2def extract_bookmark_pages(input_path):# 打开PDF文件with open(input_path, 'rb') as file:reader = PyPDF2.PdfReader(file)writer = PyPDF2.PdfWriter()# 定义递归获取书签页面的函数def get_pages(outlines):bm_pages = []for outline in outlines:if type(outline) != list: # outline不是列表，即属于一级书签的情况，将书签的目标页码写入列表bm_pagesdest_page = reader.get_destination_page_number(outline)if dest_page not in bm_pages:bm_pages.append(dest_page)else: # outline为二级书签组成的列表，则递归调用get_pages函数bm_pages.extend(get_pages(outline))return bm_pages# 将提取的页面添加到writer中input_pdf_bm_pages = get_pages(reader.outline)page_map = {original: new for new, original in enumerate(input_pdf_bm_pages)} # 创建原始页码到新页码的映射for page_num in input_pdf_bm_pages:writer.add_page(reader.pages[page_num])# 定义递归复制书签的函数def add_bookmarks_recursively(outlines, parent=None, page_map=None):for outline in outlines:if type(outline) != list: # outline不是列表，即属于一级书签的情况，则在writer中创建该书签new_parent = writer.add_outline_item(outline.title, page_map[reader.get_destination_page_number(outline)], parent) if page_map else Noneelse: # outline为二级书签组成的列表，则递归调用add_bookmarks_recursively函数add_bookmarks_recursively(outline, new_parent, page_map)# 创建新PDF文件的书签add_bookmarks_recursively(reader.outline, page_map=page_map)# 写入输出PDFoutput_path = os.path.splitext(input_path)[0] + '_提取书签对应页面.pdf'writer.set_page_mode('/UseOutlines') # 设置“初始视图 - 导览标签”为“书签面板和页面”with open(output_path, 'wb') as output_file:writer.write(output_file)class Window():def __init__(self):self.root = root = tkinter.Tk()self.label = tkinter.Label(root, text='选择文件')self.label.grid(row=0, column=0)self.entryfile_name = tkinter.Entry(root)self.entryfile_name.grid(row=0, column=1)self.BrowserDirButton = tkinter.Button(root, text='浏览', command=self.select_file)self.BrowserDirButton.grid(row=0, column=2)self.ButtonCov = tkinter.Button(root, text='开始处理', command=self.Conv1, )self.ButtonCov.grid(row=0, column=3)self.label = tkinter.Label(root, text='或者')self.label.grid(row=1, column=0)self.label = tkinter.Label(root, text='选择目录')self.label.grid(row=2, column=0)self.entryDir = tkinter.Entry(root)self.entryDir.grid(row=2, column=1)self.BrowserDirButton = tkinter.Button(root, text='浏览', command=self.BrowserDir)self.BrowserDirButton.grid(row=2, column=2)self.ButtonCov = tkinter.Button(root, text='批量处理', command=self.Conv2, )self.ButtonCov.grid(row=2, column=3)def select_file(self):# 弹出选择文件对话框file_name = tkinter.filedialog.askopenfilename()if file_name:self.entryfile_name.delete(0, tkinter.END)self.entryfile_name.insert(tkinter.END, file_name)def Conv1(self):file_name = self.entryfile_name.get()if file_name.lower().endswith('.pdf'):extract_bookmark_pages(file_name)tkinter.messagebox.showinfo("iReg", "已完成")else:tkinter.messagebox.showinfo("iReg", "请选择PDF文件")def BrowserDir(self):directory = tkinter.filedialog.askdirectory(title='iReg小程序')if directory:self.entryDir.delete(0, tkinter.END)self.entryDir.insert(tkinter.END, directory)def Conv2(self):for root, dirs, files in os.walk(self.entryDir.get()):for file in files:try:if file.lower().endswith('.pdf'):file_path = os.path.join(root, file)extract_bookmark_pages(file_path)except:passtkinter.messagebox.showinfo("iReg", "已完成")def mainloop(self):self.root.minsize(380, 120)self.root.maxsize(380, 120)self.root.title('iReg小程序 - 提取PDF书签对应页面')self.root.mainloop()if __name__ == "__main__":window = Window()window.mainloop()
