pypdf处理pdf的书签为分级书签

问题

从网上下载了一本书的pdf，发现书签没有分级，对强迫症很难受，于是用python的pypdf库处理了一下，踩了点小坑，在这里放一下代码吧。

处理前的书签

处理后的书签

代码

import PyPDF2 

#层级打印书签
def print_outLine(outline: list, sub: int = 0):
    for line in outline:
        if isinstance(line,list):
            print_outLine(line,sub+1)
        else:
            print(sub*"    "+line.title)
            
#处理无级书签为层级结构，以.的个数为层级标准           
def to_new_outline(outline, end: int, curlevel:int = 0, index: int = 0):
    new_outline = []
    while index<end:
        line = outline[index]
        level = line.title.split(" ")[0].count(".")
        if level == curlevel:
            new_outline.append(line)
            index += 1
        else:
            remain = 1
            while outline[index+remain].title.split(" ")[0].count(".") > curlevel:
                remain += 1
            sub_outline = to_new_outline(outline,index+remain,curlevel+1,index)
            new_outline.append(sub_outline)
            index = index+remain
    return new_outline

#设置书签
def set_outline(outline: list, root, writer: PyPDF2.PdfWriter):
    i = 0
    while i < len(outline):
        if not isinstance(outline[i],list):
            parent = writer.add_outline_item(outline[i].title,pdf_in.get_page_number(outline[i].page),root)
            if i+1<len(outline) and isinstance(outline[i+1],list):
                i=i+1
                set_outline(outline[i],parent,writer)
            else:
                i+=1
        else:
            i+=1


if __name__ == "__main__":
    pdf_in = PyPDF2.PdfReader("./test.pdf")
    outline = pdf_in.outline
    new_outline = to_new_outline(outline,end=len(outline))
    print_outLine(new_outline,0)

    # 写出pdf
    writer = PyPDF2.PdfWriter()
    #复制原书页
    for page in pdf_in.pages:
        writer.add_page(page)
    #根书签
    root = writer.get_outline_root()
    set_outline(new_outline,root,writer)
    #写文件
    with open("./new.pdf","wb") as f:
        writer.write(f)
        f.close()