Change order of pdf bookmarks using PyPdf2

Question

I create an application, that merges multiple pdfs with a bookmark. If the origin pdfs already have bookmarks, I want to keep them and just add a bookmark at the beginning of the pdf. I use the following code: The title and the path in the code come from user input.

from PyPDF2 import PdfFileReader, PdfFileMerger
from PyPDF2.utils import PdfReadError

merger = PdfFileMerger()
pages_total = 0
for path, title in zip(paths, titles):
    merger.append(path, import_bookmarks=True)
    with open(path, "rb") as file:
           pdf = PdfFileReader(file, "rb")
           pages_one = pdf.getNumPages()
           pages_total += pages_one

    merger.addBookmark(f"{title}", pages_total - pages_one, parent=None)

 save_path = input() + ".pdf"  
 merger.write(save_path)
 merger.close()

The new pdf has all bookmarks, that points to the right pages. However, that bookmark for the entire pdf appears after the pdfs that already existed. How can I change the order in which the bookmarks are displayed?

The picture below hopefully clarifies the problem. The last bookmark was added and should be the first in the list.

Answer 1

I found a solution. First I create a parent bookmark with the the pdf title. Then I get all existing bookmarks in the pdf with pdf.getOutlines() and iterate over the output:

parent = merger.addBookmark(f"{title}", pages_total - pages_one, parent=None)
for i in pdf.getOutlines():
     if isinstance(i, list):  # if nested bookmarks exists
         nested_boomarks(parent=child, bookmarks=i, merger=merger, pdf=pdf)
     else:
         child = merger.addBookmark(i["/Title"], pdf.getDestinationPageNumber(i), parent=parent)
                           

When there are bookmarks inside the bookmarks I use this recursive function.

def nested_boomarks(parent, bookmarks, merger, pdf):
        for i in bookmarks:
            if isinstance(i, list):
                nested_boomarks(child, i, merger, pdf)
            else:
                child = merger.addBookmark(i["/Title"], pdf.getDestinationPageNumber(i), parent=parent)

This way I get the following output: