Merge a list of PdfDocument objects into a single PdfDocument object

[hector-sherpas]

Is there a way to merge a list of preprocessed PdfDocument objects into a single PdfDocument object? I have tried with import_pages method but it gives an error on the second iteration.

[mara004]

Post a reproducible example (code, input file, traceback), then I can take a look.
We can't provide help or debug this with the insufficient information given. FWIW, import_pages() works for me.

[hector-sherpas]

import pypdfium2 as pdfium

def get_doc_page_from_doc(doc: pdfium.PdfDocument) -> list(pdfium.PdfDocument):
    pdf_pages = []
    for page_idx in range(len(doc)):
        blank_doc = pdfium.PdfDocument.new()
        blank_doc.import_pages(doc, pages=[page_idx])
        
    return pdf_pages


def gather_pages_to_unique_doc(pages: list[pdfium.PdfDocument]) -> pdfium.PdfDocument:
    new_doc = pdfium.PdfDocument.new()
    for idx_page, page in enumerate(pages):
        #width, height = page.get_page_size(0)
        #new_doc.new_page(width, height)
        new_doc.import_pages(page, pages=[idx_page])

    return new_doc

doc = pdfium.PdfDocument('path_doc')

pdf_pages = get_doc_page_from_doc(doc)
gather_doc = gather_pages_to_unique_doc(pages)

[mara004]

Looks like gather_pages_to_unique_doc(), which triggers the error, might do OOB index access.
You're counting up idx_page without checking the document's length.
The pages argument of import_pages() is, of course, relative to the document being imported, so perhaps you always want the first page (index 0) instead of counting up?

[mara004]

Aside from that, calling a variable that holds a PdfDocument page is pretty confusing. Maybe single_page_doc or something would be clearer. Again assuming that what you want to do is merge a sequence of single-page PdfDocuments back into one PdfDocument.
Note that you can also omit the pages argument completely to import all pages of a given doc.

[hector-sherpas]

That's right, setting the index to zero or clearing the pages argument merges all the PdfDocument objects in the list into a single PdfDocument object. Otherwise, it was just a small example. Thanks