Layers got deleted Automatically on saving PDF page.

[pachhaipurna]

@JorjMcKie` I am creating new pdf page based on page number , While creating new pdf page all annotation attribute are correct except Layer.Layers got automatically removed while saving pdf page. This is my code.

`output_pdf_with_color = fitz.open(filetype = "pdf")
input_pdf_page = fitz.open(input_src_path,filetype = "pdf")
for page_no in pdf_markup_rslt[subsystem_key]['Page No']:
page_no_in_str = page_no_in_str + ' Pages ' + str(page_no)
output_pdf_with_color.insert_pdf(input_pdf_page,from_page = page_no-1, to_page = page_no-1, links=1, annots=1)

pdf_with_color = os.path.join(subsystem_dir, '{}_final.pdf'.format(subsystem_key))
output_pdf_with_color.save(pdf_with_color)`

Could you please let me know How to fix it ?

Original file contains layers.

Thank you

[JorjMcKie]

For the time being, this is not a bug, but a question.
Transferring to "Discussions".

[JorjMcKie]

I am sorry, but your code is hardly readable, because of the mixture formatted and unformatted code.
What I understood is that your output PDF is filled with pages from some source PDF via insert_pdf().
Correct so far?

[pachhaipurna]

@JorjMcKie Yes, I want to add specific page number to new pdf page using insert_pdf(), While saving to pdf output all Layers that are present originally are removed, I am unable to save layers in my output file. Your help is highly appreciated.

` ###Output file
output_pdf_with_color = fitz.open(filetype = "pdf")

input_pdf_page = fitz.open(input_src_path,filetype = "pdf")
##Page Number to be prepared,response data

for page_no in pdf_markup_rslt[subsystem_key]['Page No']:
    page_no_in_str = page_no_in_str + ' Pages ' + str(page_no)

    output_pdf_with_color.insert_pdf(input_pdf_page,from_page = page_no-1, to_page = page_no-1, links=1, annots=1)

    pdf_with_color = os.path.join(subsystem_dir, '{}_final.pdf'.format(subsystem_key))
    output_pdf_with_color.save(pdf_with_color)`

Thank you.

[JorjMcKie]

Inserting pages from a source to a target PDF is a page-oriented process. Layers however are stored centrally in a PDF - not by page or inside a page.

While OCGs referenced by a page will copied over to the targt PDF - like any other referenced object - they will still not become known as OCGs by this copy.

You therefore must find a way to replicate the source Optional Content situation in the target somehow.

This is currently not supported and requires your own - probably hacky - code. I cannot think of a complete and failsafe way to do that to be honest.

[pachhaipurna]

@JorjMcKie I tried garbage=3, deflate=True, clean=False". but it takes lots of time, Could you please let me know other process from pymupdf?

[JorjMcKie]

Try reducing the garbage value to 2 or 1.

[pachhaipurna]

@JorjMcKie Already tried but no any effect .Is there any option we can delete ghost object. or while deleting page we can delete those objects permanently.

[pachhaipurna]

@JorjMcKie PDF file size is increasing than the original one, Even though I have deleted more than half of the page.

[JorjMcKie]

You could try one last thing: To remove possible "article beads" in the source PDFdo this:

import fitz
src = fitz.open(source.pdf")
for page in src:
    pxref = page.xref
    src.xref_set_key(pxref, "B", "null")  # remove any reference to article beads

# no execute the page insertions to the target
doc = fitz.open(...)  # your target PDF
doc.insert_pdf(src, ...)
doc.ez_save(...)