Some lines are missing when using the method: get_pixmap()

[zZzClarezZz]

Description of the bug

test.pdf
When I tried to convert the pdf file to an image and save it locally, I noticed that some lines were missing from the image

How to reproduce the bug

I want to convert pdf to jpg, like this:

doc = fitz.open(pdf_file)
page = doc[0]
matrix = fitz.Matrix(3, 3)
pix = page.get_pixmap(matrix=matrix)
image_save_path = "test.jpg"
pix.save(image_save_path)

but some lines were missing:

PyMuPDF version

1.25.4

Operating system

Windows

Python version

3.9

[JorjMcKie]

Cannot reproduce: Which lines do you missing, then?

[zZzClarezZz]

Thank you for your reply,
As shown in the figure, the left is the original file image, and the right is the saved image after extraction using the method get_pixmap(), with some lines missing in the image

[JorjMcKie]

Authoritative PDF viewers like Adobe Acrobat, PDF-XChange, Foxit, evince (Linux) and others do not display these lines either. Here is how Adobe renders the page:

So I'm afraid wherever your left image part came from, it did not land in the PDF as expected.

[zZzClarezZz]

When I opened the pdf in Microsoft Office and Google Chrome, the lines appeared fine. I used show_pdf_page () to copy the content into the new pdf, which also successfully displayed them in Microsoft Office and Google Chrome.

It seems that using the show_pdf_page() method can successfully copy these lines that Adobe Acrobat does not show. Is there any difference in the replication range between the two methods:get_pixmap() and show_pdf_page ().

[JorjMcKie]

Method page.show_pdf_page()? How does that come into the game?
It seems you didn't tell me the whole story.

I am going to convert this into a "Discussions" item. There is no current evidence for an issue.

[zZzClarezZz]

I originally wanted to convert the pdf file to an image and save it, so I used the method page.get_pixmap() but found that the resulting image was not what I expected（some lines missing）.These lines look normal from Google Chrome and office. so i wanted to know if the lines didn't load properly or if the page.get_pixmap() method was missing them, then I tried to copy the contents into the new pdf using page.show_pdf_page() the lines still look normal in Google Chrome.

doc = fitz.open(pdf_file)
page = doc[0]

# the originally code:
matrix = fitz.Matrix(3, 3)
pix = page.get_pixmap(matrix=matrix)
image_save_path = "test.jpg"
pix.save(image_save_path)

# the additional code:
newDoc = fitz.open()
newPage = newDoc.new_page(page.number)
newPage .show_pdf_page(page.rect, doc)
newDoc.save("new.pdf")

It appears that the page.show_pdf_page() method captures these lines normally but page.get_pixmap() does not. So I'm asking if there's a different range of replication between the two that causes this.