How to disable line weights when saving a page to image?

[cpwan]

I am currently working on pages with vector graphics. I want to save the page as an image but I encounter issues with 'line weights'.

What I expected: (with line weights disabled)

What I got: (with line weights enabled)

In adobe pdf, I can disable 'line weights' when ( and only when) viewing the pdf, following this link .

I am using the following code to save the page as image:

doc = fitz.open(pdffile)
page = doc.loadPage(page_id)
pix = page.getPixmap(annots=False, matrix=fitz.Matrix(scale, scale)) 
pix.writePNG(out_name)

How can I disable the 'line weights' when saving it as image?

Here is a minimal example. The difference becomes more prominent when zooming in.

minimal example.pdf

[JorjMcKie]

May be a problem of your pymupdf version ... at least you are using the deprecated camelCase names.
I did this:

>>> for dpi in (100, 200, 300):
	pix = page.get_pixmap(dpi=dpi)
	pix.save("%i.png" % dpi)

... and everything looks perfect!
100 dpi:

200 dpi:

300 dpi:

And to be sure, here one with 400 dpi:

pymupdf version: 1.19.3

[cpwan]

I expect the following, with the 'line weight' disabled. This is viewed with adobe pdf.

[cpwan]

demo.mp4

[JorjMcKie]

So you want an option to maintain a line width of 1 even if the image is generated with 400% resolution?
No that is not possible.
Pixmap generation scales everything evenly.

[JorjMcKie]

But your first image shown with that CL600 text should not happen either.
Can you let me have the corresponding file?

[cpwan]

Sure, please have a look: cl600 demo.pdf

[cpwan]

My current workaround is to

1. save the page as svg
2. modify the stroke-width to a small value like .01
3. print the svg to pdf
4. use the above code to save it as an image

import re
doc = fitz.open(pdffile)
page = doc.loadPage(page_id)
svg = page.get_svg_image(matrix=fitz.Identity)
svg = re.sub(r'stroke-width=".[0-9][0-9]"', 'stroke-width=".01"', svg)
print(svg, file=open("dummy.svg", "w"))
pix_doc = fitz.open("dummy.svg")
pix = pix_doc.load_page(0).get_pixmap(
     annots=False, matrix=fitz.Matrix(scale, scale)
)
pix.save("thin_lines.png")

The problem becomes how to change the line width of vector graphics in pdf.

[JorjMcKie]

This snippet also works ... at least with your example PDF:

>>> doc=fitz.open("cl600.demo.pdf")
>>> page=doc[0]
>>> page.clean_contents() # enforce standard painting syntax
>>> cont_xref = page.get_contents()[0] # we only have 1 of those now
>>> # read an modify the paint commands
>>> lines = doc.xref_stream(cont_xref).splitlines()
>>> # walk through lines and replace each width value
>>> for i in range(len(lines)):
	if lines[i].endswith(b" w"):  # command to set line width!
		lines[i] = b"0.02 w"

		
>>> # write back modified paint commands
>>> doc.update_stream(cont_xref, b"\n".join(lines))
>>> # now either make a pixmap of modified page ...
>>> pix=page.get_pixmap(dpi=300)
>>> pix.save("300.png")
>>> # ... or save a modified PDF
>>> doc.save("thin-lines.pdf")
>>> 

[cpwan]

Thanks!