The coordinates are in millimeters in PyMuPDF

[Sv9t]

Hello!
I have coordinates of an image in millimeters in pdf file. Coordinates of the top-left corner: x=24mm, y=88mm. And I have the width of the image: 77mm, height image: 7mm.
How I can use my coordinates in millimeters in function fitz.Rect()? Rect() uses points in coordinates on the pdf page. I was trying to convert millimeters to pixels (point), but this method did not help.

[JorjMcKie]

You should be able to compute mm by using these equalities: 72 points = 1 inch, 1 inch = 25.4 mm.
What did you use, that did not work?
Your point x=24mm, y=88mm should therefore compute to x: 24 / 25.4 * 72 = 68 points, y: 88 / 25.4 * 72 = 304,9 point, etc.
Counter-calculation: page size ISO A4 = 595 x 842 points => 595/722.54 = 21 cm width, and 842/722.54 = 29.7 cm.

[JorjMcKie]

oops, GitHub editor tricked me out: 595 / 72 * 2.54 and 842 / 72 * 2.54!

[Sv9t]

What did you use, that did not work?

my bad, I convert millimeters in pixels, from the calculation 1mm=3.793627px. This is wrong!

595/722.54 = 21 cm width, and 842/722.54 = 29.7 cm.

I didn't understand where the number came from 722.54?

I drew my calculations, attach jpeg. If use fitz.Rect(68, 304.9, 308.4, 517,3) then I get an error "raise ValueError("bad Rect: sequ. length") ValueError: bad Rect: sequ. length"
Please tell me what I calculated wrong?

[JorjMcKie]

rectangle constructor needs 4 numbers - you gave 5!

[JorjMcKie]

probably 517.3 - you coded a comma.

[Sv9t]

Thanks! This is my mistake! I wasn't attentive
The image insert to pdf page, but the image is much lower for X coordinate than it should be. I'll try different options...

[JorjMcKie]

The image insert to pdf page, but the image is much lower for X coordinate than it should be. I'll try different options...

There may be this issue, so that you must "wrap" the page contents first.

[JorjMcKie]

just try page.wrap_contents() before you insert.

[Sv9t]

import fitz

file = '56204820.pdf'
regnum = 'regnum.png'
# Coordinates of the top-left corner: x=24mm, y=88mm
# Width of the image: 77mm and height image: 7mm.
coords = [24, 88, 77, 7]

def func_coords(*args):
    print(args)
    point_coords = []
    for c in args[0][:2]:
        point_coords.append(c / 25.4 * 72)

    for c in args[0][2:]:
        points = c / 25.4 * 72
        if c == args[0][2]:
            result = 595 - points - args[0][0]
        else:
            result = 842 - points - args[0][1]
        point_coords.append(result)
    return tuple(point_coords)


def main():
    pdf = fitz.open(file)
    page = pdf[0]

    point_coords = func_coords(coords)
    print(point_coords)

    rect = fitz.Rect(point_coords)
    img = open(regnum, "rb").read()  # an image file
    page.wrapContents()
    page.insertImage(rect, stream=img)

    pdf.save('result.pdf')


if __name__ == "__main__":
    main()

This is my script, maybe there is a problem in the code? String page.wrapContents() it didn't help me, image is much lower for Y coordinate than it should be.

[JorjMcKie]

A rectangle must be given as rect = fitz.Rect(x0, y0, x1, y1) where (x0, y0) is the top-left point and (x1, y1) the bottom-right one. If you have the the top-left only plus the width / height, the calculation is therefore rect = fitz.Rect(x0, y0, x0 + width, y0 + height).
So in your case rect = fitz.Rect(24, 88, 24+77, 88+7). Then simply do this to convert the whole thing to pixels:

>>> rect = fitz.Rect(24, 88, 24+77, 88+7)
>>> rect = rect / 25.4 * 72
>>> rect
Rect(68.03149606299213, 249.4488188976378, 286.2992125984252, 269.29133858267716)
>>> 

[Sv9t]

Excellent!! It works!
I calculated rectangle the bottom-right wrong, I calculated from the bottom-right corner of the a4 page, but I had to calculate from the top-left corner.
Thank you for your help and your feedback! PyMuPDF it is very necessary module!