keep_aspect_ratio flag not recognized in image() method

[avascherocman]

keep_aspect_ratio flag not recognized when attempting to create an image. Also happens when a rectangle is not involved, but thought that providing code directly from the repository would make this easier to re-create

Error details

TypeError                                 Traceback (most recent call last)
Cell In[88], line 5
      3 rect1 = 30, 30, 60, 100
      4 pdf.rect(*rect1)
----> 5 pdf.image(
      6     "test_image.jpg",
      7         *rect1,
      8         keep_aspect_ratio=True,
      9 )

File c:\Users\...\fpdf\fpdf.py:150, in FPDF.check_page.<locals>.wrapper(self, *args, **kwargs)
    148     self.error("No page open, you need to call add_page() first")
    149 else:
--> 150     return fn(self, *args, **kwargs)

TypeError: FPDF.image() got an unexpected keyword argument 'keep_aspect_ratio'

Code to reproduce

from fpdf import FPDF

pdf = FPDF()
pdf.add_page()
rect1 = 30, 30, 60, 100
pdf.rect(*rect1)
pdf.image(
    "test_image.jpg",
        *rect1,
        keep_aspect_ratio=True,
)

# from https://github.com/py-pdf/fpdf2/blob/master/test/image/test_image_fit.py

Environment
Please provide the following information:

• Operating System: Windows
• Python version: Python 3.14.0
• fpdf2 version used: 2.8.5

[Lucas-C]

Based on my analysis, this can be resolved by addressing the core issue.

What do you mean more precisely?

Also, bots are not welcome on this project: are you a robot?

[Lucas-C]

keep_aspect_ratio flag not recognized when attempting to create an image. Also happens when a rectangle is not involved, but thought that providing code directly from the repository would make this easier to re-create

This is could be caused by havinh both fpdf2 & fpdf installed in your environment.
fpdf package is PyFPDF, the ancestor of fpdf2 : https://pypi.org/project/fpdf/

Under Mac or Linux, if you are using pip, this can be easily checked by running this command:
pip freeze | grep fpdf

If you are using Windows or another package manager than pip, you will have to figure what is the command to display the packages currently installed in your environment.

[avascherocman]

Hi Lucas, thanks for your help with this.

I don't think both packages being present is the issue - you can check my pip freeze output here - requirements.txt - and see that only fpdf2 is present.

And to confirm, here's what happens when I try to remove fpdf and fpdf2 from my environment.

[andersonhc]

Hi @avascherocman

On the traceback you posted, looking at the line number, the error points to fpdf.py:150 inside the check_page wrapper. That line number matches legacy fpdf 1.7.2, not fpdf2 2.8.5. In 2.8.5, the check_page wrapper is at a different location, and FPDF.image() does accept keep_aspect_ratio.

This strongly suggests the code is running against a different module than the one shown in your pip freeze.

Could you check the actual module being imported in the same session that fails?

These debugging prints can give some clues:

import sys, inspect, fpdf
print("Python:", sys.executable)
print("fpdf file:", fpdf.__file__)
print("fpdf version:", getattr(fpdf, "__version__", "no __version__"))
print("FPDF.image signature:", inspect.signature(fpdf.FPDF.image))

If fpdf.__file__ points to something like a local fpdf.py or a different site‑packages folder, that’s the mismatch. Also, your screenshots show a venv in PowerShell — can you confirm the failing run is in that activated venv?

[avascherocman]

Hi @andersonhc !

I can confirm the failing run is in that activated venv. I should probably also mention this is running in a Jupyter notebook - I'm not sure why that would cause problems, though.

Here's the output of the debugging lines you sent:

Python: c:\Users\[my stuff here]\.venv\Scripts\python.exe
fpdf file: c:\Users\[my stuff here]\.venv\Lib\site-packages\fpdf\__init__.py
fpdf version: 2.8.5
FPDF.image signature: (self, name, x=None, y=None, w=0, h=0, type='', link='', title=None, alt_text=None, dims=None, keep_aspect_ratio=False)

From what I can tell, it's picking up the correct fpdf2 2.8.5 package and fpdf.image() signature, but please let me know if I'm reading something incorrectly.