This utility compares PDF files visually by converting each page into images and then comparing them using OpenCV. It is particularly useful for identifying differences between PDF files that may not be apparent through text comparison alone.
- Compares PDF files visually, page by page.
- Supports multi-page PDF files.
- Reports differences between PDF files, specifying the page number and source file.
- Python 3.x
- PyMuPDF (
fitz) library - OpenCV (
cv2) library
For detailed requirements, see below.
-
Clone the repository:
git clone https://github.com/Formartha/compare-pdf.git -
Install the required dependencies:
pip install pymupdf opencv-python
pip install compare-pdf
compare_pdf --pdf <path_to_pdf1> --pdf <path_to_pdf2> ... [--showdiffs={sidebyside|diffimage}]
- Replace
<path_to_pdf1>,<path_to_pdf2>, etc. with the paths to the PDF files you want to compare. At least two PDF files are required for comparison. - The
--showdiffsoption requests the display of differing pages in a window as they are discovered. A diff window is dismissed by typing any character and the processing continues- sidebyside places the differing pdf pages horizontally next to each other in the diff window
- diffimage places an OpenCV absdiff result in the diff window, a white background with non-white pixels showing where the images differed.
The program returns a zero result code when there are no differences, otherwise -1
compare_pdf --pdf file1.pdf --pdf file2.pdf
This will compare full/path/to/file1.pdf and full/path/to/file2.pdf visually, reporting any differences found.
The requirements.txt file specifying exactly which modules are needed in a virtual environenment was generated using the commands below
pip3 install pipreqs
pip3 install pip-tools
pipreqs --savepath=requirements.in && pip-compile
The result was not perfect, however, so that manual additions were needed:
PyMuPDF==1.25.5
setuptools==75.8.0
wheel==0.45.1
This project is licensed under the MIT License - see the LICENSE file for details.
