pdf-diff

Finds differences between two PDF documents:

1. Compares the text layers of two PDF documents and outputs the bounding boxes of changed text in JSON.
2. Rasterizes the changed pages in the PDFs to a PNG and draws red outlines around changed text.

The script is written in Python 3, and it relies on the pdftotext program.

Installation

sudo apt-get install libxml2-dev libxslt1-dev poppler-utils # on Ubuntu, at least
sudo pip3 install pillow lxml

# get my Python extension module for the Google Diff Match Patch library
# so we can compute differences in text very quickly
git clone --recursive https://github.com/JoshData/diff_match_patch-python
cd diff_match_patch-python
sudo apt-get install python3-dev
python3 setup.py build
sudo python3 setup.py install

Running

Turn two PDFs into one large PNG image showing the differences:

python3 pdf-diff.py before.pdf after.pdf > test.png

Node Version

There's also a node.js version for computing the changes that uses Mozilla's pdf.js instead of pdftotext. Getting pdf.js to work in node.js isn't straightforward:

npm install

git clone https://github.com/mozilla/pdf.js
cd pdf.js
node make singlefile
cd ..

Compute the changes (writes a JSON file):

node index.js before.pdf after.pdf | grep -v "^Warning:" > changes.json

(Unfortunately the pdf.js library prints warnings on STDOUT, so we have to filter those out.)

Render the changes:

python3 pdf-diff.py --changes < changes.json > test.png

pdftotext gives bounding boxes for each word in the document while pdf.js only gives bounding boxes for text runs, so the granularity is not as good.