A python script that extracts self assesment questions (SAQs) from Open University textbooks into Anki cards
This script uses the pdftotext package to convert textbook PDFs into parseable text.
Please note that pdftotext has some OS specific dependencies that must be present for a successful installation.
You can find OS specific instructions for installing pdftotext here https://github.com/jalan/pdftotext#os-dependencies
In my case (Debian on WSL) installation consisted of:
sudo apt install build-essential libpoppler-cpp-dev pkg-config python3-dev
pip install pdftotext
You need to have the textbook PDFs downloaded into a directory you know the path to. You can find these textbook PDFs in the TM354 resources section.
Clone this repo, and cd into the newly created directory
git clone https://github.com/chowieuk/SAQ-to-Anki
cd SAQ-to-Anki
You might find it easier if SAQripper.py and the textbook PDFs are in the same directory, so I would suggest either moving the script to where you have the textbooks, or vice versa.
py SAQripper.py path/to/textbook.pdf
This will create a file named "textbook cards.txt" containing an Anki card for each SAQ present in the given textbook
So for example if I had a textbook named Block 1.pdf in the same directory as the script, I would run:
py SAQripper.py "Block 1.pdf"
And I would expect to see a file named Block 1 cards.txt
Please take note of the quotes - if your path contains any spaces then you must wrap it in quotes.
- Open Anki
- Create a new Anki deck e.g. Block 1
- Click Import File
- Choose the relevant text file
- Choose Type "Basic (with code block styling)"
- Choose the Deck you created earlier
- My other settings (which I believe are the default) looked like this:
You should receive a notice that looks something like this:
Importing complete.
149 notes added, 0 notes updated, 0 notes unchanged.
And that's it, you've got your Anki cards!
If you're noticing some issues you can supply the debug flag to get some more information
py SAQripper.py path/to/textbook.pdf debug
This will create two additional files: textbook filtered.txt and textbook raw.txt so you can compare the raw output of pdftotext with the output from the script after some initial filtering. These two files will help you compose better regexs should you want to improve the results of the script.
Debug mode will also print a significant amount info to the terminal. I found it useful to pipe the debug output to less via py SAQripper.py "Block 1.pdf" | less
- Sometimes answer blocks are incomplete
- This results in incompete answers, or missing subanswers
- Write and include a test suite
- Fix known issues
- Add a flag to make tagging cards optional
- Add tags to each card for which Unit the card belongs to
- Remove tags per page number, as they are included in the card body and it creates too many tags
- Add option for cards in two different forms:
- SAQ + answer block
- Cards for each subquestions ((a), (b)... etc) (this is the current default mode)
- Add more options for filenames
- Add guards when overwriting files