PDF WAM

This is a rewrite of Egovmon/Tingtun PDF WAM rewrite. This rewrite focuses on,

1. Update to the latest PyPDF2 library.
2. Drop dependencies on deprecated libraries.
3. Merge and remove ununsed modules.
4. Use consistent naming of functions, such as replacing camelCase functions with snake_case
5. Make the code run as a HTTP server backend similar to current PDF WAM.

Setup

Make sure you have python3 with virtual environment support. This can be done by installing,

sudo apt install python3-venv -y

Create a virtual environment, say pdfwam and switch to it

python3 -m venv pdfwam; source pdfwam/bin/activate

Install the requirements.

pip install -r requirements.txt

Running Tests

There is a local test suite for PDF files for each type of test we are support in WCAG 2.0 PDF Techniques. For running the checker against one of those files.

python pdfchecker.py <filename> -r

For example,

python pdfchecker.py testfiles/wcag.pdf.01/images-with-and-without-ALT.pdf -r

The following shell command can be used as a short-cut to run the checker against all files in the test suite.

for i in $(find ./testfiles -name \*.pdf); do python pdfchecker.py "$i" -r; done

Code Structure

Core

 PdfStruct <--- PdfWCAG <---<testing>-- PdfReaderWrapper --<read>-> PyPDF2.PdfReader 

This gives you the AWAMs.
1. A WAM -> Accessibility WAM
2. B WAM -> Barrier WAMs (compose AWAM data) 
3. M WAM -> Metadata WAMs -> wraps up on top of metadata like creator, producer etc.

Service

 extractAWAMIndicators (pdfAWAM.py)

 Wrap this up in some simple service
   * protobuf (if you want to get funky on this)
   * RestFUL (What I'd suggest)
     * uwsgi --> 
	 * 

Architecture

backend - standalone backend as of now, FE -> do it in modern angular/react/vue ?