I am attempting to parse through NSDUH data which is publicly available here :
I have seen Python packages like PDFMiner and tabula-py, but I have not been able to have success when I come across a table that is delimited by white space. Here is an example of a survey response that is difficult to mark as a "table" object :
The most promising piece that I have together is leveraging "xpdf command line tools" available here : https://www.xpdfreader.com/download.html
I put the "pdftotext.exe" file in my Python directory and can create near-perfect text copies of the pdf files I am manipulating with this command :
C:\Users\UserName\Python\Python37> pdftotext.exe -layout NameOfPDF.pdf
Unfortunately, this output would still require a dependency on a series of regular expressions. If there is an open-source software or python package which can detect that the image attached should be treated as one contiguous block with data points within it, then that would solve my problem.