Different Cell /Rectangle Boxes Sizes

[thoughtfuldata]

I am having trouble extracting data from this old "job application" style pages where the cell/rectangles are of different sizes and not lined up.. Any ideas on how to approach?

Current idea is to just search for the specific titles and expand the rect into a larger one, but I was hoping to get simpler/better ideas. possibly one with .find_table(s) or .extract_table(s)

keep in mind the "Acid %" is blank and that's a value I want to extract.

The second empty table does not need to be extracted but I showed it for context

example.pdf

[jsvine]

The main issue you're going to run into is that this is a scanned, image-based PDF — not a digital-native/"true" PDF — and so it's not going to have the rich details re. graphical object positions, text characteristics, et cetera. As noted in the documentation, pdfplumber will generally be less useful for these types of PDFs, since there's less information to work with.

That said, you might use page.search(...) to find the location of the text you care about, and then page.crop(...) to select the area beneath/around it, and then cropped_selection.extract_text(...) to grab the text there.