Better understanding memory consumption when using pdfplumber

[EdmundsEcho]

When I parse large files the memory goes up accordingly. I have a couple of profiles that describe what happens in a relatively small and large file. It makes me wonder if there is a way to work through the pdf in a "chunked" manner that permits a more constant memory profile.

Notwithstanding that I could be thinking about this incorrectly, here are two profile charts describing the situation. Note the increasing memory consumed by pdfplumber/internals.