Nice to have feature/idea - Support for extracting bounding box values of header-footer regions.

[Akshay0799]

Looks like the library looks at text directly after extraction. But it would be nice to fetch the bounding box values for header/footer regions based on the algorithm's outcome to define th scope of header footer regions so that those regions can be eliminated during text extraction.

[lpi-tn]

What would be the advantage over what the library does today?

[Akshay0799]

What would be the advantage over what the library does today?

Right now, the library only does text -to -text matching and returns text values directly for the header-footers. Ideally since headers/footers are localized to a particular region in the page. It would nice to obtain the bbox region values so that there is a foolproof way to only remove text from that region when this component to linked to other parts of the codebase.

[lpi-tn]

Isee what's you mean but I have the feeling it's would be almost a complete different code