Add .extract_table(...) logic to avoid assigning characters to multiple cells

[jsvine]

(See #1006 (reply in thread) for context.)

Currently, Page.extract_table(...) uses .crop(...) internally for each cell, capturing all characters that overlap at all with the cell's bounding box. So if a character straddles multiple cells in a table, its text would appear in multiple cells of the .extract_table(...) output. This seems undesirable. Perhaps the method should either:

(a) "keep track" of characters it has already assigned to a cell (and then not use them again in another cell), or

(b) only assign characters to cells if they are more than 50% inside? (And then what to do if a cell is perfectly divided 50/50 between two cells? Or 30/30/40 across three cells?)

Thoughts / suggestions / other approaches?

[jsvine]

Whoops, my assessment was incorrect here, and based on an assumption I failed to double-check in the code. The good news is we already handle this in a reasonable-seeming way:

pdfplumber/pdfplumber/table.py

Lines 399 to 410 in 94da66c

	def extract(self, **kwargs: Any) -> List[List[Optional[str]]]:
	chars = self.page.chars
	table_arr = []
	def char_in_bbox(char: T_obj, bbox: T_bbox) -> bool:
	v_mid = (char["top"] + char["bottom"]) / 2
	h_mid = (char["x0"] + char["x1"]) / 2
	x0, top, x1, bottom = bbox
	return bool(
	(h_mid >= x0) and (h_mid < x1) and (v_mid >= top) and (v_mid < bottom)
	)

Thanks to @cmdlineluser in #1006 (reply in thread)