[RAFT] The chunking function doesn't separate PDF document pages in the chunks

[cedricvidal]

In the get_chunks function, the PDF pages are not separated, yielding chunks where spaces, dots or new lines are missing.

gorilla/raft/raft.py

Lines 90 to 92 in 2fc82a9

	for page_num in range(num_pages):
	page = reader.pages[page_num]
	text += page.extract_text()

Unclear how to properly handle this situation in a generic way as depending on the context, this might require either a space, a new line or a dot.

This may impact the quality of the generated dataset.