More structure and tests around splitting the page into chunks' DOMs.

[paperdoorknob] / glowfic.py

# paperdoorknob: Print glowfic
#
# This program is free software: you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by the
# Free Software Foundation, version 3.


import itertools

from typing import Iterable

from bs4 import BeautifulSoup
from bs4.element import Tag

# We avoid the name "post" because the Glowfic community uses the term
# inconsistently:
#  * The Glowfic software sometimes uses "post" to refer to a whole thread
#    (eg: in the URL), but more often uses "post" to refer to just the first
#    chunk in a thread.  The non-first chunks are "replies".
#  * Readers and this software don't need to distinguish first-chunks and
#    non-first-chunks.
#  * Humans in the community tend to use "posts" to mean "chunks" ("replies"
#    in the Glowfic software's lexicon).


def chunkDOMs(html: BeautifulSoup) -> Iterable[Tag]:
    def text() -> Tag:
        body = html.body
        assert body
        text = body.find_next("div", class_="post-post")
        assert isinstance(text, Tag)
        return text

    def the_replies() -> Iterable[Tag]:
        rs = html.find_all("div", class_="post-reply")
        assert all(isinstance(r, Tag) for r in rs)
        return rs

    return itertools.chain([text()], the_replies())