X-Git-Url: http://git.scottworley.com/paperdoorknob/blobdiff_plain/170e50a0efa18ed0f5f2ff644793469894983029..f1dec72014657ab33bffa8f68064d85cc862ea65:/paperdoorknob.py diff --git a/paperdoorknob.py b/paperdoorknob.py index fb2df1f..3b86cb8 100644 --- a/paperdoorknob.py +++ b/paperdoorknob.py @@ -9,37 +9,14 @@ from argparse import ArgumentParser import itertools import os.path -from typing import Iterable +from typing import IO, Iterable from bs4 import BeautifulSoup from bs4.element import Tag -import requests -import requests_cache from xdg_base_dirs import xdg_cache_home - -class Post: - def __init__(self, html: BeautifulSoup) -> None: - self._html = html - for eb in self._html.find_all("div", class_="post-edit-box"): - eb.decompose() - for footer in self._html.find_all("div", class_="post-footer"): - footer.decompose() - - def text(self) -> Tag: - body = self._html.body - assert body - text = body.find_next("div", class_="post-post") - assert isinstance(text, Tag) - return text - - def replies(self) -> Iterable[Tag]: - replies = self._html.find_all("div", class_="post-reply") - assert all(isinstance(r, Tag) for r in replies) - return replies - - def entries(self) -> Iterable[Tag]: - return itertools.chain([self.text()], self.replies()) +from fetch import CachingFetcher, Fetcher +from texify import PandocTexifier, Texifier def command_line_parser() -> ArgumentParser: @@ -49,6 +26,12 @@ def command_line_parser() -> ArgumentParser: metavar='PATH', help='Where to keep the http cache (instead of %(default)s)', default=os.path.join(xdg_cache_home(), "paperdoorknob")) + parser.add_argument( + '--out', + help='The filename stem at which to write output ' + + '(eg: "%(default)s" produces %(default)s.text, %(default)s.pdf, etc.)', + default='book') + parser.add_argument('--pandoc', help='Location of the pandoc executable') parser.add_argument( '--timeout', help='How long to wait for HTTP requests, in seconds', @@ -57,17 +40,52 @@ def command_line_parser() -> ArgumentParser: return parser -def fetch(url: str, session: requests.Session, timeout: int) -> BeautifulSoup: - with session.get(url, timeout=timeout) as r: - r.raise_for_status() - return BeautifulSoup(r.text, 'html.parser') +def parse(content: bytes) -> BeautifulSoup: + return BeautifulSoup(content, 'html.parser') + + +def clean(html: BeautifulSoup) -> BeautifulSoup: + for eb in html.find_all("div", class_="post-edit-box"): + eb.decompose() + for footer in html.find_all("div", class_="post-footer"): + footer.decompose() + return html + + +def replies(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()) + + +def process( + url: str, + fetcher: Fetcher, + texifier: Texifier, + texout: IO[bytes]) -> None: + texout.write(b'\\documentclass{article}\n\\begin{document}\n') + html = clean(parse(fetcher.fetch(url))) + for r in replies(html): + texout.write(texifier.texify(r)) + texout.write(b'\\end{document}\n') def main() -> None: args = command_line_parser().parse_args() - with requests_cache.CachedSession(args.cache_path, cache_control=True) as session: - html = fetch(args.url, session, args.timeout) - Post(html) + texifier = PandocTexifier(args.pandoc or 'pandoc') + with CachingFetcher(args.cache_path, args.timeout) as fetcher: + with open(args.out + '.tex', 'wb') as texout: + process(args.url, fetcher, texifier, texout) if __name__ == '__main__':