git.scottworley.com Git - paperdoorknob/blob - paperdoorknob.py

   1 # paperdoorknob: Print glowfic
   2 #
   3 # This program is free software: you can redistribute it and/or modify it
   4 # under the terms of the GNU General Public License as published by the
   5 # Free Software Foundation, version 3.
   6
   7
   8 from argparse import ArgumentParser
   9 import itertools
  10 import os.path
  11 import subprocess
  12
  13 from typing import IO, Iterable
  14
  15 from bs4 import BeautifulSoup
  16 from bs4.element import Tag
  17 from xdg_base_dirs import xdg_cache_home
  18
  19 from fetch import CachingFetcher, Fetcher
  20
  21
  22 def command_line_parser() -> ArgumentParser:
  23     parser = ArgumentParser(prog='paperdoorknob', description='Print glowfic')
  24     parser.add_argument(
  25         '--cache_path',
  26         metavar='PATH',
  27         help='Where to keep the http cache (instead of %(default)s)',
  28         default=os.path.join(xdg_cache_home(), "paperdoorknob"))
  29     parser.add_argument(
  30         '--out',
  31         help='The filename stem at which to write output ' +
  32              '(eg: "%(default)s" produces %(default)s.text, %(default)s.pdf, etc.)',
  33         default='book')
  34     parser.add_argument('--pandoc', help='Location of the pandoc executable')
  35     parser.add_argument(
  36         '--timeout',
  37         help='How long to wait for HTTP requests, in seconds',
  38         default=30)
  39     parser.add_argument('url', help='URL to retrieve')
  40     return parser
  41
  42
  43 def parse(content: bytes) -> BeautifulSoup:
  44     return BeautifulSoup(content, 'html.parser')
  45
  46
  47 def clean(html: BeautifulSoup) -> BeautifulSoup:
  48     for eb in html.find_all("div", class_="post-edit-box"):
  49         eb.decompose()
  50     for footer in html.find_all("div", class_="post-footer"):
  51         footer.decompose()
  52     return html
  53
  54
  55 def replies(html: BeautifulSoup) -> Iterable[Tag]:
  56     def text() -> Tag:
  57         body = html.body
  58         assert body
  59         text = body.find_next("div", class_="post-post")
  60         assert isinstance(text, Tag)
  61         return text
  62
  63     def the_replies() -> Iterable[Tag]:
  64         rs = html.find_all("div", class_="post-reply")
  65         assert all(isinstance(r, Tag) for r in rs)
  66         return rs
  67
  68     return itertools.chain([text()], the_replies())
  69
  70
  71 def html_to_tex(pandoc: str, tag: Tag) -> bytes:
  72     return subprocess.run([pandoc, '--from=html', '--to=latex'],
  73                           input=tag.encode(),
  74                           stdout=subprocess.PIPE,
  75                           check=True).stdout
  76
  77
  78 def process(
  79         url: str,
  80         fetcher: Fetcher,
  81         texout: IO[bytes],
  82         pandoc: str) -> None:
  83     texout.write(b'\\documentclass{article}\n\\begin{document}\n')
  84     html = clean(parse(fetcher.fetch(url)))
  85     for r in replies(html):
  86         texout.write(html_to_tex(pandoc, r))
  87     texout.write(b'\\end{document}\n')
  88
  89
  90 def main() -> None:
  91     args = command_line_parser().parse_args()
  92     with CachingFetcher(args.cache_path, args.timeout) as fetcher:
  93         with open(args.out + '.tex', 'wb') as texout:
  94             process(args.url, fetcher, texout, args.pandoc or 'pandoc')
  95
  96
  97 if __name__ == '__main__':
  98     main()