from argparse import ArgumentParser
import itertools
import os.path
+import subprocess
-from typing import Iterable
+from typing import IO, Iterable
from bs4 import BeautifulSoup
from bs4.element import Tag
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',
return itertools.chain([text()], the_replies())
+def html_to_tex(pandoc: str, tag: Tag) -> bytes:
+ return subprocess.run([pandoc, '--from=html', '--to=latex'],
+ input=tag.encode(),
+ stdout=subprocess.PIPE,
+ check=True).stdout
+
+
def process(
url: str,
session: requests.Session,
- timeout: int) -> Iterable[Tag]:
+ timeout: int,
+ texout: IO[bytes],
+ pandoc: str,
+) -> None:
html = clean(fetch(url, session, timeout))
- return replies(html)
+ for r in replies(html):
+ texout.write(html_to_tex(pandoc, r))
def main() -> None:
args = command_line_parser().parse_args()
with requests_cache.CachedSession(args.cache_path, cache_control=True) as session:
- process(args.url, session, args.timeout)
+ with open(args.out + '.tex', 'wb') as texout:
+ process(
+ args.url,
+ session,
+ args.timeout,
+ texout,
+ args.pandoc or 'pandoc')
if __name__ == '__main__':