Extensible, flag-controlled DOM filters

[paperdoorknob] / args.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.


from argparse import ArgumentParser
from contextlib import contextmanager
import os.path

from typing import Iterator

from xdg_base_dirs import xdg_cache_home

from domfilter import ApplyDOMFilters, DOMFilters
from fetch import CachingFetcher
from htmlfilter import ApplyHTMLFilters, HTMLFilters
from spec import Spec
from texify import PandocTexifier


def _command_line_parser() -> ArgumentParser:
    parser = ArgumentParser(prog='paperdoorknob', description='Print glowfic')
    parser.add_argument(
        '--cache_path',
        metavar='PATH',
        help='Where to keep the http cache (instead of %(default)s)',
        default=os.path.join(xdg_cache_home(), "paperdoorknob"))
    parser.add_argument(
        '--domfilters',
        help='Which DOM filters to use (default: %(default)s)',
        default=','.join(f[0] for f in DOMFilters))
    parser.add_argument(
        '--htmlfilters',
        help='Which HTML filters to use (default: %(default)s)',
        default=','.join(f[0] for f in HTMLFilters))
    parser.add_argument(
        '--out',
        help='The filename stem at which to write output ' +
             '(eg: "%(default)s" produces %(default)s.tex, %(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',
        default=30)
    parser.add_argument('url', help='URL to retrieve')
    return parser


@contextmanager
def spec_from_commandline_args() -> Iterator[Spec]:
    args = _command_line_parser().parse_args()
    with CachingFetcher(args.cache_path, args.timeout) as fetcher:
        with open(args.out + '.tex', 'wb') as texout:
            yield Spec(
                args.url,
                fetcher,
                lambda x: ApplyHTMLFilters(args.htmlfilters, x),
                lambda x: ApplyDOMFilters(args.domfilters, x),
                PandocTexifier(args.pandoc or 'pandoc'),
                texout)