X-Git-Url: http://git.scottworley.com/paperdoorknob/blobdiff_plain/79631507ecb69945b7cedb10c3aaed4b8dd788b0..refs/heads/main:/paperdoorknob.py diff --git a/paperdoorknob.py b/paperdoorknob.py index 3b86cb8..5ad341f 100644 --- a/paperdoorknob.py +++ b/paperdoorknob.py @@ -4,88 +4,42 @@ # under the terms of the GNU General Public License as published by the # Free Software Foundation, version 3. - -from argparse import ArgumentParser -import itertools -import os.path - -from typing import IO, Iterable - -from bs4 import BeautifulSoup -from bs4.element import Tag -from xdg_base_dirs import xdg_cache_home - -from fetch import CachingFetcher, Fetcher -from texify import PandocTexifier, Texifier - - -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( - '--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', - default=30) - parser.add_argument('url', help='URL to retrieve') - return 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') +from args import spec_from_commandline_args +from glowfic import Thread +from spec import Spec + + +def process(spec: Spec) -> None: + spec.texout.write(br'''\documentclass{article} +\usepackage{amssymb} +\usepackage{booktabs} +\usepackage{graphicx} +\usepackage{longtable} +\usepackage{soul} +\usepackage{varwidth} +\usepackage{wrapstuff} +''') + if spec.geometry is not None: + spec.texout.write(br'\usepackage[' + + spec.geometry.encode('UTF-8') + + b']{geometry}\n') + spec.texout.write(br'''\begin{document} +\newcommand{\href}[2]{#2\footnote{\detokenize{#1}}} +\def\glowiconsize{%fmm} +\newcommand{\glowicon}[1]{\includegraphics[ + width=\glowiconsize,height=\glowiconsize,keepaspectratio +]{#1}} +\newcommand{\ifnotempty}[2]{\expandafter\ifx\expandafter\relax + \detokenize{#1}\relax\else #2\fi} +%s +''' % (spec.icon_size, spec.layout)) + Thread(spec).emit() + spec.texout.write(b'\\end{document}\n') def main() -> None: - args = command_line_parser().parse_args() - 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) + with spec_from_commandline_args() as spec: + process(spec) if __name__ == '__main__':