Extensible, flag-controlled DOM filters

[paperdoorknob] / paperdoorknob.py

diff --git a/paperdoorknob.py b/paperdoorknob.py
index a31b659d2d403c67e0a44d4e32ebfd218168cc1d..2bcf31afe3c1b38407eec51f79babed956847fc9 100644 (file)
--- a/paperdoorknob.py
+++ b/paperdoorknob.py
@@ -5,52 +5,19 @@
 # Free Software Foundation, version 3.
 
 
-from argparse import ArgumentParser
 import itertools
-import os.path
 
-from typing import IO, Iterable
+from typing import Iterable
 
 from bs4 import BeautifulSoup
 from bs4.element import Tag
-import requests
-import requests_cache
-from xdg_base_dirs import xdg_cache_home
-
-
-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 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 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
+
+from args import spec_from_commandline_args
+from spec import Spec
+
+
+def parse(content: bytes) -> BeautifulSoup:
+    return BeautifulSoup(content, 'html.parser')
 
 
 def replies(html: BeautifulSoup) -> Iterable[Tag]:
@@ -69,22 +36,18 @@ def replies(html: BeautifulSoup) -> Iterable[Tag]:
     return itertools.chain([text()], the_replies())
 
 
-def process(
-        url: str,
-        session: requests.Session,
-        timeout: int,
-        texout: IO[str],
-) -> None:
-    html = clean(fetch(url, session, timeout))
-    replies(html)
-    print("soon", file=texout)
+def process(spec: Spec) -> None:
+    spec.texout.write(b'\\documentclass{article}\n\\begin{document}\n')
+    html = parse(spec.htmlfilter(spec.fetcher.fetch(spec.url)))
+    for r in replies(html):
+        spec.domfilter(r)
+        spec.texout.write(spec.texifier.texify(r))
+    spec.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:
-        with open(args.out + '.tex', 'w', encoding='UTF-8') as texout:
-            process(args.url, session, args.timeout, texout)
+    with spec_from_commandline_args() as spec:
+        process(spec)
 
 
 if __name__ == '__main__':