Strip all  

[paperdoorknob] / paperdoorknob.py

diff --git a/paperdoorknob.py b/paperdoorknob.py
index b7e4349b98ab1d97ae2ff0ad6f50e4740bb6440f..34223e4c35eb7636370c1032764de6455c075933 100644 (file)
--- a/paperdoorknob.py
+++ b/paperdoorknob.py
@@ -5,37 +5,56 @@
 # Free Software Foundation, version 3.
 
 
-from argparse import ArgumentParser
-import os.path
-import requests
-import requests_cache
-from xdg_base_dirs import xdg_cache_home
+import itertools
 
+from typing import Iterable
 
-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(
-        '--timeout',
-        help='How long to wait for HTTP requests, in seconds',
-        default=30)
-    parser.add_argument('url', help='URL to retrieve')
-    return parser
+from bs4 import BeautifulSoup
+from bs4.element import Tag
 
+from args import spec_from_commandline_args
+from spec import Spec
 
-def fetch(url: str, session: requests.Session, timeout: int) -> None:
-    with session.get(url, timeout=timeout) as r:
-        r.raise_for_status()
+
+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(spec: Spec) -> None:
+    spec.texout.write(b'\\documentclass{article}\n\\begin{document}\n')
+    html = clean(parse(spec.htmlfilter(spec.fetcher.fetch(spec.url))))
+    for r in replies(html):
+        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:
-        fetch(args.url, session, args.timeout)
+    with spec_from_commandline_args() as spec:
+        process(spec)
 
 
 if __name__ == '__main__':