Extensible, flag-controlled DOM filters

[paperdoorknob] / paperdoorknob.py

diff --git a/paperdoorknob.py b/paperdoorknob.py
index 6cc76a2f56d465fdf0c897b601dfbaa25dd86d03..2bcf31afe3c1b38407eec51f79babed956847fc9 100644 (file)
--- a/paperdoorknob.py
+++ b/paperdoorknob.py
@@ -5,65 +5,49 @@
 # Free Software Foundation, version 3.
 
 
 # Free Software Foundation, version 3.
 
 

-from argparse import ArgumentParser

 import itertools
 import itertools

-import os.path

 
 from typing import Iterable
 
 from bs4 import BeautifulSoup
 from bs4.element import Tag
 
 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

 
 

+from args import spec_from_commandline_args
+from spec import Spec

 
 

-class Post:
-    def __init__(self, html: BeautifulSoup) -> None:
-        self._html = html

 
 

-    def text(self) -> Tag:
-        body = self._html.body
+def parse(content: bytes) -> BeautifulSoup:
+    return BeautifulSoup(content, 'html.parser')
+
+
+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
 
         assert body
         text = body.find_next("div", class_="post-post")
         assert isinstance(text, Tag)
         return text
 

-    def replies(self) -> Iterable[Tag]:
-        replies = self._html.find_all("div", class_="post-reply")
-        assert all(isinstance(r, Tag) for r in replies)
-        return replies
-
-    def entries(self) -> Iterable[Tag]:
-        return itertools.chain([self.text()], self.replies())
-
+    def the_replies() -> Iterable[Tag]:
+        rs = html.find_all("div", class_="post-reply")
+        assert all(isinstance(r, Tag) for r in rs)
+        return rs

 
 

-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
+    return itertools.chain([text()], the_replies())

 
 
 
 

-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 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:
 
 
 def main() -> None:

-    args = command_line_parser().parse_args()
-    with requests_cache.CachedSession(args.cache_path, cache_control=True) as session:
-        html = fetch(args.url, session, args.timeout)
-        Post(html)
+    with spec_from_commandline_args() as spec:
+        process(spec)

 
 
 if __name__ == '__main__':
 
 
 if __name__ == '__main__':