]> git.scottworley.com Git - paperdoorknob/blobdiff - paperdoorknob.py
Drop Post as a class
[paperdoorknob] / paperdoorknob.py
index bb8bdd1ae38a660b3a6f90aa3d5aa317505beee3..8329f83e69e026791f6997a296a10af19ab4d2e1 100644 (file)
@@ -6,8 +6,13 @@
 
 
 from argparse import ArgumentParser
+import itertools
 import os.path
+
+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
@@ -34,10 +39,42 @@ def fetch(url: str, session: requests.Session, timeout: int) -> BeautifulSoup:
         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
+
+
+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,
+        session: requests.Session,
+        timeout: int) -> Iterable[Tag]:
+    html = clean(fetch(url, session, timeout))
+    return replies(html)
+
+
 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)
+        process(args.url, session, args.timeout)
 
 
 if __name__ == '__main__':