More structure and tests around splitting the page into chunks' DOMs.

[paperdoorknob] / paperdoorknob.py

diff --git a/paperdoorknob.py b/paperdoorknob.py
index 34223e4c35eb7636370c1032764de6455c075933..84ed70228d3d9b4040fc765c3ced2be68d1d58ea 100644 (file)
--- a/paperdoorknob.py
+++ b/paperdoorknob.py
@@ -5,14 +5,10 @@
 # Free Software Foundation, version 3.
 
 
-import itertools
-
-from typing import Iterable
-
 from bs4 import BeautifulSoup
-from bs4.element import Tag
 
 from args import spec_from_commandline_args
+from glowfic import chunkDOMs
 from spec import Spec
 
 
@@ -20,34 +16,16 @@ 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(b'\\documentclass{article}\n')
+    if spec.geometry is not None:
+        spec.texout.write(b'\\usepackage[' +
+                          spec.geometry.encode('UTF-8') +
+                          b']{geometry}\n')
+    spec.texout.write(b'\\begin{document}\n')
+    html = parse(spec.htmlfilter(spec.fetcher.fetch(spec.url)))
+    for r in chunkDOMs(html):
+        spec.domfilter(r)
         spec.texout.write(spec.texifier.texify(r))
     spec.texout.write(b'\\end{document}\n')