Always have Thread.__init__ fetch the HTML

[paperdoorknob] / glowfic.py

diff --git a/glowfic.py b/glowfic.py
index 012bd069aca5a3066b443640bc66904d796e44ee..5ffaac31921e360961967832ad090074e5cd2aaa 100644 (file)
--- a/glowfic.py
+++ b/glowfic.py
@@ -15,6 +15,7 @@ from bs4 import BeautifulSoup
 from bs4.element import Tag
 
 from images import ImageStore
+from spec import Spec
 from texify import Texifier
 
 
@@ -54,20 +55,34 @@ class Chunk:
 #  * Humans in the community tend to use "posts" to mean chunks.
 
 
-def chunkDOMs(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
+class Thread:
 
-    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 __init__(self, spec: Spec) -> None:
+        spec.log('Fetching HTML...\r')
+        html = spec.fetcher.fetch(flatURL(spec.url))
+        spec.log('Parsing HTML...\r')
+        self._dom = BeautifulSoup(spec.htmlfilter(html), 'html.parser')
 
-    return itertools.chain([text()], the_replies())
+    def title(self) -> str | None:
+        span = self._dom.findChild("span", id="post-title")
+        if not isinstance(span, Tag):
+            return None
+        return span.text.strip()
+
+    def chunkDOMs(self) -> Iterable[Tag]:
+        def text() -> Tag:
+            body = self._dom.body
+            assert body
+            text = body.find_next("div", class_="post-post")
+            assert isinstance(text, Tag)
+            return text
+
+        def the_replies() -> Iterable[Tag]:
+            rs = self._dom.find_all("div", class_="post-reply")
+            assert all(isinstance(r, Tag) for r in rs)
+            return rs
+
+        return itertools.chain([text()], the_replies())
 
 
 def makeChunk(chunk_dom: Tag, image_store: ImageStore) -> Chunk: