X-Git-Url: http://git.scottworley.com/paperdoorknob/blobdiff_plain/1fac41bf2d403d35d43dd532a37a8225f682bae2..53f396b459889b46e6feea48e8fa8e89b2452e9f:/glowfic.py

diff --git a/glowfic.py b/glowfic.py
index 012bd06..81a5342 100644
--- 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 +55,48 @@ 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:
+        def find_next_thread(dom: BeautifulSoup) -> str | None:
+            for c in dom.findChildren('div', class_='post-navheader'):
+                for a in c.findChildren('a'):
+                    if 'Next Post' in a.text and 'href' in a.attrs and isinstance(
+                            a.attrs['href'], str):
+                        return a.attrs['href']
+            return None
+
+        spec.log('Fetching HTML...\r')
+        html = spec.fetcher.fetch(spec.url)
+        flat_html = spec.fetcher.fetch(flatURL(spec.url))
+        spec.log('Parsing HTML...\r')
+        self._next_thread = find_next_thread(
+            BeautifulSoup(spec.htmlfilter(html), 'html.parser'))
+        self._dom = BeautifulSoup(spec.htmlfilter(flat_html), 'html.parser')
+
+    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 next_thread(self) -> str | None:
+        return self._next_thread
+
+    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())
+        return itertools.chain([text()], the_replies())
 
 
 def makeChunk(chunk_dom: Tag, image_store: ImageStore) -> Chunk: