1

srt_downloader.py

@@ -64,14 +64,27 @@ def parse_knowledge_entries(html_path: Path) -> List[KnowledgeEntry]:
     """Scan the saved HTML page and extract all knowledge entries."""
     soup = BeautifulSoup(html_path.read_text(encoding="utf-8"), "html.parser")
     entries: Dict[str, KnowledgeEntry] = {}
-    for span in soup.select("span.posCatalog_name"):
-        onclick = span.get("onclick", "")
-        match = _ONCLICK_RE.search(onclick)
-        if not match:
+    for div in soup.select("div.posCatalog_select"):
+        div_id = div.get("id", "")
+        if not div_id.startswith("cur"):
             continue
-        knowledge_id = match["knowledge"]
-        title_attr = span.get("title") or span.get_text(strip=True)
-        title = " ".join(title_attr.split())
+        knowledge_id = div_id[3:]
+        span = div.select_one("span.posCatalog_name")
+        if not span:
+            continue
+        
+        title_attr = span.get("title")
+        em = span.select_one("em.posCatalog_sbar")
+        
+        if title_attr:
+            if em:
+                title = f"{em.get_text(strip=True)} {title_attr}"
+            else:
+                title = title_attr
+        else:
+            title = span.get_text(strip=True)
+            
+        title = " ".join(title.split())
         if knowledge_id not in entries:
             entries[knowledge_id] = KnowledgeEntry(knowledge_id=knowledge_id, title=title)
     return list(entries.values())