Update app.py

app.py

@@ -64,9 +64,15 @@ def extract_synopsis(description: str) -> str:
 
 
 def extract_title_and_description(html: str) -> tuple[str, str]:
-    """Pull the best available title and synopsis/description from a page's HTML."""
+    """Pull the best available title and synopsis/description from a page's HTML.
+
+    For WebNovel pages the meta description is deliberately truncated with '...'.
+    The full synopsis lives inside a <div class="... j_synopsis ..."> in the body.
+    We prefer that element; the meta tags are only a fallback for other sites.
+    """
     soup = BeautifulSoup(html, "html.parser")
 
+    # --- Title ---
     title = ""
     og_title = soup.find("meta", property="og:title")
     if og_title and og_title.get("content"):
@@ -74,20 +80,47 @@ def extract_title_and_description(html: str) -> tuple[str, str]:
     if not title and soup.title and soup.title.string:
         title = soup.title.string.strip()
 
-    description = ""
-    meta_desc = soup.find("meta", attrs={"name": "description"})
-    if meta_desc and meta_desc.get("content"):
-        description = meta_desc["content"].strip()
-    if not description:
+    # --- Synopsis: try full body element first (WebNovel and similar sites) ---
+    synopsis = ""
+
+    # 1. WebNovel: <div class="... j_synopsis ...">
+    j_syn = soup.find(class_=lambda c: c and "j_synopsis" in c.split())
+    if j_syn:
+        # Replace <br> tags with newlines before extracting text
+        for br in j_syn.find_all("br"):
+            br.replace_with("\n")
+        # Remove the "read more" toggle span that WebNovel appends
+        for rm in j_syn.find_all("span", class_="_readmore"):
+            rm.decompose()
+        synopsis = j_syn.get_text(separator=" ").strip()
+
+    # 2. Fallback: any element whose class contains common synopsis keywords
+    if not synopsis:
+        for cls_hint in ("book-intro", "book_intro", "bookIntro",
+                         "detail-desc", "book-desc", "bookDesc",
+                         "synopsis", "j_intro"):
+            el = soup.find(class_=lambda c: c and cls_hint in c)
+            if el:
+                for br in el.find_all("br"):
+                    br.replace_with("\n")
+                synopsis = el.get_text(separator=" ").strip()
+                if synopsis:
+                    break
+
+    # 3. Last resort: meta tags (always truncated on WebNovel, but better than nothing)
+    if not synopsis:
+        meta_desc = soup.find("meta", attrs={"name": "description"})
+        if meta_desc and meta_desc.get("content"):
+            synopsis = extract_synopsis(meta_desc["content"].strip())
+    if not synopsis:
         og_desc = soup.find("meta", property="og:description")
         if og_desc and og_desc.get("content"):
-            description = og_desc["content"].strip()
+            synopsis = extract_synopsis(og_desc["content"].strip())
 
-    # Keep everything on a single line in the combined output
-    title = re.sub(r"\s+", " ", title)
-    description = re.sub(r"\s+", " ", description)
+    # Normalise whitespace: collapse runs of spaces/tabs but keep newlines
+    title = re.sub(r"[ \t]+", " ", title).strip()
+    synopsis = re.sub(r"[ \t]+", " ", synopsis).strip()
 
-    synopsis = extract_synopsis(description)
     return title, synopsis