Update app.py

app.py

@@ -9,7 +9,7 @@ The app will:
   4. Produce one combined .txt file with: URL | Title | Description
   5. Zip all the saved .html files for download
 
-Built for Gradio on Hugging Face .
+Built for Gradio on Hugging Face Spaces.
 """
 
 import os
@@ -50,8 +50,21 @@ def slugify_for_filename(url: str, index: int) -> str:
     return f"{index:05d}_{path}.html"
 
 
+def extract_synopsis(description: str) -> str:
+    """WebNovel-style descriptions read like:
+    'Read X Online for Free, written by ... and the synopsis is: <actual synopsis>'
+    Strip everything before the marker so only the real synopsis remains.
+    If the marker isn't found (non-WebNovel page), just return the description as-is.
+    """
+    marker = "the synopsis is:"
+    idx = description.lower().find(marker)
+    if idx != -1:
+        return description[idx + len(marker):].strip()
+    return description.strip()
+
+
 def extract_title_and_description(html: str) -> tuple[str, str]:
-    """Pull the best available title and description from a page's HTML."""
+    """Pull the best available title and synopsis/description from a page's HTML."""
     soup = BeautifulSoup(html, "html.parser")
 
     title = ""
@@ -73,7 +86,9 @@ def extract_title_and_description(html: str) -> tuple[str, str]:
     # Keep everything on a single line in the combined output
     title = re.sub(r"\s+", " ", title)
     description = re.sub(r"\s+", " ", description)
-    return title, description
+
+    synopsis = extract_synopsis(description)
+    return title, synopsis
 
 
 def fetch_one(index: int, url: str, save_html: bool, timeout: int):
@@ -186,7 +201,7 @@ def run_scraper(
         for res in results:
             f.write(f"URL: {res['url']}\n")
             f.write(f"Title: {res['title']}\n")
-            f.write(f"Description: {res['description']}\n")
+            f.write(f"Synopsis: {res['description']}\n")
             f.write(f"Status: {res['status']}\n")
             f.write("-" * 60 + "\n")