Update app.py

app.py

@@ -15,70 +15,69 @@ faiss_index = None  # Renamed from 'index' to 'faiss_index'
 bookmarks = []
 fetch_cache = {}
 
-# Helper functions remain the same...
+# Helper functions
 
 def parse_bookmarks(file_content):
-    # [Same as before]
+    soup = BeautifulSoup(file_content, 'html.parser')
+    extracted_bookmarks = []
+    for link in soup.find_all('a'):
+        url = link.get('href')
+        title = link.text
+        if url and title:
+            extracted_bookmarks.append({'url': url, 'title': title})
+    return extracted_bookmarks
 
 def fetch_url_info(bookmark):
-    # [Same as before]
+    url = bookmark['url']
+    if url in fetch_cache:
+        bookmark.update(fetch_cache[url])
+        return bookmark
+
+    try:
+        response = requests.get(url, timeout=5)
+        bookmark['etag'] = response.headers.get('ETag', 'N/A')
+        bookmark['status_code'] = response.status_code
+
+        if response.status_code >= 400:
+            bookmark['dead_link'] = True
+            bookmark['content'] = ''
+        else:
+            bookmark['dead_link'] = False
+            soup = BeautifulSoup(response.content, 'html.parser')
+            meta_tags = {meta.get('name', ''): meta.get('content', '') for meta in soup.find_all('meta')}
+            bookmark['meta_tags'] = meta_tags
+            bookmark['content'] = soup.get_text(separator=' ', strip=True)
+    except Exception as e:
+        bookmark['dead_link'] = True
+        bookmark['etag'] = 'N/A'
+        bookmark['status_code'] = 'N/A'
+        bookmark['meta_tags'] = {}
+        bookmark['content'] = ''
+    finally:
+        fetch_cache[url] = {
+            'etag': bookmark.get('etag'),
+            'status_code': bookmark.get('status_code'),
+            'dead_link': bookmark.get('dead_link'),
+            'meta_tags': bookmark.get('meta_tags'),
+            'content': bookmark.get('content'),
+        }
+    return bookmark
 
 def generate_summary(bookmark):
-    # [Same as before]
+    content = bookmark.get('content', '')
+    if content:
+        # Limit content to first 2000 characters to save resources
+        content = content[:2000]
+        summary = summarizer(content, max_length=50, min_length=25, do_sample=False)
+        bookmark['summary'] = summary[0]['summary_text']
+    else:
+        bookmark['summary'] = 'No content available to summarize.'
+    return bookmark
 
 def vectorize_and_index(bookmarks):
-    # [Same as before]
-
-def process_uploaded_file(file):
-    global bookmarks, faiss_index
-    if file is None:
-        return "Please upload a bookmarks HTML file."
-
-    # Since 'file' is now bytes, decode it directly
-    file_content = file.decode('utf-8')
-    bookmarks = parse_bookmarks(file_content)
-
-    for bookmark in bookmarks:
-        fetch_url_info(bookmark)
-        generate_summary(bookmark)
-
-    faiss_index, embeddings = vectorize_and_index(bookmarks)
-    return f"Successfully processed {len(bookmarks)} bookmarks."
-
-def chatbot_response(user_query):
-    # [Same as before]
-
-def display_bookmarks():
-    # [Same as before]
-
-def edit_bookmark(bookmark_idx, new_title, new_url):
-    # [Same as before]
-
-def delete_bookmark(bookmark_idx):
-    # [Same as before]
-
-def build_app():
-    with gr.Blocks() as demo:
-        gr.Markdown("# Bookmark Manager App")
-
-        with gr.Tab("Upload and Process Bookmarks"):
-            upload = gr.File(label="Upload Bookmarks HTML File", type='bytes')  # Updated here
-            process_button = gr.Button("Process Bookmarks")
-            output_text = gr.Textbox(label="Output")
-
-            process_button.click(
-                process_uploaded_file,
-                inputs=upload,
-                outputs=output_text
-            )
-
-        with gr.Tab("Chat with Bookmarks"):
-            # [Same as before]
-
-        with gr.Tab("Manage Bookmarks"):
-            # [Same as before]
-
-    demo.launch()
-
-if __name__ == "__main__":
-    build_app()
+    summaries = [bookmark['summary'] for bookmark in bookmarks]
+    embeddings = embedding_model.encode(summaries)
+    dimension = embeddings.shape[1]
+    faiss_idx = faiss.IndexFlatL2(dimension)
+    faiss_idx.add(np.array(embeddings))
+    return f