Update app.py

app.py

@@ -34,22 +34,22 @@ def clean_response_text(response_text):
         cleaned_text = cleaned_text[:-len("```")].strip()
     return cleaned_text
 
-def generate_ideas(tag):
+def generate_ideas(user_input):
     """
-    Generate a diverse set of ideas related to the tag using the LLM.
+    Generate a diverse set of ideas based on the user's input concept using the LLM.
     
     Args:
-        tag (str): The tag to base the ideas on.
+        user_input (str): The user's input concept or idea (e.g., "blindfolded Rubik's Cube challenge").
     
     Returns:
         list: A list of ideas as strings.
     """
     prompt = f"""
-    Generate a list of 5 diverse and creative ideas related to {tag} that can be used for a TikTok video.
-    Each idea should be a short sentence describing a specific scene or concept.
+    The user has provided the concept: "{user_input}". Based on this concept, generate a list of 5 diverse and creative ideas for a TikTok video that directly relate to "{user_input}".
+    Each idea should be a short sentence describing a specific scene or concept, ensuring the core theme of "{user_input}" is central to each idea.
     Return the response as a JSON object with a single key 'ideas' containing a list of 5 ideas.
     Ensure the response is strictly in JSON format.
-    Example: {{"ideas": ["A neon-lit gaming setup with RGB lights flashing", "A futuristic robot assembling a gadget"]}}
+    Example: {{"ideas": ["A neon-lit Rubik's Cube challenge in the dark", "A blindfolded speedcubing competition with dramatic music"]}}
     """
     try:
         response = client.models.generate_content(
@@ -68,19 +68,19 @@ def generate_ideas(tag):
     except Exception as e:
         print(f"Error generating ideas: {e}")
         return [
-            f"A vibrant {tag} scene at sunset",
-            f"A close-up of {tag} with neon lights",
-            f"A futuristic take on {tag} with holograms",
-            f"A cozy {tag} moment with warm lighting",
-            f"An action-packed {tag} scene with dynamic colors"
+            f"A dramatic {user_input} scene with cinematic lighting",
+            f"A close-up of {user_input} in a futuristic setting",
+            f"A high-energy {user_input} moment with vibrant colors",
+            f"A serene {user_input} scene with soft focus",
+            f"An action-packed {user_input} challenge with dynamic angles"
         ]
 
-def generate_item(tag, ideas, generate_video=False, max_retries=3):
+def generate_item(user_input, ideas, generate_video=False, max_retries=3):
     """
     Generate a single feed item (image and optionally video) using one of the ideas.
     
     Args:
-        tag (str): The tag to base the content on.
+        user_input (str): The user's input concept or idea.
         ideas (list): List of ideas to choose from.
         generate_video (bool): Whether to generate a video from the image.
         max_retries (int): Maximum number of retries if image generation fails.
@@ -93,13 +93,13 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
     for attempt in range(max_retries):
         selected_idea = random.choice(ideas)
         prompt = f"""
-        Based on the idea "{selected_idea}", create content for a TikTok video about {tag}.
+        The user has provided the concept: "{user_input}". Based on this concept and the specific idea "{selected_idea}", create content for a TikTok video.
         Return a JSON object with two keys:
-        - 'caption': A short, viral TikTok-style caption with hashtags.
-        - 'image_prompt': A detailed image prompt for generating a high-quality visual scene.
+        - 'caption': A short, viral TikTok-style caption with hashtags that reflects "{user_input}".
+        - 'image_prompt': A detailed image prompt for generating a high-quality visual scene, ensuring the theme of "{user_input}" is central.
         The image prompt should describe the scene vividly, specify a perspective and style, and ensure no text or letters are included.
         Ensure the response is strictly in JSON format.
-        Example: {{"caption": "Neon vibes only! 🌌 #tech", "image_prompt": "A close-up view of a neon-lit gaming setup with RGB lights flashing, in a futuristic style, no text or letters"}}
+        Example: {{"caption": "Blindfolded Rubik's Cube MAGIC! 🤯 #rubiks", "image_prompt": "A close-up view of a person solving a Rubik's Cube blindfolded, in a dramatic style, no text or letters"}}
         """
         try:
             response = client.models.generate_content(
@@ -118,8 +118,8 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
             image_prompt = response_json['image_prompt']
         except Exception as e:
             print(f"Error generating item (attempt {attempt + 1}): {e}")
-            text = f"Obsessed with {tag}! 🔥 #{tag}"
-            image_prompt = f"A vivid scene of {selected_idea}, in a vibrant pop art style, no text or letters"
+            text = f"Amazing {user_input}! 🔥 #{user_input.replace(' ', '')}"
+            image_prompt = f"A vivid scene of {selected_idea} related to {user_input}, in a vibrant pop art style, no text or letters"
 
         # Attempt to generate the image
         try:
@@ -148,7 +148,8 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
                     try:
                         # Enhance the image prompt for video generation
                         video_prompt = f"""
-                        {image_prompt} Create a close-up shot with a slow dolly shot circling around the subject, 
+                        The user concept is "{user_input}". Based on this and the scene: {image_prompt}, create a video.
+                        Use a close-up shot with a slow dolly shot circling around the subject, 
                         using shallow focus on the main subject to emphasize details, in a realistic style with cinematic lighting.
                         """
                         operation = client.models.generate_videos(
@@ -160,6 +161,7 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
                                 number_of_videos=2,
                                 duration_seconds=8,
                                 negative_prompt="blurry, low quality, text, letters",
+                                enhance_prompt=True
                             )
                         )
                         # Wait for videos to generate
@@ -167,6 +169,12 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
                             time.sleep(20)
                             operation = client.operations.get(operation)
                         
+                        # Check if operation succeeded and has a valid response
+                        if operation.response is None:
+                            raise ValueError("Video generation operation failed: No response")
+                        if not hasattr(operation.response, 'generated_videos'):
+                            raise ValueError("Video generation operation failed: No generated_videos in response")
+                        
                         for n, video in enumerate(operation.response.generated_videos):
                             fname = f'with_image_input{n}.mp4'
                             print(f"Generated video: {fname}")
@@ -200,7 +208,7 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
                         'ideas': ideas
                     }
                 # Retry with new ideas
-                ideas = generate_ideas(tag)
+                ideas = generate_ideas(user_input)
                 continue
         except Exception as e:
             print(f"Error generating image (attempt {attempt + 1}): {e}")
@@ -217,33 +225,33 @@ def generate_item(tag, ideas, generate_video=False, max_retries=3):
                     'ideas': ideas
                 }
             # Retry with new ideas
-            ideas = generate_ideas(tag)
+            ideas = generate_ideas(user_input)
             continue
 
-def start_feed(tag, generate_video, current_index, feed_items):
+def start_feed(user_input, generate_video, current_index, feed_items):
     """
-    Start or update the feed based on the tag.
+    Start or update the feed based on the user input.
     
     Args:
-        tag (str): The tag to generate content for.
+        user_input (str): The user's input concept or idea.
         generate_video (bool): Whether to generate a video.
         current_index (int): The current item index.
         feed_items (list): The current list of feed items.
     
     Returns:
-        tuple: (current_tag, current_index, feed_items, html_content, share_links, is_loading)
+        tuple: (current_user_input, current_index, feed_items, html_content, share_links, is_loading)
     """
-    if not tag.strip():
-        tag = "trending"
+    if not user_input.strip():
+        user_input = "trending"
     
     # Set loading state
     is_loading = True
-    html_content = generate_html([], False, 0, tag, is_loading)
+    html_content = generate_html([], False, 0, user_input, is_loading)
     share_links = ""
 
     try:
-        ideas = generate_ideas(tag)
-        item = generate_item(tag, ideas, generate_video=generate_video)
+        ideas = generate_ideas(user_input)
+        item = generate_item(user_input, ideas, generate_video=generate_video)
         feed_items = [item]
         current_index = 0
         share_links = generate_share_links(
@@ -274,41 +282,41 @@ def start_feed(tag, generate_video, current_index, feed_items):
         </div>
         """
         is_loading = False
-        return tag, current_index, feed_items, html_content, share_links, is_loading
+        return user_input, current_index, feed_items, html_content, share_links, is_loading
 
     # Set loading state to False and update UI
     is_loading = False
-    html_content = generate_html(feed_items, False, current_index, tag, is_loading)
-    return tag, current_index, feed_items, html_content, share_links, is_loading
+    html_content = generate_html(feed_items, False, current_index, user_input, is_loading)
+    return user_input, current_index, feed_items, html_content, share_links, is_loading
 
-def load_next(tag, generate_video, current_index, feed_items):
+def load_next(user_input, generate_video, current_index, feed_items):
     """
     Load the next item in the feed.
     
     Args:
-        tag (str): The tag to generate content for.
+        user_input (str): The user's input concept or idea.
         generate_video (bool): Whether to generate a video.
         current_index (int): The current item index.
         feed_items (list): The current list of feed items.
     
     Returns:
-        tuple: (current_tag, current_index, feed_items, html_content, share_links, is_loading)
+        tuple: (current_user_input, current_index, feed_items, html_content, share_links, is_loading)
     """
     is_loading = True
-    html_content = generate_html(feed_items, False, current_index, tag, is_loading)
+    html_content = generate_html(feed_items, False, current_index, user_input, is_loading)
     share_links = ""
 
     try:
         if current_index + 1 < len(feed_items):
             current_index += 1
         else:
-            ideas = feed_items[-1]['ideas'] if feed_items else generate_ideas(tag)
-            new_item = generate_item(tag, ideas, generate_video=generate_video)
+            ideas = feed_items[-1]['ideas'] if feed_items else generate_ideas(user_input)
+            new_item = generate_item(user_input, ideas, generate_video=generate_video)
             feed_items.append(new_item)
             current_index = len(feed_items) - 1
         share_links = generate_share_links(
             feed_items[current_index]['image_base64'],
-            feed_items[current_index]['video_base64_list'],  # Fixed typo: IRfeed_items -> feed_items
+            feed_items[current_index]['video_base64_list'],
             feed_items[current_index]['text']
         )
     except Exception as e:
@@ -332,34 +340,34 @@ def load_next(tag, generate_video, current_index, feed_items):
         </div>
         """
         is_loading = False
-        return tag, current_index, feed_items, html_content, share_links, is_loading
+        return user_input, current_index, feed_items, html_content, share_links, is_loading
 
     is_loading = False
-    html_content = generate_html(feed_items, False, current_index, tag, is_loading)
-    return tag, current_index, feed_items, html_content, share_links, is_loading
+    html_content = generate_html(feed_items, False, current_index, user_input, is_loading)
+    return user_input, current_index, feed_items, html_content, share_links, is_loading
 
-def load_previous(tag, generate_video, current_index, feed_items):
+def load_previous(user_input, generate_video, current_index, feed_items):
     """
     Load the previous item in the feed.
     
     Args:
-        tag (str): The tag to generate content for.
+        user_input (str): The user's input concept or idea.
         generate_video (bool): Whether to generate a video (not used here).
         current_index (int): The current item index.
         feed_items (list): The current list of feed items.
     
     Returns:
-        tuple: (current_tag, current_index, feed_items, html_content, share_links, is_loading)
+        tuple: (current_user_input, current_index, feed_items, html_content, share_links, is_loading)
     """
     if current_index > 0:
         current_index -= 1
-    html_content = generate_html(feed_items, False, current_index, tag, False)
+    html_content = generate_html(feed_items, False, current_index, user_input, False)
     share_links = generate_share_links(
         feed_items[current_index]['image_base64'],
         feed_items[current_index]['video_base64_list'],
         feed_items[current_index]['text']
     )
-    return tag, current_index, feed_items, html_content, share_links, False
+    return user_input, current_index, feed_items, html_content, share_links, False
 
 def generate_share_links(image_base64, video_base64_list, caption):
     """
@@ -460,7 +468,7 @@ def generate_share_links(image_base64, video_base64_list, caption):
     """
     return share_links
 
-def generate_html(feed_items, scroll_to_latest=False, current_index=0, tag="", is_loading=False):
+def generate_html(feed_items, scroll_to_latest=False, current_index=0, user_input="", is_loading=False):
     """
     Generate an HTML string to display the current feed item with click navigation.
     
@@ -468,18 +476,18 @@ def generate_html(feed_items, scroll_to_latest=False, current_index=0, tag="", i
         feed_items (list): List of dictionaries containing 'text' and 'image_base64'.
         scroll_to_latest (bool): Whether to auto-scroll to the latest item (not used here).
         current_index (int): The index of the item to display.
-        tag (str): The current tag for loading messages.
+        user_input (str): The user's input concept or idea for loading messages.
         is_loading (bool): Whether the feed is currently loading.
     
     Returns:
         str: HTML string representing the feed.
     """
     loading_messages = [
-        f"Cooking up a {tag} masterpiece... 🍳",
-        f"Snapping a vibrant {tag} moment... 📸",
-        f"Creating a {tag} vibe that pops... ✨",
-        f"Getting that perfect {tag} shot... 🎥",
-        f"Bringing {tag} to life... 🌟"
+        f"Cooking up a {user_input} masterpiece... 🍳",
+        f"Snapping a vibrant {user_input} moment... 📸",
+        f"Creating a {user_input} vibe that pops... ✨",
+        f"Getting that perfect {user_input} shot... 🎥",
+        f"Bringing {user_input} to life... 🌟"
     ]
 
     if is_loading:
@@ -640,7 +648,7 @@ with gr.Blocks(
     title="Create Your Feed"
 ) as demo:
     # State variables
-    current_tag = gr.State(value="")
+    current_user_input = gr.State(value="")
     current_index = gr.State(value=0)
     feed_items = gr.State(value=[])
     is_loading = gr.State(value=False)
@@ -649,7 +657,7 @@ with gr.Blocks(
     # Input section
     with gr.Column(elem_classes="gr-form"):
         gr.Markdown("### Create Your Feed")
-        tag_input = gr.Textbox(
+        user_input = gr.Textbox(
             label="Enter Concept or Ideas",
             value="",
             placeholder="e.g., sushi adventure, neon tech",
@@ -667,17 +675,17 @@ with gr.Blocks(
 
     # Event handlers
     # Handle Enter keypress in the concept input
-    tag_input.submit(
+    user_input.submit(
         fn=start_feed,
-        inputs=[tag_input, generate_video_checkbox, current_index, feed_items],
-        outputs=[current_tag, current_index, feed_items, feed_html, share_html, is_loading]
+        inputs=[user_input, generate_video_checkbox, current_index, feed_items],
+        outputs=[current_user_input, current_index, feed_items, feed_html, share_html, is_loading]
     )
 
     # Handle magic button click to generate next item
     magic_button.click(
         fn=load_next,
-        inputs=[current_tag, generate_video_checkbox, current_index, feed_items],
-        outputs=[current_tag, current_index, feed_items, feed_html, share_html, is_loading]
+        inputs=[current_user_input, generate_video_checkbox, current_index, feed_items],
+        outputs=[current_user_input, current_index, feed_items, feed_html, share_html, is_loading]
     )
 
     # Hidden button for previous item navigation
@@ -686,8 +694,8 @@ with gr.Blocks(
     # Handle click to go to previous item
     previous_button.click(
         fn=load_previous,
-        inputs=[current_tag, generate_video_checkbox, current_index, feed_items],
-        outputs=[current_tag, current_index, feed_items, feed_html, share_html, is_loading]
+        inputs=[current_user_input, generate_video_checkbox, current_index, feed_items],
+        outputs=[current_user_input, current_index, feed_items, feed_html, share_html, is_loading]
     )
 
 # Launch the app