Update space

app.py

@@ -1,160 +1,106 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-import pandas as pd
 import requests
 from bs4 import BeautifulSoup
+import pandas as pd
 
-# Initialize HF client
-client = InferenceClient("meta-llama/Llama-2-7b-chat-hf")
+"""
+For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
+"""
+client = InferenceClient("meta-llama/Llama-3.2-3B-Instruct")
+
+
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
+
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+
+    messages.append({"role": "user", "content": message})
+
+    response = ""
+
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
+
+        response += token
+        yield response
 
-def respond(message, history, max_tokens=512, temperature=0.7, top_p=0.95):
-    try:
-        # Format messages including history
-        messages = []
-        for user_msg, assistant_msg in history:
-            messages.append({"role": "user", "content": user_msg})
-            messages.append({"role": "assistant", "content": assistant_msg})
-        messages.append({"role": "user", "content": message})
-        
-        # Generate response
-        response = ""
-        for chunk in client.chat_completion(
-            messages,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-            stream=True,
-        ):
-            if hasattr(chunk.choices[0].delta, 'content'):
-                token = chunk.choices[0].delta.content
-                if token:
-                    response += token
-        return response
-        
-    except Exception as e:
-        return f"Error: {str(e)}"
 
-def extract_schedule(url):
+def extract_table(url):
+    global data
     try:
-        # Fetch and parse webpage
         response = requests.get(url)
         response.raise_for_status()
         soup = BeautifulSoup(response.text, 'html.parser')
-        
-        # Find table and extract data
         table = soup.find('table')
         if not table:
             return "<p>No table found on page</p>"
-            
-        schedule_data = []
+        
+        data = []
         rows = table.find_all('tr')
-        for row in rows[1:]:
+        for i, row in enumerate(rows[1:]):
             cells = row.find_all('td')
-            if len(cells) >= 4:
-                date = cells[0].text.strip()
-                topic = cells[1].text.strip()
-                
-                if date and topic and not topic.startswith('See Canvas'):
-                    schedule_data.append({
-                        'Date': date[:10],
-                        'Topic': topic,
-                        'Action': f"""
-                        <button 
-                            onclick="triggerChatbotPreparation('{topic.replace("'", "")}')" 
-                            class="prepare-btn">
-                            Prepare
-                        </button>
-                        """
-                    })
-
-        df = pd.DataFrame(schedule_data)
-        
-        # Convert to HTML with styling and JavaScript
-        html = f"""
-        <style>
-        table {{ 
-            border-collapse: collapse; 
-            width: 100%;
-            font-size: 12px;
-        }}
-        th, td {{ 
-            border: 1px solid black;
-            padding: 6px;
-            text-align: left;
-            font-family: Arial, sans-serif;
-        }}
-        .prepare-btn {{
-            padding: 4px 8px;
-            font-size: 11px;
-            cursor: pointer;
-        }}
-        </style>
-        <script>
-        function triggerChatbotPreparation(topic) {{
-            // Find all Gradio textareas
-            const textareas = document.querySelectorAll('.gradio-container textarea');
-            
-            // Find the first textarea (assuming it's the input)
-            const textbox = textareas[0];
-            
-            if (textbox) {{
-                // Set the value
-                const preparationMessage = `prepare 5 minutes reading important parts related to ${topic}`;
-                textbox.value = preparationMessage;
-                
-                // Trigger input and change events
-                const inputEvent = new Event('input', {{ bubbles: true }});
-                const changeEvent = new Event('change', {{ bubbles: true }});
-                textbox.dispatchEvent(inputEvent);
-                textbox.dispatchEvent(changeEvent);
-                
-                // Find and click the send button
-                const sendButtons = document.querySelectorAll('.gradio-container button');
-                for (let button of sendButtons) {{
-                    if (button.getAttribute('aria-label') === 'Send') {{
-                        button.click();
-                        break;
-                    }}
-                }}
-            }}
-        }}
-        </script>
-        {df.to_html(index=False, escape=False)}
-        """
-        return html
+            if len(cells) >= 2:
+                data.append({
+                    'Date': cells[0].text.strip()[:10],
+                    'Topic': cells[1].text.strip(),
+                })
         
+        df = pd.DataFrame(data)
+        return df.to_html(escape=False, index=False)
     except Exception as e:
         return f"<p>Error: {str(e)}</p>"
 
-def display_schedule(url):
-    try:
-        html_table = extract_schedule(url)
-        return html_table  # Already HTML string
-    except Exception as e:
-        return str(e)
 
+def display_table(url):
+    return extract_table(url)
+
+
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
 with gr.Blocks() as demo:
     with gr.Row():
-        # Left Column - Schedule
         with gr.Column(scale=1):
             url_input = gr.Textbox(
                 value="https://id2223kth.github.io/schedule/",
-                label="Schedule URL"
+                label="Table URL"
             )
-            schedule_output = gr.HTML(label="Extracted Schedule")
-            extract_btn = gr.Button("Extract Schedule")
+            table_output = gr.HTML(label="Extracted Table")
+            extract_btn = gr.Button("Extract Table")
             
             extract_btn.click(
-                fn=display_schedule,
+                fn=display_table,
                 inputs=[url_input],
-                outputs=[schedule_output]
+                outputs=[table_output]
             )
         
-        # Right Column - Chatbot
         with gr.Column(scale=2):
+            
             chatbot = gr.ChatInterface(
-                respond
+                respond,
+                additional_inputs=[
+                    gr.Textbox(value="Student class preparation companion.", label="System message"),
+                ],
             )
+            
 
 if __name__ == "__main__":
     demo.launch()