File size: 10,435 Bytes
e272f4f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7877316
 
e272f4f
 
7877316
 
e272f4f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7877316
 
e272f4f
 
 
7877316
 
e272f4f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7877316
e272f4f
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
import gradio as gr
import uuid
import logging
from typing import List, Tuple
from src.main import RAGSystem
import asyncio

# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Initialize the RAG system
rag = RAGSystem()

def create_session() -> str:
    """Create a new session ID"""
    return str(uuid.uuid4())

def index_website(url: str, session_id: str) -> Tuple[bool, str]:
    """Index a website for a given session"""
    try:
        result = rag.crawl_and_index(session_id, url)
        if result["status"] == "success":
            return True, f"Successfully indexed {len(result.get('urls_processed', []))} pages"
        return False, result.get("message", "Unknown error during indexing")
    except Exception as e:
        logger.error(f"Indexing error: {str(e)}")
        return False, f"Error during indexing: {str(e)}"

def chat_response(
    session_id: str,
    message: str,
    model_choice: str,
    ollama_url: str,
    gemini_api_key: str,
    chat_history: List[dict]
) -> Tuple[List[dict], str]:
    """Generate a chat response with proper error handling"""
    if not session_id:
        chat_history.append({"role": "user", "content": "πŸ§‘ " + message})
        chat_history.append({"role": "assistant", "content": "πŸ€– Please index a website first or enter a valid session ID"})
        return chat_history, ""
    
    try:
        response = asyncio.run(rag.chat(
            session_id=session_id,
            question=message,
            model=model_choice.lower(),
            ollama_url=ollama_url if model_choice == "mistral" else None,
            gemini_api_key=gemini_api_key if model_choice == "gemini" else None
        ))
        
        if response["status"] == "success":
            answer = response["response"]
            sources = "\n\nSources:\n" + "\n".join(
                f"- {src['source_url']}" for src in response.get("sources", [])
            ) if response.get("sources") else ""
            full_response = answer + sources
        else:
            full_response = f"Error: {response.get('message', 'Unknown error')}"
            
        chat_history.append({"role": "user", "content": "πŸ§‘ " + message})
        chat_history.append({"role": "assistant", "content": "πŸ€– " + full_response})
        return chat_history, ""
    except Exception as e:
        logger.error(f"Chat error: {str(e)}")
        chat_history.append({"role": "user", "content": "πŸ§‘ " + message})
        chat_history.append({"role": "assistant", "content": f"πŸ€– System error: {str(e)}"})
        return chat_history, ""

def toggle_model_inputs(model_choice: str) -> List[gr.update]:
    """Show/hide model-specific inputs"""
    if model_choice == "mistral":
        return [gr.update(visible=True), gr.update(visible=False)]
    return [gr.update(visible=False), gr.update(visible=True)]

def load_session(existing_session_id: str) -> Tuple[str, str]:
    """Load an existing session"""
    if existing_session_id:
        # Here you might want to add validation if the session exists
        return existing_session_id, f"Loaded existing session: {existing_session_id}"
    return "", "Please enter a valid session ID"

def get_session(self, session_id: str):
    # If session exists in memory, return it
    if session_id in self.sessions:
        return self.sessions[session_id]
    # If not, check if Qdrant collection exists and has documents
    collection_name = self.get_collection_name(session_id)
    try:
        results = self.qdrant_client.scroll(collection_name=collection_name, limit=1)
        if results and results[0]:
            # Rehydrate session in memory
            self.sessions[session_id] = {
                "documents": [],  # Optionally, you can fetch all docs if needed
                "history": []
            }
            return self.sessions[session_id]
    except Exception as e:
        logger.warning(f"Session {session_id} not found in Qdrant: {e}")
    # If not found, return None or raise
    raise ValueError("No documents indexed for this session")

# Custom CSS for better styling
custom_css = """
.gradio-container {
    max-width: 1200px !important;
    margin: 0 auto !important;
}
.dark .gradio-container {
    background: #1e1e2e !important;
}
#chatbot {
    min-height: 500px;
    border-radius: 12px !important;
}
.message.user {
    border-left: 4px solid #4f46e5 !important;
}
.message.assistant {
    border-left: 4px solid #10b981 !important;
}
.btn-primary {
    background: linear-gradient(to right, #4f46e5, #7c3aed) !important;
    border: none !important;
}
.btn-primary:hover {
    background: linear-gradient(to right, #4338ca, #6d28d9) !important;
}
.prose {
    max-width: 100% !important;
}
"""

with gr.Blocks(title="RAG Chat with Mistral/Gemini", css=custom_css, theme="soft") as demo:
    # Header section
    with gr.Row():
        gr.Markdown("""
        # 🌐 RAG Chat Assistant
        ### Chat with any website using Mistral or Gemini
        """)
    
    # Session state
    session_id = gr.State("")
    
    with gr.Tabs():
        with gr.TabItem("πŸ“š Index Website"):
            with gr.Row():
                with gr.Column():
                    gr.Markdown("### Step 1: Configure and Index")
                    with gr.Group():
                        url_input = gr.Textbox(
                            label="Website URL to index",
                            placeholder="https://example.com",
                            interactive=True,
                            lines=1
                        )
                        
                        with gr.Row():
                            model_choice = gr.Radio(
                                choices=["mistral", "gemini"],
                                label="Select Model",
                                value="mistral",
                                interactive=True
                            )
                            
                            index_btn = gr.Button(
                                "πŸš€ Index Website", 
                                variant="primary",
                                scale=0
                            )
                    
                    with gr.Accordion("πŸ” Model Settings", open=False):
                        ollama_url = gr.Textbox(
                            label="Ollama URL (required for Mistral)",
                            placeholder="http://localhost:11434",
                            visible=True
                        )
                        
                        gemini_api_key = gr.Textbox(
                            label="Gemini API Key (required for Gemini)",
                            placeholder="your-api-key-here",
                            visible=False,
                            type="password"
                        )
                    
                    status_output = gr.Textbox(
                        label="Status",
                        interactive=False,
                        elem_classes="prose"
                    )
                    
                    gr.Markdown("""
                    **Instructions:**
                    1. Enter a website URL
                    2. Select your preferred model
                    3. Configure model settings if needed
                    4. Click 'Index Website'
                    """)
        
        with gr.TabItem("πŸ’¬ Chat"):
            with gr.Row():
                with gr.Column(scale=2):
                    # New session ID input for resuming sessions
                    with gr.Accordion("πŸ” Resume Previous Session", open=False):
                        existing_session_input = gr.Textbox(
                            label="Enter existing Session ID",
                            placeholder="Paste your session ID here...",
                            interactive=True
                        )
                        load_session_btn = gr.Button(
                            "πŸ” Load Session",
                            variant="secondary"
                        )
                        session_status = gr.Textbox(
                            label="Session Status",
                            interactive=False
                        )
                    
                    chatbot = gr.Chatbot(
                    label="Chat History",
                    height=500,
                    avatar_images=(None, None),
                    show_copy_button=True,
                    type="messages"  # Use OpenAI-style messages
                )

                with gr.Row():
                    message_input = gr.Textbox(
                        label="Type your message",
                        placeholder="Ask about the website content...",
                            interactive=True,
                            container=False,
                            scale=7,
                            autofocus=True
                        )
                        
                    send_btn = gr.Button(
                            "Send", 
                            variant="primary",
                            scale=1,
                            min_width=100
                        )
    
    # Event handlers
    model_choice.change(
        fn=toggle_model_inputs,
        inputs=model_choice,
        outputs=[ollama_url, gemini_api_key]
    )
    
    index_btn.click(
        fn=create_session,
        outputs=session_id
    ).success(
        fn=index_website,
        inputs=[url_input, session_id],
        outputs=[status_output]
    )
    
    # New handler for loading existing sessions
    load_session_btn.click(
        fn=load_session,
        inputs=[existing_session_input],
        outputs=[session_id, session_status]
    )
    
    send_btn.click(
        fn=chat_response,
        inputs=[session_id, message_input, model_choice, ollama_url, gemini_api_key, chatbot],
        outputs=[chatbot, message_input]
    )
    
    # Allow submitting with Enter key
    message_input.submit(
        fn=chat_response,
        inputs=[session_id, message_input, model_choice, ollama_url, gemini_api_key, chatbot],
        outputs=[chatbot, message_input]
    )

if __name__ == "__main__":
    demo.launch(
        server_name="0.0.0.0",
        server_port=7860,
        #favicon_path="assets/favicon.ico"  # Local path, not URL
    )