app as fast apis

.gitignore

@@ -1,3 +1,9 @@
+#custom
+.env
+*.pdf
+app copy.py
+myenv/
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -73,6 +79,4 @@ htmlcov/
 *.sock
 
 # Ignore any large files not tracked by Git LFS
-!*.lfs
-
-.env
+!*.lfs

app.py

@@ -1,40 +1,48 @@
-from fastapi import FastAPI, Request, Form
-from fastapi.responses import RedirectResponse
-from fastapi.templating import Jinja2Templates
-from fastapi.staticfiles import StaticFiles
+from fastapi import FastAPI, Request
 # from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+from typing import List, Optional
 from helpmate_ai import initialize_conversation, retreive_results, rerank_with_cross_encoder, generate_response
-import re
 import google.generativeai as genai
 import os
 from dotenv import load_dotenv
+import re
 
-# Configure Gemini API
-# gemini_api_key = open("gemini_api_key.txt", "r").read().strip()
+# Load environment variables
 load_dotenv()
-
 gemini_api_key = os.getenv("GEMINI_API_KEY")
 genai.configure(api_key=gemini_api_key)
 
 # Initialize FastAPI app
 app = FastAPI()
 
-# Set up templates
-templates = Jinja2Templates(directory="templates")
-
-# Serve static files (if needed)
-app.mount("/static", StaticFiles(directory="static"), name="static")
-
-# Enable CORS middleware if needed
+# # Enable CORS
 # app.add_middleware(
 #     CORSMiddleware,
-#     allow_origins=["*"],  # Adjust origins as per requirements
+#     allow_origins=["*"],
 #     allow_credentials=True,
 #     allow_methods=["*"],
 #     allow_headers=["*"],
 # )
 
-def format_rag_response(response_text):
+# Pydantic models for request/response validation
+class Message(BaseModel):
+    role: str
+    content: str
+
+class ChatRequest(BaseModel):
+    message: str
+
+class ChatResponse(BaseModel):
+    response: str
+    conversation: List[Message]
+
+# Initialize conversation and model
+conversation_bot = []
+conversation = initialize_conversation()
+model = genai.GenerativeModel("gemini-1.5-flash", system_instruction=conversation)
+
+def format_rag_response(response_text: str) -> str:
     formatted_text = response_text.replace("\n", "<br>")
     formatted_text = re.sub(r'(\*\*.*?\*\*)', r'<strong>\1</strong>', formatted_text).replace("**", "")
     formatted_text = re.sub(r'(\d+\.\s)', r'<br><strong>\1</strong>', formatted_text)
@@ -44,53 +52,56 @@ def format_rag_response(response_text):
     formatted_text = re.sub(r'\n\|\s*', r'<tr><td>', formatted_text)
     return formatted_text
 
-conversation_bot = []
-conversation = initialize_conversation()
-
-# Initialize Gemini model
-model = genai.GenerativeModel("gemini-1.5-flash", system_instruction=conversation)
-
-def get_gemini_completions(conversation):
+def get_gemini_completions(conversation: str) -> str:
     response = model.generate_content(conversation)
     return response.text
 
-introduction = get_gemini_completions(conversation)
-conversation_bot.append({'bot': introduction})
-top_3_laptops = None
-
-@app.get("/")
-async def default_func(request: Request):
+# Initialize conversation endpoint
+@app.get("/init", response_model=ChatResponse)
+async def initialize_chat():
     global conversation_bot
-    return templates.TemplateResponse("index_invite.html", {"request": request, "name_xyz": conversation_bot})
-
-@app.post("/end_conv")
-async def end_conv():
-    global conversation_bot, conversation, top_3_laptops
-    conversation_bot = []
-    conversation = initialize_conversation()
     introduction = get_gemini_completions(conversation)
-    conversation_bot.append({'bot': introduction})
-    top_3_laptops = None
-    return RedirectResponse(url="/", status_code=303)
-
-@app.post("/invite")
-async def invite(user_input_message: str = Form(...)):
-    global conversation_bot, conversation, top_3_laptops
-    user_input = user_input_message
-    conversation_bot.append({'user': user_input})
-
-    results_df = retreive_results(user_input)
-    top_docs = rerank_with_cross_encoder(user_input, results_df)
-
+    conversation_bot = [Message(role="bot", content=introduction)]
+    return ChatResponse(
+        response=introduction,
+        conversation=conversation_bot
+    )
+
+# Chat endpoint
+@app.post("/chat", response_model=ChatResponse)
+async def chat(request: ChatRequest):
+    global conversation_bot
+    
+    # Add user message to conversation
+    user_message = Message(role="user", content=request.message)
+    conversation_bot.append(user_message)
+    
     # Generate response
-    messages = generate_response(user_input, top_docs)
+    results_df = retreive_results(request.message)
+    top_docs = rerank_with_cross_encoder(request.message, results_df)
+    messages = generate_response(request.message, top_docs)
     response_assistant = get_gemini_completions(messages)
+    formatted_response = format_rag_response(response_assistant)
+    
+    # Add bot response to conversation
+    bot_message = Message(role="bot", content=formatted_response)
+    conversation_bot.append(bot_message)
+    
+    return ChatResponse(
+        response=formatted_response,
+        conversation=conversation_bot
+    )
+
+# Reset conversation endpoint
+@app.post("/reset")
+async def reset_conversation():
+    global conversation_bot, conversation
+    conversation_bot = []
+    conversation = initialize_conversation()
+    introduction = get_gemini_completions(conversation)
+    conversation_bot.append(Message(role="bot", content=introduction))
+    return {"status": "success", "message": "Conversation reset"}
 
-    conversation_bot.append({'bot': format_rag_response(response_assistant)})
-    return RedirectResponse(url="/", status_code=303)
-
-# Run the application
-# if __name__ == '__main__':
-#     main()
-    # import uvicorn
-    # uvicorn.run(app, host="0.0.0.0", port=8000, debug=True)
+# if __name__ == "__main__":
+#     import uvicorn
+#     uvicorn.run(app, host="0.0.0.0", port=8000)

requirements.txt

@@ -6,5 +6,5 @@ fastapi
 uvicorn[standard] 
 jinja2 
 python-multipart
-sentence_transformers
-python-dotenv
+sentence-transformers
+python-dotenv