Spaces:
Paused
Paused
File size: 4,917 Bytes
ddaa426 |
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 |
import os
import uvicorn
import tempfile
from openai import AsyncOpenAI
from fastapi import FastAPI, Body, UploadFile, File, Depends, HTTPException
from fastapi.staticfiles import StaticFiles
from fastapi.responses import StreamingResponse, JSONResponse
from aimakerspace.openai_utils.prompts import (
UserRolePrompt,
SystemRolePrompt,
)
from aimakerspace.text_utils import CharacterTextSplitter, TextFileLoader, PDFLoader
from qdrant_client import QdrantClient
from fastapi.security import APIKeyHeader
import uuid
from typing import Dict, Optional
system_template = """\
Use the following context to answer a users question.
If you cannot find the answer in the context, say you don't know the answer.
"""
system_role_prompt = SystemRolePrompt(system_template)
user_prompt_template = """\
Context:
{context}
Question:
{question}
"""
user_role_prompt = UserRolePrompt(user_prompt_template)
app = FastAPI()
openai = AsyncOpenAI()
vector_db = QdrantClient(":memory:")
text_splitter = CharacterTextSplitter()
sessions: Dict[str, dict] = {}
api_key_header = APIKeyHeader(name="X-Session-ID", auto_error=False)
async def get_session(session_id: Optional[str] = Depends(api_key_header)):
if not session_id:
# Create new session
session_id = str(uuid.uuid4())
sessions[session_id] = {
"vector_db": None,
"vector_db_retriever": None,
}
elif session_id not in sessions:
raise HTTPException(status_code=404, detail="Session not found")
return session_id, sessions[session_id]
def process_file(file: UploadFile):
print(f"Processing file: {file.filename}")
# Create a temporary file with the correct extension
suffix = f".{file.filename.split('.')[-1]}"
with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as temp_file:
# Write the uploaded file content to the temporary file
content = file.file.read()
temp_file.write(content)
temp_file.flush()
print(f"Created temporary file at: {temp_file.name}")
# Create appropriate loader
if file.filename.lower().endswith('.pdf'):
loader = PDFLoader(temp_file.name)
else:
loader = TextFileLoader(temp_file.name)
try:
# Load and process the documents
documents = loader.load_documents()
texts = text_splitter.split_texts(documents)
return texts
finally:
# Clean up the temporary file
try:
os.unlink(temp_file.name)
except Exception as e:
print(f"Error cleaning up temporary file: {e}")
async def get_response(msg: str, session_id: str, vector_db: QdrantClient):
context_list = vector_db.query(
collection_name=session_id,
query_text=msg,
limit=4,
)
context_prompt = ""
for context in context_list:
context_prompt += context.document + "\n"
formatted_system_prompt = system_role_prompt.create_message()
formatted_user_prompt = user_role_prompt.create_message(question=msg, context=context_prompt)
openai_stream = await openai.chat.completions.create(
model="gpt-3.5-turbo",
messages=[
formatted_system_prompt,
formatted_user_prompt,
],
temperature=0.0,
stream=True,
)
async def generate_response():
async for chunk in openai_stream:
if chunk.choices[0].delta.content is not None:
yield chunk.choices[0].delta.content
yield ""
return StreamingResponse(generate_response(), media_type="text/event-stream")
@app.post("/api/chat")
async def get_bot_response(
msg: str = Body(...),
session_data: tuple = Depends(get_session)
):
session_id, _ = session_data
print(f"Session ID: {session_id}")
response = await get_response(msg, session_id, vector_db)
return response
@app.post("/api/file")
async def get_file_response(
file: UploadFile = File(..., description="A text file to process"),
session_data: tuple = Depends(get_session)
):
session_id, _ = session_data
print(f"Session ID: {session_id}")
if not file.filename:
return {"error": "No file uploaded"}
try:
chunks = process_file(file)
vector_db.add(
collection_name=session_id,
documents=chunks,
)
return {
"message": "File processed successfully",
"session_id": session_id
}
except Exception as e:
return JSONResponse(
status_code=422,
content={"detail": str(e)}
)
app.mount("/", StaticFiles(directory="dist", html=True), name="static")
app.get("/")(StaticFiles(directory="dist", html=True))
if __name__ == "__main__":
uvicorn.run("server:app")
|