Spaces:
Sleeping
Sleeping
File size: 3,241 Bytes
6e94f25 555a055 6e94f25 555a055 6e94f25 555a055 6e94f25 b64d6a4 6e94f25 8eaaf5c 6e94f25 |
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 |
import os
import logging
import uvicorn
import json
import hashlib
# Import our existing pipeline components
from app.utils.util import verify_api_key
from app.core.models import llm_response_generator
from app.schema.schema import QuestionRequest, AnswerResponse
from contextlib import asynccontextmanager
from fastapi import FastAPI, HTTPException, Depends, status
from fastapi.responses import FileResponse, RedirectResponse
logging.basicConfig(format='%(asctime)s - %(levelname)s - Line: %(lineno)d - %(message)s',
datefmt='%Y-%m-%d %H:%M:%S',
level=logging.INFO)
os.environ["TOKENIZERS_PARALLELISM"] = "false"
# Load config.json at startup
@asynccontextmanager
async def lifespan(app: FastAPI):
import json
config_path = os.path.join(os.path.dirname(__file__), "app", "config", "config.json")
with open(config_path, "r") as f:
app.state.config = json.load(f)
logging.info("Config loaded successfully.")
yield
logging.info("Shutting down FastAPI app...")
# FastAPI app
app = FastAPI(title="HackRx PDF RAG API", version="1.0.0", lifespan=lifespan)
@app.get("/", include_in_schema=False)
async def docs_redirect():
return RedirectResponse("/docs")
@app.post("/api/v1/hackrx/run", response_model=AnswerResponse)
async def process_questions(request: QuestionRequest, api_key: str = Depends(verify_api_key)):
try:
url = str(request.documents)
config = app.state.config
questions = request.questions
logging.info(f"Received {(questions)} questions for processing. Documents URL: {url}")
# Create cache directory if not exists
cache_dir = "/tmp/cache"
os.makedirs(cache_dir, exist_ok=True)
# Create a cache key from url and questions
cache_key = hashlib.sha256((url + json.dumps(questions, sort_keys=True)).encode()).hexdigest()
cache_file = os.path.join(cache_dir, f"{cache_key}.json")
# Try to load from cache
if os.path.exists(cache_file):
logging.info(f"Cache hit for key: {cache_key}")
with open(cache_file, "r") as f:
cached_response = json.load(f)
return cached_response
# Otherwise, call LLM and cache response
response, status_code = await llm_response_generator(config=config, url=url, questions=questions)
if status_code != status.HTTP_200_OK:
raise HTTPException(status_code=status_code, detail="Error processing questions")
else:
with open(cache_file, "w") as f:
json.dump(response, f)
logging.info(f"Cache saved for key: {cache_key}")
return response
except Exception as e:
logging.error(f"Error processing questions: {e}")
raise HTTPException(status_code=500, detail="Internal Server Error")
@app.get("/health")
async def health_check():
"""
Health check endpoint to verify API is running
"""
return {"status": "Healthy", "message": "API is running smoothly.", "code": status.HTTP_200_OK}
if __name__ == "__main__":
uvicorn.run(
"main:app",
host="0.0.0.0",
port=8000,
reload=True,
workers=1
) |