File size: 5,498 Bytes
17c13db 8b3b01c 17c13db 0553d6a 17c13db 8b3b01c 17c13db 8b3b01c 275e33e 8b3b01c 17c13db 8b3b01c 17c13db 8b3b01c 275e33e 17c13db 2819245 17c13db 2819245 17c13db 2819245 17c13db 2819245 17c13db 2819245 17c13db 094984f 17c13db |
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 |
from fastapi import FastAPI, HTTPException, File, UploadFile
from contextlib import asynccontextmanager
from pydantic import BaseModel
import os
from lightrag import LightRAG, QueryParam
from lightrag.llm.ollama import ollama_embed, ollama_model_complete
from lightrag.utils import EmbeddingFunc
from typing import Optional
import asyncio
import nest_asyncio
import aiofiles
from lightrag.kg.shared_storage import initialize_pipeline_status
# Apply nest_asyncio to solve event loop issues
nest_asyncio.apply()
DEFAULT_RAG_DIR = "index_default"
DEFAULT_INPUT_FILE = "book.txt"
INPUT_FILE = os.environ.get("INPUT_FILE", f"{DEFAULT_INPUT_FILE}")
print(f"INPUT_FILE: {INPUT_FILE}")
# Configure working directory
WORKING_DIR = os.environ.get("RAG_DIR", f"{DEFAULT_RAG_DIR}")
print(f"WORKING_DIR: {WORKING_DIR}")
if not os.path.exists(WORKING_DIR):
os.mkdir(WORKING_DIR)
async def init():
rag = LightRAG(
working_dir=WORKING_DIR,
llm_model_func=ollama_model_complete,
llm_model_name="gemma2:9b",
llm_model_max_async=4,
llm_model_max_token_size=8192,
llm_model_kwargs={
"host": "http://localhost:11434",
"options": {"num_ctx": 8192},
},
embedding_func=EmbeddingFunc(
embedding_dim=768,
max_token_size=8192,
func=lambda texts: ollama_embed(
texts, embed_model="nomic-embed-text", host="http://localhost:11434"
),
),
)
# Add initialization code
await rag.initialize_storages()
await initialize_pipeline_status()
return rag
@asynccontextmanager
async def lifespan(app: FastAPI):
global rag
rag = await init()
print("done!")
yield
app = FastAPI(
title="LightRAG API", description="API for RAG operations", lifespan=lifespan
)
# Data models
class QueryRequest(BaseModel):
query: str
mode: str = "hybrid"
only_need_context: bool = False
class InsertRequest(BaseModel):
text: str
class Response(BaseModel):
status: str
data: Optional[str] = None
message: Optional[str] = None
# API routes
@app.post("/query", response_model=Response)
async def query_endpoint(request: QueryRequest):
try:
loop = asyncio.get_event_loop()
result = await loop.run_in_executor(
None,
lambda: rag.query(
request.query,
param=QueryParam(
mode=request.mode, only_need_context=request.only_need_context
),
),
)
return Response(status="success", data=result)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
# insert by text
@app.post("/insert", response_model=Response)
async def insert_endpoint(request: InsertRequest):
try:
loop = asyncio.get_event_loop()
await loop.run_in_executor(None, lambda: rag.insert(request.text))
return Response(status="success", message="Text inserted successfully")
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
# insert by file in payload
@app.post("/insert_file", response_model=Response)
async def insert_file(file: UploadFile = File(...)):
try:
file_content = await file.read()
# Read file content
try:
content = file_content.decode("utf-8")
except UnicodeDecodeError:
# If UTF-8 decoding fails, try other encodings
content = file_content.decode("gbk")
# Insert file content
loop = asyncio.get_event_loop()
await loop.run_in_executor(None, lambda: rag.insert(content))
return Response(
status="success",
message=f"File content from {file.filename} inserted successfully",
)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
# insert by local default file
@app.post("/insert_default_file", response_model=Response)
@app.get("/insert_default_file", response_model=Response)
async def insert_default_file():
try:
# Read file content from book.txt
async with aiofiles.open(INPUT_FILE, "r", encoding="utf-8") as file:
content = await file.read()
print(f"read input file {INPUT_FILE} successfully")
# Insert file content
loop = asyncio.get_event_loop()
await loop.run_in_executor(None, lambda: rag.insert(content))
return Response(
status="success",
message=f"File content from {INPUT_FILE} inserted successfully",
)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e))
@app.get("/health")
async def health_check():
return {"status": "healthy"}
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=8020)
# Usage example
# To run the server, use the following command in your terminal:
# python lightrag_api_openai_compatible_demo.py
# Example requests:
# 1. Query:
# curl -X POST "http://127.0.0.1:8020/query" -H "Content-Type: application/json" -d '{"query": "your query here", "mode": "hybrid"}'
# 2. Insert text:
# curl -X POST "http://127.0.0.1:8020/insert" -H "Content-Type: application/json" -d '{"text": "your text here"}'
# 3. Insert file:
# curl -X POST "http://127.0.0.1:8020/insert_file" -H "Content-Type: multipart/form-data" -F "file=@path/to/your/file.txt"
# 4. Health check:
# curl -X GET "http://127.0.0.1:8020/health"
|