Files
matagent/_backend/api.py
2025-02-24 15:28:57 +08:00

420 lines
15 KiB
Python

import json
import logging
import os
from typing import Any, Awaitable, Callable, Optional, Sequence
import uuid
import aiofiles
import yaml
import cv2
import base64
import asyncio
import numpy as np
import time
import subprocess
import ffmpeg
import io
from PIL import Image
from collections import deque
from fastapi import FastAPI, HTTPException, WebSocket, WebSocketDisconnect
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import FileResponse
from fastapi.staticfiles import StaticFiles
from fastapi.responses import HTMLResponse
from aiortc import RTCPeerConnection, RTCSessionDescription, VideoStreamTrack
from aiortc.contrib.media import MediaRelay
from aiortc.contrib.media import MediaPlayer
from autogen_agentchat.agents import AssistantAgent, UserProxyAgent
from autogen_agentchat.base import TaskResult
from autogen_agentchat.messages import TextMessage, UserInputRequestedEvent
from autogen_agentchat.teams import RoundRobinGroupChat
from autogen_core import CancellationToken
from autogen_core.models import ChatCompletionClient
from autogen_ext.models.openai import OpenAIChatCompletionClient
from autogen_agentchat.conditions import MaxMessageTermination, TextMentionTermination
from autogen_agentchat.teams import SelectorGroupChat, RoundRobinGroupChat
from autogen_agentchat.messages import AgentEvent, ChatMessage, TextMessage, ToolCallExecutionEvent
from constant import MODEL, OPENAI_API_KEY, OPENAI_BASE_URL
from scientist_team import create_scientist_team
from engineer_team import create_engineer_team
from robot_platform import create_robot_team
from analyst_team import create_analyst_team
from utils import load_agent_configs
logger = logging.getLogger(__name__)
relay = MediaRelay()
model_client = OpenAIChatCompletionClient(
model=MODEL,
base_url=OPENAI_BASE_URL,
api_key=OPENAI_API_KEY,
model_info={
"vision": True,
"function_calling": True,
"json_output": True,
"family": "unknown",
},
)
async def get_team(
user_input_func: Callable[[str, Optional[CancellationToken]], Awaitable[str]],
session_dir: str
) -> RoundRobinGroupChat | SelectorGroupChat:
# Create the team.
scientist_team = create_scientist_team(model_client=model_client)
engineer_team = create_engineer_team()
robot_platform = create_robot_team()
analyst_team = create_analyst_team()
user = UserProxyAgent(
name="user",
input_func=user_input_func, # Use the user input function.
)
cur_path = os.path.dirname(os.path.abspath(__file__))
planning_agent_system_message = load_agent_configs(os.path.join(cur_path, "agent_config/planning.yaml"))
planning_agent = AssistantAgent(
"PlanningAgent",
description="An agent for planning tasks, this agent should be the first to engage when given a new task.",
model_client=model_client,
system_message=planning_agent_system_message["PlanningAgent"],
reflect_on_tool_use=False
)
# The termination condition is a combination of text mention termination and max message termination.
text_mention_termination = TextMentionTermination("TERMINATE")
max_messages_termination = MaxMessageTermination(max_messages=200)
termination = text_mention_termination | max_messages_termination
# The selector function is a function that takes the current message thread of the group chat
# and returns the next speaker's name. If None is returned, the LLM-based selection method will be used.
def selector_func(messages: Sequence[AgentEvent | ChatMessage]) -> str | None:
if messages[-1].source != planning_agent.name:
return planning_agent.name # Always return to the planning agent after the other agents have spoken.
elif "HUMAN" in messages[-1].content:
return user.name
return None
team = SelectorGroupChat(
[planning_agent, user, scientist_team, engineer_team, robot_platform, analyst_team],
model_client=model_client, # Use a smaller model for the selector.
termination_condition=termination,
selector_func=selector_func,
)
# Load state from file.
state_path = os.path.join(session_dir, "team_state.json")
if not os.path.exists(state_path):
return team
async with aiofiles.open(state_path, "r") as file:
state = json.loads(await file.read())
await team.load_state(state)
return team
app = FastAPI()
current_task = None # 用于跟踪当前任务
# Add CORS middleware
app.add_middleware(
CORSMiddleware,
allow_origins=["*"], # Allows all origins
allow_credentials=True,
allow_methods=["*"], # Allows all methods
allow_headers=["*"], # Allows all headers
)
# model_config_path = "model_config.yaml"
# state_path = "team_state.json"
# history_path = "team_history.json"
# Serve static files
app.mount("/static", StaticFiles(directory="."), name="static")
async def get_session_history(session_dir: str) -> list[dict[str, Any]]:
"""Get chat history from file using UUID."""
session_history_path = os.path.join(session_dir, "team_history.json")
if not os.path.exists(session_history_path):
return []
async with aiofiles.open(session_history_path, "r") as file:
content = await file.read()
if content:
return json.loads(content)
else:
return []
@app.websocket("/history/{session_uuid}")
async def history(websocket: WebSocket) -> list[dict[str, Any]]:
await websocket.accept()
data = await websocket.receive_json()
session_uuid = data["uuid"]
try:
session_history = await get_session_history(session_uuid)
await websocket.send_json(session_history)
except Exception as e:
raise HTTPException(status_code=500, detail=str(e)) from e
@app.websocket("/sessions")
async def sessions(websocket: WebSocket) -> list[dict[str, str]]:
"""Get all history UUIDs and their main content."""
await websocket.accept()
cur_path = os.path.dirname(os.path.abspath(__file__))
history_dir = os.path.join(cur_path, "history")
session_data = []
for dir_name in os.listdir(history_dir):
session_dir = os.path.join(history_dir, dir_name)
if os.path.isdir(session_dir): # Check if it's a directory
try:
history = await get_session_history(session_dir)
main_content = history[0]["content"] if history and "content" in history[0] else ""
session_data.append({"uuid": dir_name, "content": main_content})
except Exception as e:
print(f"Error reading history for {dir_name}: {e}") # Log the error but continue
await websocket.send_json(session_data)
@app.websocket("/ws/chat")
async def chat(websocket: WebSocket):
await websocket.accept()
# User input function used by the team.
async def _user_input(prompt: str, cancellation_token: CancellationToken | None) -> str:
data = await websocket.receive_json()
return data['content']
try:
while True:
# Get user message.
data = await websocket.receive_json()
if 'session_uuid' not in data:
# New session
request = TextMessage.model_validate(data)
# request = data['content']
session_uuid = str(uuid.uuid4()) # Initialize a unique UUID for each session
cur_path = os.path.dirname(os.path.abspath(__file__))
session_dir = os.path.join(cur_path, "history", session_uuid) # Directory for session states
os.makedirs(session_dir, exist_ok=True) # ensure the directory is created.
history = []
else:
session_uuid = data['session_uuid']
cur_path = os.path.dirname(os.path.abspath(__file__))
session_dir = os.path.join(cur_path, "history", session_uuid) # Directory for session states
history = await get_session_history(session_dir)
new_data = {k: v for k, v in data.items() if k != "session_uuid"}
request = TextMessage.model_validate(new_data)
request = history + request
try:
# Get the team and respond to the message.
team = await get_team(_user_input, session_dir)
stream = team.run_stream(task=request)
async for message in stream:
if isinstance(message, TaskResult):
continue
print(f"----------------{message.source}----------------\n {message.content}")
await websocket.send_json(message.model_dump())
if not isinstance(message, UserInputRequestedEvent):
history.append(message.model_dump())
# Save chat history to file.
session_history_path = os.path.join(session_dir, "team_history.json")
async with aiofiles.open(session_history_path, "w") as file:
await file.write(json.dumps(history))
# # Save team state to file.
# session_state_path = os.path.join(session_dir, "team_state.json")
# async with aiofiles.open(session_state_path, "w") as file:
# state = await team.save_state()
# await file.write(json.dumps(state))
except Exception as e:
# Send error message to client
error_message = {
"type": "error",
"content": f"Error: {str(e)}",
"source": "system"
}
await websocket.send_json(error_message)
# Re-enable input after error
await websocket.send_json({
"type": "UserInputRequestedEvent",
"content": "An error occurred. Please try again.",
"source": "system"
})
except WebSocketDisconnect:
logger.info("Client disconnected")
except Exception as e:
logger.error(f"Unexpected error: {str(e)}")
try:
await websocket.send_json({
"type": "error",
"content": f"Unexpected error: {str(e)}",
"source": "system"
})
except:
pass
RTSP_STREAMS = {
"camera1": "rtsp://admin:@192.168.1.13:554/live",
"camera2": "rtsp://admin:@192.168.1.10:554/live",
}
@app.websocket("/video_stream/{camera_id}")
async def websocket_endpoint(websocket: WebSocket, camera_id: str):
await websocket.accept()
rtsp_url = RTSP_STREAMS.get(camera_id)
if not rtsp_url:
await websocket.close(code=4001)
return
fps = 15
frame_interval = 1 / fps
frame_buffer = deque() # 每个连接独立的队列
# 启动FFmpeg进程
process = (
ffmpeg
.input(rtsp_url, rtsp_transport='tcp')
.output('pipe:', format='image2pipe', vcodec='mjpeg', r=fps)
.run_async(pipe_stdout=True, pipe_stderr=True, quiet=True)
)
async def capture_frames():
"""读取并分割完整的JPEG帧"""
buffer = b''
loop = asyncio.get_event_loop()
while True:
# 异步读取防止阻塞
data = await loop.run_in_executor(None, process.stdout.read, 4096)
if not data:
break
buffer += data
# 提取完整帧
while True:
start = buffer.find(b'\xff\xd8') # JPEG开始标记
if start == -1:
break
end = buffer.find(b'\xff\xd9', start + 2) # JPEG结束标记
if end == -1:
break
# 提取帧数据并编码
frame_data = buffer[start:end+2]
frame = base64.b64encode(frame_data).decode('utf-8')
frame_buffer.append(frame)
buffer = buffer[end+2:] # 移除已处理数据
# 清理FFmpeg进程
process.kill()
async def send_frames():
"""发送帧到客户端"""
while True:
if frame_buffer:
frame = frame_buffer.popleft()
await websocket.send_text(frame)
await asyncio.sleep(frame_interval)
# 启动任务并处理退出
capture_task = asyncio.create_task(capture_frames())
send_task = asyncio.create_task(send_frames())
try:
await asyncio.gather(capture_task, send_task)
except asyncio.CancelledError:
capture_task.cancel()
send_task.cancel()
await asyncio.gather(capture_task, send_task, return_exceptions=True)
finally:
process.kill()
await websocket.close()
@app.get("/")
async def debug_page():
return HTMLResponse("""
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>RTSP Stream</title>
<style>
#video {
width: 100%;
height: auto;
}
</style>
</head>
<body>
<!-- 视频流 1 的画布 -->
<canvas id="canvas1"></canvas>
<script>
const canvas1 = document.getElementById('canvas1');
const ctx1 = canvas1.getContext('2d');
const img1 = new Image();
img1.onload = () => {
ctx1.drawImage(img1, 0, 0, canvas1.width, canvas1.height);
};
const ws1 = new WebSocket('ws://100.85.52.31:8000/video_stream/camera1');
ws1.onmessage = (event) => {
img1.src = 'data:image/jpeg;base64,' + event.data;
};
// 动态调整 Canvas 尺寸
function resizeCanvas1() {
canvas1.width = 720; // 设置宽度为屏幕的一半
canvas1.height = 480; // 设置高度为屏幕的一半
}
resizeCanvas1();
window.addEventListener('resize', resizeCanvas1);
</script>
<!-- 视频流 2 的画布 -->
<canvas id="canvas2"></canvas>
<script>
const canvas2 = document.getElementById('canvas2');
const ctx2 = canvas2.getContext('2d');
const img2 = new Image();
img2.onload = () => {
ctx2.drawImage(img2, 0, 0, canvas2.width, canvas2.height);
};
const ws2 = new WebSocket('ws://100.85.52.31:8000/video_stream/camera2');
ws2.onmessage = (event) => {
img2.src = 'data:image/jpeg;base64,' + event.data;
};
// 动态调整 Canvas 尺寸
function resizeCanvas2() {
canvas2.width = 720; // 设置宽度为屏幕的一半
canvas2.height = 480; // 设置高度为屏幕的一半
}
resizeCanvas2();
window.addEventListener('resize', resizeCanvas2);
</script>
</body>
</html>
""")
# Example usage
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=8000)