更新canvas 3d接口 ,使用异步mq队列处理

Dockerfile

@@ -24,8 +24,6 @@ RUN playwright install
 #CMD ["tail","-f","/dev/null"]
 # Run the application.
 CMD ["gunicorn", "main:app_server", \
-     "-w", "4", \
+     "-c", "/app/gunicorn.conf.py", \
-     "-k", "uvicorn.workers.UvicornWorker", \
-     "--bind", "0.0.0.0:80", \
      "--access-logfile", "-", \
      "--error-logfile", "-"]

docker-compose.yml

@@ -1,4 +1,5 @@
 name: fida-python-prod
+
 services:
   server:
     container_name: "FiDA_${SERVE_ENV}_Server"
@@ -12,3 +13,55 @@ services:
       - /etc/localtime:/etc/localtime:ro
     ports:
       - "${SERVE_PORT}:80"
+    environment:
+      - SERVE_ENV=${SERVE_ENV}
+    restart: unless-stopped
+
+  # ==================== Worker 1: img_to_3d（重资源，建议只跑1个） ====================
+  img_worker:
+    container_name: "FiDA_${SERVE_ENV}_ImgWorker"
+    build:
+      context: .
+      dockerfile: Dockerfile
+    working_dir: /app
+    command: >
+      celery -A src.server.canvas_generate_3D.celery_app worker
+      -n img_worker@%h
+      -Q img_to_3d_queue
+      --concurrency=1
+      --prefetch-multiplier=1
+      --max-tasks-per-child=1
+      --loglevel=INFO
+    volumes:
+      - ./:/app
+      - ./.env:/app/.env
+      - /etc/localtime:/etc/localtime:ro
+    environment:
+      - SERVE_ENV=${SERVE_ENV}
+    depends_on:
+      - server          # 可选：等 server 启动后再启动 worker
+    restart: unless-stopped
+
+  # ==================== Worker 2: 3d_to_3views ====================
+  views_worker:
+    container_name: "FiDA_${SERVE_ENV}_ViewsWorker"
+    build:
+      context: .
+      dockerfile: Dockerfile
+    working_dir: /app
+    command: >
+      celery -A src.server.canvas_generate_3D.celery_app worker
+      -n views_worker@%h
+      -Q 3d_to_3views_queue
+      --concurrency=2
+      --prefetch-multiplier=1
+      --loglevel=INFO
+    volumes:
+      - ./:/app
+      - ./.env:/app/.env
+      - /etc/localtime:/etc/localtime:ro
+    environment:
+      - SERVE_ENV=${SERVE_ENV}
+    depends_on:
+      - server
+    restart: unless-stopped

pyproject.toml

@@ -29,9 +29,7 @@ dependencies = [
     "psycopg[binary]>=3.3.3",
     "postgres>=4.0",
     "langchain-huggingface>=1.2.0",
-    "sentence-transformers>=5.2.3",
     "rank-bm25>=0.2.2",
-    "torch>=2.10.0",
     "faiss-cpu>=1.13.2",
     "terminate>=0.0.9",
     "report-generator>=0.1.10",
@@ -51,4 +49,15 @@ dependencies = [
     "tool>=0.8.0",
     "langchain-daytona>=0.0.3",
     "langgraph-cli[inmem]>=0.4.19",
+    "start>=0.2",
+    "end>=1.3.1",
+    "annotated>=0.0.2",
+    "field>=0.2.0",
+    "aio-pika>=9.6.2",
+    "celery[redis]>=5.6.3",
+    "python3-pika>=0.9.14",
+    "tasks>=2.8.0",
+    "pika>=1.3.0",
+    "kombu>=5.4.0",
+    "sentence-transformers[onnx]>=5.3.0",
 ]

src/core/config.py

@@ -28,6 +28,11 @@ class Settings(BaseSettings):
     MINIO_SECRET: str = Field(default='', description="")
     MINIO_SECURE: bool = Field(default=True, description="")
 
+    # --- redis 配置信息 ---
+    REDIS_HOST: str = Field(default='', description="")
+    REDIS_PORT: str = Field(default='', description="")
+    REDIS_DB: int = Field(default=0, description="")
+
     # ---  mongodb配置信息 ---
     MONGODB_USERNAME: str = Field(default="", description="")
     MONGODB_PASSWORD: str = Field(default="", description="")
@@ -38,6 +43,7 @@ class Settings(BaseSettings):
     IMAGE_TO_3D_MODEL_URL: str = Field(default='', description="")
     FLUX2_GEN_IMG_MODEL_URL: str = Field(default='', description="")
     SEG_ANYTHING: str = Field(default='', description="")
+    RABBITMQ_URL: str = Field(default='', description="")
 
     # ---  外部工具api配置信息 ---
     TAVILY_API_KEY: str = Field(default="", description="")

src/routers/generate_3D.py

@@ -1,17 +1,23 @@
+import asyncio
 import json
 import logging
 
 import httpx
 import requests
+import uuid
 from fastapi import APIRouter
 
 from src.core.config import settings
 from src.schemas.generate_3D import ImageTo3DRequest, ToSVGRequest
 from src.schemas.response_template import ResponseModel
+from src.server.canvas_generate_3D.server import submit_img_to_3d_task, submit_three_d_to_3views_task
+from src.server.canvas_generate_3D.tasks import img_to_3d_task
 
 router = APIRouter(prefix="/canvas", tags=["Furniture Canvas"])
 logger = logging.getLogger(__name__)
 
+img_to_3d_semaphore = asyncio.Semaphore(1)
+
 
 @router.post("/img_to_3D")
 async def img_to_3D(request_data: ImageTo3DRequest):
@@ -106,6 +112,15 @@ async def img_to_3D(request_data: ImageTo3DRequest):
         logger.warning(f"img_to_3D Run Exception: {e}")
 
 
+@router.post("/img_to_3D_mq")
+async def img_to_3d_endpoint(request_data: ImageTo3DRequest):
+    result = submit_img_to_3d_task(
+        input_images=request_data.input_images,
+        model=request_data.model
+    )
+
+    return result
+
 
 @router.post("/3d_to_3views")
 async def to_3views(request_data: ToSVGRequest):
@@ -149,3 +164,9 @@ async def to_3views(request_data: ToSVGRequest):
 
     except Exception as e:
         logger.warning(f"img_to_3D Run Exception: {e}")
+
+
+@router.post("/3d_to_3views_mq")
+async def to_3views(request_data: ToSVGRequest):
+    result = submit_three_d_to_3views_task(minio_glb_path=request_data.minio_glb_path)
+    return result

src/server/canvas_generate_3D/celery_app.py

@@ -0,0 +1,60 @@
+from celery import Celery
+import os
+
+from kombu import Queue
+
+from src.core.config import settings
+
+# RabbitMQ 连接（请改成你的真实配置）
+BROKER_URL = settings.RABBITMQ_URL  # 用户名:密码@主机:端口/vhost
+
+celery_app = Celery(
+    "canvas_generate_3D",
+    broker=BROKER_URL,
+    backend=f"redis://{settings.REDIS_HOST}:{settings.REDIS_PORT}/{settings.REDIS_DB}",  # 推荐用 Redis 存任务结果
+    include=["src.server.canvas_generate_3D.tasks"],  # 明确包含任务模块
+)
+
+# 重要配置：限制并发为 1（一次只处理一个 img_to_3D 请求）
+celery_app.conf.update(
+    imports=[
+        'src.server.canvas_generate_3D.tasks',  # ← 加上这一行（或你的实际路径）
+    ],
+    task_serializer="json",
+    accept_content=["json"],
+    result_serializer="json",
+    timezone="Asia/Hong_Kong",
+    enable_utc=True,
+
+    # ==================== 新增：定义多个队列 ====================
+    task_queues=(
+        Queue("img_to_3d_queue", durable=True),
+        Queue("three_d_to_3views_queue", durable=True),
+    ),
+
+    task_routes={
+        'src.server.canvas_generate_3D.tasks.img_to_3d_task': {
+            'queue': 'img_to_3d_queue'
+        },
+        'src.server.canvas_generate_3D.tasks.three_d_to_3views_task': {  # 注意任务名称要一致
+            'queue': 'three_d_to_3views_queue'
+        },
+    },
+    task_default_queue="img_to_3d_queue",
+
+    # 全局或针对该队列的限制
+    worker_concurrency=1,  # 同时只跑 1 个
+    worker_prefetch_multiplier=1,  # 严格一次只预取 1 个
+    worker_max_tasks_per_child=1,  # 处理完一个后重启子进程（推荐用于重资源任务）
+    # 可选：任务 ack 策略（长任务建议晚 ack）
+    task_acks_late=True,
+    task_reject_on_worker_lost=True,
+)
+
+
+# 可选：打印已注册的任务，帮助调试
+@celery_app.on_after_configure.connect
+def setup_periodic_tasks(sender, **kwargs):
+    print("✅ Celery 已启动，以下任务已注册：")
+    for task_name in sorted(sender.tasks.keys()):
+        print(f"   - {task_name}")

src/server/canvas_generate_3D/server.py

@@ -0,0 +1,80 @@
+from celery import current_app
+from src.server.canvas_generate_3D.tasks import img_to_3d_task, three_d_to_3views_task
+
+
+def submit_img_to_3d_task(input_images: list, model: str = "single", **kwargs):
+    """
+    提交 3D 生成任务 - 队列最多堆积 10 个
+    """
+    queue_name = "img_to_3d_queue"
+    max_queue_length = 10
+
+    try:
+        with current_app.connection() as conn:  # 使用 Celery 的连接（最推荐）
+            with conn.channel() as channel:
+                queue_info = channel.queue_declare(queue=queue_name, durable=True, auto_delete=False, passive=False)
+                current_length = queue_info.message_count
+
+        # 队列已满
+        if current_length >= max_queue_length:
+            return {
+                "state": "queue_full",
+                "message": "当前 3D 生成请求较多，请等待片刻后重试。",
+                "queue_length": current_length,
+                "max_length": max_queue_length
+            }
+
+        task = img_to_3d_task.delay(input_images, model, **kwargs)
+
+        return {
+            "state": "success",
+            "task_id": task.id,
+            "message": "任务已成功提交，正在处理中...",
+            "queue_length": current_length + 1
+        }
+
+    except Exception as e:
+        return {
+            "state": "fail",
+            "message": f"提交失败，请稍后重试。错误: {str(e)}",
+            "error": str(e)
+        }
+
+
+def submit_three_d_to_3views_task(minio_glb_path: str, **kwargs):
+    """
+        提交 3D 生成任务 - 队列最多堆积 10 个
+        """
+    queue_name = "three_d_to_3views_queue"
+    max_queue_length = 3
+
+    try:
+        with current_app.connection() as conn:  # 使用 Celery 的连接（最推荐）
+            with conn.channel() as channel:
+                queue_info = channel.queue_declare(queue=queue_name, durable=True, auto_delete=False, passive=False)
+                current_length = queue_info.message_count
+
+        # 队列已满
+        if current_length >= max_queue_length:
+            return {
+                "state": "queue_full",
+                "message": "当前 3 视图 生成请求较多，请等待片刻后重试。",
+                "queue_length": current_length,
+                "max_length": max_queue_length
+            }
+
+        task = three_d_to_3views_task.delay(minio_glb_path, **kwargs)
+
+        return {
+            "state": "success",
+            "task_id": task.id,
+            "message": "任务已成功提交，正在处理中...",
+            "queue_length": current_length + 1
+        }
+
+    except Exception as e:
+        return {
+            "state": "fail",
+            "message": f"提交失败，请稍后重试。错误: {str(e)}",
+            "error": str(e)
+        }

src/server/canvas_generate_3D/tasks.py

@@ -0,0 +1,139 @@
+# src/server/canvas_generate_3D/tasks.py
+import json
+import time
+import httpx
+from celery import shared_task
+
+from src.core.config import settings
+from src.server.canvas_generate_3D.celery_app import celery_app
+import logging
+
+logger = logging.getLogger(__name__)
+
+
+def send_result_to_rabbitmq(result: dict, job_id: str, status: str = "completed"):
+    """发送结果到 RabbitMQ（建议后续移到 mq_util.py）"""
+    try:
+        # 你已经有 mq_util.py，可以调用那里面的函数
+        # 这里先用简单实现，如果你想用 mq_util.py 的方式，后面我再帮你调整
+        import pika
+        from pika import DeliveryMode
+
+        connection = pika.BlockingConnection(pika.URLParameters(settings.RABBITMQ_URL))
+        channel = connection.channel()
+
+        EXCHANGE_NAME = "img_to_3d_exchange"
+        ROUTING_KEY = "img_to_3d_results"
+        QUEUE_NAME = "img_to_3d_results"
+
+        channel.exchange_declare(exchange=EXCHANGE_NAME, exchange_type="direct", durable=True)
+        channel.queue_declare(queue=QUEUE_NAME, durable=True)
+        channel.queue_bind(exchange=EXCHANGE_NAME, queue=QUEUE_NAME, routing_key=ROUTING_KEY)
+
+        message_body = {
+            "job_id": job_id,
+            "status": status,
+            "timestamp": time.time(),
+            "result": result
+        }
+
+        channel.basic_publish(
+            exchange=EXCHANGE_NAME,
+            routing_key=ROUTING_KEY,
+            body=json.dumps(message_body).encode("utf-8"),
+            properties=pika.BasicProperties(delivery_mode=DeliveryMode.Persistent)
+        )
+        logger.info(f"✅ RabbitMQ 发送成功 | job_id: {job_id}")
+        connection.close()
+
+    except Exception as e:
+        logger.error(f"❌ RabbitMQ 发送失败 | job_id: {job_id} | {e}")
+
+
+@shared_task(bind=True, queue="img_to_3d_queue", max_retries=3, name='src.server.canvas_generate_3D.tasks.img_to_3d_task')
+def img_to_3d_task(self, input_images: list, model: str = "single"):
+    """img_to_3D 主任务"""
+
+    # ====================== 处理 job_id ======================
+    job_id = self.request.id  # 如果没传 job_id，就使用 Celery 自带的 task id
+
+    logger.info(f"开始处理 img_to_3D 任务 | job_id: {job_id} | celery_task_id: {self.request.id}")
+
+    try:
+        input_data = {
+            "image_paths": input_images,  # 注意：后端服务用的是 image_paths，不是 input_images
+            "model": model,
+        }
+
+        # 调用模型服务（推荐使用同步 httpx，避免 asyncio.run 在 worker 中的潜在问题）
+        with httpx.Client(timeout=300.0) as client:  # 改成同步 Client
+            resp = client.post(
+                f"http://{settings.IMAGE_TO_3D_MODEL_URL}/canvas/img_to_3D",
+                json=input_data
+            )
+            resp.raise_for_status()  # 自动抛出 HTTP 错误
+            result = resp.json()
+
+        logger.info(f"任务处理完成 | job_id: {job_id}")
+
+        # 发送 RabbitMQ 通知
+        send_result_to_rabbitmq(result=result, job_id=job_id, status="completed")
+
+        return result
+
+    except Exception as exc:
+        logger.error(f"任务失败 | job_id: {job_id} | error: {exc}", exc_info=True)
+
+        # 发送失败通知
+        send_result_to_rabbitmq(
+            result={"error": str(exc)},
+            job_id=job_id,
+            status="failed"
+        )
+
+        # 重试
+        raise self.retry(exc=exc, countdown=60, max_retries=3)
+
+
+@shared_task(bind=True, queue="three_d_to_3views_task", max_retries=3, name='src.server.canvas_generate_3D.tasks.three_d_to_3views_task')
+def three_d_to_3views_task(self, minio_glb_path: str):
+    """3D to 3views 主任务"""
+
+    # ====================== 处理 job_id ======================
+    job_id = self.request.id  # 如果没传 job_id，就使用 Celery 自带的 task id
+
+    logger.info(f"开始处理 three_d_to_3views_task 任务 | job_id: {job_id} | celery_task_id: {self.request.id}")
+
+    try:
+        input_data = {
+            "minio_glb_path": minio_glb_path,  # 注意：后端服务用的是 image_paths，不是 input_images
+        }
+
+        # 调用模型服务（推荐使用同步 httpx，避免 asyncio.run 在 worker 中的潜在问题）
+        with httpx.Client(timeout=300.0) as client:  # 改成同步 Client
+            resp = client.post(
+                f"http://{settings.IMAGE_TO_3D_MODEL_URL}/canvas/3d_to_3views",
+                json=input_data
+            )
+            resp.raise_for_status()  # 自动抛出 HTTP 错误
+            result = resp.json()
+
+        logger.info(f"任务处理完成 | job_id: {job_id}")
+
+        # 发送 RabbitMQ 通知
+        send_result_to_rabbitmq(result=result, job_id=job_id, status="completed")
+
+        return result
+
+    except Exception as exc:
+        logger.error(f"任务失败 | job_id: {job_id} | error: {exc}", exc_info=True)
+
+        # 发送失败通知
+        send_result_to_rabbitmq(
+            result={"error": str(exc)},
+            job_id=job_id,
+            status="failed"
+        )
+
+        # 重试
+        raise self.retry(exc=exc, countdown=60, max_retries=3)