AI图片

api/__init__.py

@@ -1,9 +1,10 @@
 from fastapi import APIRouter
-from api import devices, contents, todos
+from api import devices, contents, todos, ai
 
 api_router = APIRouter()
 
 # 注册所有路由，并添加全局安全要求
 api_router.include_router(devices.router, prefix="/devices")
 api_router.include_router(contents.router, prefix="/contents")
-api_router.include_router(todos.router, prefix="/todos")
+api_router.include_router(todos.router, prefix="/todos")
+api_router.include_router(ai.router, prefix="/ai", tags=["AI生成"])

api/ai.py

@@ -0,0 +1,157 @@
+from fastapi import APIRouter, HTTPException, Depends
+from typing import Dict, Any
+import httpx
+import json
+import logging
+from config import settings
+from api.ai_schemas import AIGenerationRequest, AITaskResponse, AITaskResult
+
+router = APIRouter()
+logger = logging.getLogger(__name__)
+
+DASHSCOPE_API_URL = "https://dashscope.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation"
+DASHSCOPE_TASK_URL = "https://dashscope.aliyuncs.com/api/v1/tasks"
+
+@router.post("/generate", response_model=AITaskResponse, summary="提交AI图片生成任务")
+async def generate_image(request: AIGenerationRequest):
+    """
+    提交AI图片生成任务，使用阿里云DashScope服务
+    """
+    if not settings.dashscope_api_key:
+        raise HTTPException(status_code=500, detail="DashScope API Key not configured")
+
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {settings.dashscope_api_key}",
+        "X-DashScope-Async": "enable"  # 确保异步任务提交
+    }
+
+    # 构建请求体
+    payload = {
+        "model": request.model,
+        "input": {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {
+                            "text": request.prompt
+                        }
+                    ]
+                }
+            ]
+        },
+        "parameters": {
+            "prompt_extend": True,
+            "watermark": False,
+            "n": request.n,
+            "size": request.size
+        }
+    }
+    
+    if request.negative_prompt:
+        payload["parameters"]["negative_prompt"] = request.negative_prompt
+
+    try:
+        async with httpx.AsyncClient() as client:
+            response = await client.post(
+                DASHSCOPE_API_URL,
+                headers=headers,
+                json=payload,
+                timeout=30.0
+            )
+            
+            if response.status_code != 200:
+                logger.error(f"DashScope API error: {response.text}")
+                try:
+                    error_detail = response.json()
+                except:
+                    error_detail = {"message": response.text}
+                raise HTTPException(status_code=response.status_code, detail=error_detail)
+            
+            result = response.json()
+            # 检查是否有task_id，因为如果是同步返回可能没有task_id，但在这种模型通常是异步的
+            # 这里的curl示例似乎是直接返回task_id
+            
+            if "output" in result and "task_id" in result["output"]:
+                 task_id = result["output"]["task_id"]
+            elif "task_id" in result: # 或者是这种结构
+                 task_id = result["task_id"]
+            else:
+                 # 有些情况下直接返回 output.task_id
+                 # 根据文档 https://help.aliyun.com/zh/dashscope/developer-reference/api-details-10
+                 # 异步提交返回结构通常包含 output.task_id
+                 if "output" in result and "task_id" in result["output"]:
+                     task_id = result["output"]["task_id"]
+                 else:
+                     # 如果是同步返回，可能直接给结果，或者结构不同
+                     # 但 wan2.6-t2i 通常是异步任务
+                     logger.warning(f"Unexpected response structure: {result}")
+                     # 尝试直接取，或者抛错
+                     # 假设是标准异步结构
+                     task_id = result.get("output", {}).get("task_id")
+            
+            if not task_id:
+                 raise HTTPException(status_code=500, detail="Failed to retrieve task_id from DashScope response")
+
+            return AITaskResponse(
+                task_id=task_id,
+                request_id=result.get("request_id")
+            )
+
+    except httpx.RequestError as e:
+        logger.error(f"Request error: {str(e)}")
+        raise HTTPException(status_code=500, detail=f"Request failed: {str(e)}")
+
+@router.get("/tasks/{task_id}", response_model=AITaskResult, summary="查询AI任务结果")
+async def get_task_result(task_id: str):
+    """
+    查询AI生成任务的结果
+    """
+    if not settings.dashscope_api_key:
+        raise HTTPException(status_code=500, detail="DashScope API Key not configured")
+
+    headers = {
+        "Authorization": f"Bearer {settings.dashscope_api_key}"
+    }
+
+    try:
+        async with httpx.AsyncClient() as client:
+            response = await client.get(
+                f"{DASHSCOPE_TASK_URL}/{task_id}",
+                headers=headers,
+                timeout=30.0
+            )
+            
+            if response.status_code != 200:
+                logger.error(f"DashScope Task API error: {response.text}")
+                raise HTTPException(status_code=response.status_code, detail="Failed to fetch task status")
+            
+            result = response.json()
+            
+            # 构建返回结果
+            # DashScope 任务查询返回结构：
+            # {
+            #     "request_id": "...",
+            #     "output": {
+            #         "task_id": "...",
+            #         "task_status": "SUCCEEDED",
+            #         "results": [...]
+            #     },
+            #     "usage": ...
+            # }
+            
+            task_status = result.get("output", {}).get("task_status", "UNKNOWN")
+            
+            return AITaskResult(
+                task_id=task_id,
+                status=task_status,
+                code=result.get("code"),
+                message=result.get("message"),
+                output=result.get("output"),
+                usage=result.get("usage")
+            )
+
+    except httpx.RequestError as e:
+        logger.error(f"Request error: {str(e)}")
+        raise HTTPException(status_code=500, detail=f"Request failed: {str(e)}")

api/ai_schemas.py

@@ -0,0 +1,30 @@
+from pydantic import BaseModel, Field
+from typing import Optional, List, Dict, Any
+
+# AI生成相关模型
+class AIGenerationRequest(BaseModel):
+    prompt: str = Field(..., description="生成图片的提示词")
+    negative_prompt: Optional[str] = Field(None, description="反向提示词")
+    size: str = Field("1024*1024", description="图片尺寸")
+    n: int = Field(1, description="生成数量", ge=1, le=4)
+    model: str = Field("wan2.6-t2i", description="使用的模型")
+
+class AITemplateGenerationRequest(BaseModel):
+    template_id: str = Field(..., description="提示词模板ID")
+    params: Dict[str, str] = Field(default_factory=dict, description="提示词参数")
+    negative_prompt: Optional[str] = Field(None, description="反向提示词")
+    size: str = Field("1024*1024", description="图片尺寸")
+    n: int = Field(1, description="生成数量", ge=1, le=4)
+    model: str = Field("wan2.6-t2i", description="使用的模型")
+
+class AITaskResponse(BaseModel):
+    task_id: str = Field(..., description="任务ID")
+    request_id: Optional[str] = Field(None, description="请求ID")
+
+class AITaskResult(BaseModel):
+    task_id: str
+    status: str
+    code: Optional[str] = None
+    message: Optional[str] = None
+    output: Optional[Dict[str, Any]] = None
+    usage: Optional[Dict[str, Any]] = None

api/prompts.py

@@ -0,0 +1,31 @@
+# 预设提示词模板库
+# 使用Python的format语法进行参数替换，例如 {keyword}
+
+PROMPTS = {
+    "flower_shop": "一间有着精致窗户的花店，漂亮的木质门，摆放着{flower_type}，风格为{style}",
+    "landscape": "宏伟的{season}自然风景，包含{element}，光线为{lighting}，高分辨率，写实风格",
+    "cyberpunk_city": "赛博朋克风格的未来城市，霓虹灯闪烁，{weather}天气，街道上有{vehicle}",
+    "portrait": "一张{gender}的肖像照，{expression}表情，背景是{background}，专业摄影布光",
+    "default": "一间有着精致窗户的花店，漂亮的木质门，摆放着花朵"
+}
+
+def get_prompt(template_id: str, **kwargs) -> str:
+    """
+    获取并格式化提示词
+    :param template_id: 提示词模板ID
+    :param kwargs: 替换参数
+    :return: 格式化后的提示词
+    """
+    template = PROMPTS.get(template_id)
+    if not template:
+        return None
+    
+    try:
+        # 使用 safe_substitute 避免参数缺失报错? 
+        # Python的format如果缺参数会报错，这里直接用format，让调用者负责提供完整参数
+        # 或者我们可以在这里处理默认值
+        return template.format(**kwargs)
+    except KeyError as e:
+        # 如果缺少参数，抛出异常或返回包含未替换占位符的字符串
+        # 这里为了简单，如果出错，我们尽量保留原样或者报错
+        raise ValueError(f"缺少提示词参数: {e}")