Update app.py

app.py

@@ -1,4 +1,4 @@
-from fastapi import FastAPI, HTTPException, Depends, Request
+from fastapi import FastAPI, HTTPException, Depends, Request, Header
 from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel
 import os
@@ -6,7 +6,6 @@ import numpy as np
 from sentence_transformers import SentenceTransformer
 from typing import List, Optional
 import logging
-from fastapi import Header
 
 # 配置日志
 logging.basicConfig(
@@ -33,7 +32,7 @@ app.add_middleware(
 # 模型映射：OpenAI模型名 → 开源模型名
 MODEL_MAPPING = {
     "text-embedding-3-small": "BAAI/bge-small-en-v1.5",
-    "text-embedding-3-large": "BAAI/bge-large-en-v1.5"  # 新增大模型映射
+    "text-embedding-3-large": "BAAI/bge-large-en-v1.5"
 }
 
 # 加载模型（懒加载，首次请求时加载）
@@ -42,12 +41,10 @@ models = {}
 def get_model(model_name: str):
     logger.info(f"尝试获取模型: {model_name}")
     if model_name not in models:
-        # 检查是否支持该模型
         if model_name not in MODEL_MAPPING:
             error_msg = f"不支持的模型: {model_name}"
             logger.error(error_msg)
             raise HTTPException(status_code=400, detail=error_msg)
-        # 加载模型
         logger.info(f"开始加载模型: {MODEL_MAPPING[model_name]}")
         try:
             models[model_name] = SentenceTransformer(MODEL_MAPPING[model_name])
@@ -59,8 +56,8 @@ def get_model(model_name: str):
     return models[model_name]
 
 # 验证API密钥
-def verify_api_key(authorization: Optional[str] = None):
-    logger.info("验证API密钥")
+def verify_api_key(authorization: Optional[str] = Header(None)):
+    logger.info("执行API密钥验证")
     logger.info(f"Authorization头部内容: {authorization}")
     if not authorization or not authorization.startswith("Bearer "):
         logger.warning("未提供有效的API密钥格式")
@@ -72,13 +69,13 @@ def verify_api_key(authorization: Optional[str] = None):
     logger.info("API密钥验证通过")
     return True
 
-# 请求体模型（对齐OpenAI格式）
+# 请求体模型
 class EmbeddingRequest(BaseModel):
     input: str or List[str]
     model: str
-    encoding_format: Optional[str] = "float"  # 仅支持float，忽略base64
+    encoding_format: Optional[str] = "float"
 
-# 响应体模型（对齐OpenAI格式）
+# 响应体模型
 class EmbeddingData(BaseModel):
     object: str = "embedding"
     embedding: List[float]
@@ -90,23 +87,26 @@ class EmbeddingResponse(BaseModel):
     model: str
     usage: dict = {"prompt_tokens": 0, "total_tokens": 0}
 
-@app.post("/embeddings", response_model=EmbeddingResponse)
+@app.post("/v1/embeddings", response_model=EmbeddingResponse)
 async def create_embedding(
-    request: Request,  # 接收Request对象
-    req: EmbeddingRequest,
-    _: bool = Depends(verify_api_key)
+    request: Request,
+    req: EmbeddingRequest
 ):
-    # 打印完整请求信息
+    # 先打印完整请求信息（在验证之前）
     logger.info("\n===== 接收到的完整请求信息 =====")
     logger.info(f"请求方法: {request.method}")
     logger.info(f"请求URL: {request.url}")
     logger.info("请求头部:")
     for name, value in request.headers.items():
         logger.info(f"  {name}: {value}")
-    logger.info(f"请求体: {await request.body()}")  # 打印原始请求体
+    logger.info(f"请求体: {await request.body()}")
     logger.info("===============================\n")
-
-    # 原有逻辑保持不变
+    
+    # 手动执行验证（在打印日志之后）
+    authorization = request.headers.get("Authorization")
+    verify_api_key(authorization)
+    
+    # 原有嵌入处理逻辑
     logger.info(f"收到嵌入请求，模型: {req.model}, 输入类型: {type(req.input)}")
     try:
         model = get_model(req.model)
@@ -135,18 +135,6 @@ async def create_embedding(
         logger.error(error_msg)
         raise HTTPException(status_code=500, detail=error_msg)
 
-# 健康检查接口也打印完整请求
-@app.post("/v1/embeddings")
-async def v1_check(request: Request):
-    logger.info("\n===== v1请求信息 =====")
-    logger.info(f"请求方法: {request.method}")
-    logger.info(f"请求URL: {request.url}")
-    logger.info("请求头部:")
-    for name, value in request.headers.items():
-        logger.info(f"  {name}: {value}")
-    logger.info("===============================\n")
-    return {"status": "healthy", "models": list(MODEL_MAPPING.keys())}
-
 @app.get("/health")
 async def health_check(request: Request):
     logger.info("\n===== 健康检查请求信息 =====")