zltest01
/
huoyan-enterprise
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update

This commit is contained in:
silk 2026-06-15 19:58:47 +08:00
parent eb44dc3491
commit b4b37240d0
3 changed files with 18 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
backend/.env.example
View File

@ -79,6 +79,7 @@ EMBEDDING_DIMENSION=1536 # Embedding 维度
OCR_ACCESS_KEY_ID=修改此项为阿里云 OCR 访问密钥ID OCR_ACCESS_KEY_ID=修改此项为阿里云 OCR 访问密钥ID
OCR_ACCESS_KEY_SECRET=修改此项为阿里云 OCR 访问密钥Secret OCR_ACCESS_KEY_SECRET=修改此项为阿里云 OCR 访问密钥Secret
OCR_ENDPOINT=修改此项为阿里云 OCR 终端节点 OCR_ENDPOINT=修改此项为阿里云 OCR 终端节点
OCR_TIMEOUT_SECONDS=120
OCR_USE_LOCAL=false OCR_USE_LOCAL=false
MODERATION_ENABLED=false MODERATION_ENABLED=false

1
backend/core/config.py
View File

@ -105,6 +105,7 @@ class Settings(BaseSettings):
ocr_access_key_id: Optional[str] = None ocr_access_key_id: Optional[str] = None
ocr_access_key_secret: Optional[str] = None ocr_access_key_secret: Optional[str] = None
ocr_endpoint: str = "ocr-api.cn-hangzhou.aliyuncs.com" # OCR 服务端点 ocr_endpoint: str = "ocr-api.cn-hangzhou.aliyuncs.com" # OCR 服务端点
ocr_timeout_seconds: float = 120.0 # OCR 请求超时（含上传图片 body）
# 微信小程序配置 # 微信小程序配置
wechat_app_id: Optional[str] = None wechat_app_id: Optional[str] = None

21
backend/services/vector_service.py
View File

@ -205,15 +205,21 @@ class VectorService:
# 初始化阿里云 OCR（图片、扫描 PDF、DOCX 内嵌图均依赖云端识别） # 初始化阿里云 OCR（图片、扫描 PDF、DOCX 内嵌图均依赖云端识别）
self.ocr_engine = None self.ocr_engine = None
self._ocr_timeout_ms = int(settings.ocr_timeout_seconds * 1000)
if ALIYUN_OCR_AVAILABLE and settings.ocr_access_key_id and settings.ocr_access_key_secret: if ALIYUN_OCR_AVAILABLE and settings.ocr_access_key_id and settings.ocr_access_key_secret:
try: try:
config = open_api_models.Config( config = open_api_models.Config(
access_key_id=settings.ocr_access_key_id, access_key_id=settings.ocr_access_key_id,
access_key_secret=settings.ocr_access_key_secret, access_key_secret=settings.ocr_access_key_secret,
endpoint=settings.ocr_endpoint endpoint=settings.ocr_endpoint,
connect_timeout=self._ocr_timeout_ms,
read_timeout=self._ocr_timeout_ms,
) )
self.ocr_engine = OcrClient(config) self.ocr_engine = OcrClient(config)
logger.info("✅ 阿里云 OCR 已启用，将使用云端 OCR 服务识别图片文字") logger.info(
"✅ 阿里云 OCR 已启用，将使用云端 OCR 服务识别图片文字 "
f"(endpoint={settings.ocr_endpoint}, timeout={settings.ocr_timeout_seconds}s)"
)
except Exception as e: except Exception as e:
logger.warning(f"⚠️ 阿里云 OCR 初始化失败: {e}") logger.warning(f"⚠️ 阿里云 OCR 初始化失败: {e}")
elif not ALIYUN_OCR_AVAILABLE: elif not ALIYUN_OCR_AVAILABLE:
@ -261,7 +267,7 @@ class VectorService:
image_bytes = f.read() image_bytes = f.read()
image_size_kb = len(image_bytes) / 1024 image_size_kb = len(image_bytes) / 1024
logger.debug(f"📊 [阿里云OCR] 图片大小: {image_size_kb:.2f}KB") logger.info(f"📊 [阿里云OCR] 图片大小: {image_size_kb:.2f}KB")
# 使用 StreamClient 读取字节流（阿里云 SDK 要求的格式） # 使用 StreamClient 读取字节流（阿里云 SDK 要求的格式）
body_stream = StreamClient.read_from_bytes(image_bytes) body_stream = StreamClient.read_from_bytes(image_bytes)
@ -270,8 +276,13 @@ class VectorService:
# 构建请求 # 构建请求
request = ocr_models.RecognizeGeneralRequest(body=body_stream) request = ocr_models.RecognizeGeneralRequest(body=body_stream)
# 运行时选项 # 运行时选项（默认超时过短会导致大图片 write timeout）
runtime = util_models.RuntimeOptions() runtime = util_models.RuntimeOptions(
connect_timeout=self._ocr_timeout_ms,
read_timeout=self._ocr_timeout_ms,
autoretry=True,
max_attempts=2,
)
logger.debug(f"☁️ [阿里云OCR] 调用 API: recognize_general_with_options") logger.debug(f"☁️ [阿里云OCR] 调用 API: recognize_general_with_options")
# 调用阿里云 OCR API（使用 with_options 版本） # 调用阿里云 OCR API（使用 with_options 版本）