OCR接口:支持识别pdf

api/models.py

@@ -37,6 +37,7 @@ class OCRRequest(BaseModel):
 
     image: str
     type: str
+    file_type: int = 1
 
 
 class MessageCreateBill(BaseModel):

api/routes.py

@@ -56,7 +56,9 @@ async def orc(request: OCRRequest):
             raise HTTPException(400, f"图片格式错误: {str(e)}")
 
         # 2. OCR识别
-        result = await doc_service.pure_ocr(image_bytes=image_bytes)
+        result = await doc_service.pure_ocr(
+            image_bytes=image_bytes, file_type=request.file_type
+        )
 
         # 3. 返回结果
         return result

core/document_processor/document_service.py

@@ -76,13 +76,14 @@ class DocumentProcessingService:
         return result
 
     async def pure_ocr(
-        self, image_bytes: bytes, ocr_options: Dict = None
+        self, image_bytes: bytes, file_type: int = 1, ocr_options: Dict = None
     ) -> Dict[str, Any]:
         """
         扫描图片并返回OCR识别结果
 
         Args:
             image_bytes: 图片字节
+            file_type: 文件类型，1: 图片, 0: PDF
             ocr_options: OCR选项
 
         Returns:
@@ -90,7 +91,7 @@ class DocumentProcessingService:
         """
         # 1. OCR识别
         ocr_result = await self.ocr_service.recognize_image_async(
-            image_bytes, **(ocr_options or {})
+            image_bytes, file_type, **(ocr_options or {})
         )
 
         # 2. 提取文本