python-qlv-helper 0.6.0__py3-none-any.whl → 0.9.3__py3-none-any.whl

qlv_helper/utils/ocr_helper.py

@@ -9,11 +9,13 @@
 # Copyright ©2011-2025. Hunan xxxxxxx Company limited. All rights reserved.
 # ---------------------------------------------------------------------------------------------------------
 """
+import json
+import asyncio
 import ddddocr
 import requests
-from typing import Union, Tuple
+from typing import Tuple
 from aiohttp import ClientSession
-from playwright.async_api import Page, TimeoutError as PlaywrightTimeoutError
+from ocr_helper.core.baidu import ImageContentOCR
 
 # 复用 OCR 实例，不用每次都重新加载模型（更快）
 _ocr = ddddocr.DdddOcr(show_ad=False)
@@ -39,45 +41,38 @@ async def async_fetch_and_ocr_captcha(url: str) -> Tuple[str, bytes]:
     return result, img_bytes
 
 
-def recognize_captcha(image: Union[str, bytes]) -> str:
-    """
-    识别验证码图片，返回识别文本。
-    参数:
-        image: 图片路径 str，或图片的二进制 bytes
-    返回:
-        识别出的验证码字符串
-    """
-    try:
-        # 如果是路径，读取文件
-        if isinstance(image, str):
-            with open(image, "rb") as f:
-                img_bytes = f.read()
+async def get_image_text(image_path: str, captcha_type: int, api_key, secret_key: str) -> str:
+    if captcha_type == 0:
+        with open(image_path, "rb") as f:
+            img_bytes = f.read()
+        for _ in range(100):
+            text = _ocr.classification(img_bytes).strip()
+            if len(text) == 4:
+                return text
+        raise RuntimeError("ddddocr识别验证码失败")
+    else:
+        api = ImageContentOCR(api_key=api_key, secret_key=secret_key)
+        response = await api.get_access_token(is_end=False)
+        if not response.get("access_token"):
+            raise RuntimeError(f"获取百度API的认证Token失败，原因：{response}")
+        token = response.get("access_token")
+        response = await api.submit_request(
+            question='图片中的文字是什么，如果含有运算信息，请将运算结果返回。注意给我返回一个json格式数据包，例如：{"content":"xxxx", result: xxx}, 如果无运算信息，设置为空串就行',
+            image_path=image_path,
+            token=token,
+            is_end=False
+        )
+        task_id: str = response.get("result", dict()).get("task_id")
+        if not task_id:
+            raise RuntimeError(f"提交图片至百度API接口失败，原因：{response}")
+        await asyncio.sleep(delay=10)
+        response = await api.get_result(task_id=task_id, token=token, is_end=True)
+        if response.get("result").get("ret_code") == 0 and response.get("result").get("ret_msg") == "success":
+            description = response.get("result").get("description")
+            description = json.loads(description[description.find("{"):description.find("}") + 1])
+            if description.get("result"):
+                return str(description.get("result")).strip()
+            else:
+                return description.get("content").strip()
         else:
-            img_bytes = image
-
-        result = _ocr.classification(img_bytes)
-        return result
-
-    except Exception as e:
-        raise RuntimeError(f"OCR 识别失败: {e}")
-
-
-async def get_image_text(page: Page, selector: str, timeout: float = 5.0) -> Tuple[bool, str]:
-    try:
-        # 找到 img
-        locator = page.locator(selector)
-        if locator:
-            img = await locator.element_handle(timeout=timeout * 1000)
-
-            # 直接截图获取原始图片字节，不刷新图片
-            img_bytes = await img.screenshot(timeout=timeout * 1000)
-
-            # OCR 识别
-            text = _ocr.classification(img_bytes)
-            return True, text.strip()
-        else:
-            return False, f'没有找到当前页面中的【{selector}】图片'
-    except PlaywrightTimeoutError:
-        return False, f"元素 '{selector}' 未在 {timeout} 秒内找到"
-    except Exception as e:
-        return False, f"检查元素时发生错误: {str(e)}"
+            raise RuntimeError(f"调用百度API，获取图片识别结果失败，原因{response}")