Google Vision Python客户端快速启动报TypeError: bases must be types

本地运行Google Vision Python库的quickstart示例，执行python3 quickstart.py命令时抛出如下报错：

Traceback (most recent call last):
  File "/Users/penguin/.../quickstart.py", line 53, in <module>
    print(run_quickstart())
  File "/Users/penguin/.../quickstart.py", line 25, in run_quickstart
    from google.cloud import vision
  File "/Users/pengui/anaconda3/envs/vision/lib/python3.9/site-packages/google/cloud/vision/__init__.py", line 18, in <module>
    from google.cloud.vision_v1.services.image_annotator.async_client import (
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/cloud/vision_v1/__init__.py", line 17, in <module>
    from google.cloud.vision_helpers.decorators import add_single_feature_methods
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/cloud/vision_helpers/__init__.py", line 16, in <module>
    import proto
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/proto/__init__.py", line 15, in <module>
    from .enums import Enum
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/proto/enums.py", line 17, in <module>
    from google.protobuf import descriptor_pb2
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/protobuf/descriptor_pb2.py", line 5, in <module>
    from google.protobuf.internal import builder as _builder
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/protobuf/internal/builder.py", line 42, in <module>
    from google.protobuf import reflection as _reflection
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/protobuf/reflection.py", line 51, in <module>
    from google.protobuf import message_factory
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/protobuf/message_factory.py", line 42, in <module>
    from google.protobuf.internal import api_implementation
  File "/Users/penguin/anaconda3/envs/vision/lib/python3.9/site-packages/google/protobuf/internal/api_implementation.py", line 104, in <module>
    from google.protobuf.pyext import _message
TypeError: bases must be types

• Google Cloud Shell环境中运行同一份quickstart示例无异常
• 接口凭证配置正确，环境变量GOOGLE_APPLICATION_CREDENTIALS=key.json设置无误
• 本地运行环境Python版本为3.9.12
• 已通过pip3安装requirements.txt指定的全部依赖，对应google-cloud-vision版本为2.7.2
• python3与pip3均指向/anaconda3/envs/vision虚拟环境路径，不存在多环境混用问题
• 本地操作系统为macOS Monterey 12.4

报错触发点在protobuf C扩展加载环节，属于Anaconda环境下的典型依赖冲突：当前环境中安装的protobuf版本过高，和google-cloud-vision 2.7.2依赖的proto-plus库存在兼容性问题，部分通过conda源安装的protobuf版本还存在macOS平台C扩展编译适配问题，直接触发TypeError: bases must be types错误。

1. 激活目标虚拟环境：
   conda activate vision
2. 彻底卸载环境中已安装的冲突包：

pip uninstall -y protobuf proto-plus
conda remove -y protobuf

3. 安装经过兼容性验证的依赖版本：

pip install "protobuf>=3.19.0,<3.20.0" "proto-plus>=1.22.0"

4. 校验安装结果：执行pip list | grep proto，确认protobuf版本处于3.19.x区间、proto-plus版本不低于1.22.0即可。
5. 重新执行python3 quickstart.py即可正常运行。

注意：修复过程中不要使用conda安装protobuf相关依赖，conda源的protobuf包在macOS平台下容易出现C扩展适配问题，直接通过pip安装官方预编译的wheel包可规避该类问题。

内容的提问来源于stack exchange，提问作者William Torkington