使用Sightengine API时遇AttributeError：模块无Client属性

以下是针对你遇到的API调用问题的具体解决方案：

• 升级并确认Sightengine SDK版本
  你的代码使用的是旧版SDK的调用逻辑，当前官方Python SDK已更新了核心类名。先执行命令确保安装最新版本：

  pip install --upgrade sightengine
  

• 修正客户端初始化代码
  替换原有导入和客户端创建逻辑：
  原代码片段：

  import sightengine
  client = sightengine.Client(sightengine_user, sightengine_secret)
  

  修改为：

  from sightengine.client import SightengineClient
  client = SightengineClient(sightengine_user, sightengine_secret)
  

• 更新视频工作流检查调用方式
  新版SDK中视频检查的方法已拆分，原client.check需替换为专门的视频工作流检查方法：
  原代码片段：

  response = client.check('workflow', workflow_id, video_path)
  

  修改为：

  response = client.video.check_workflow(workflow_id, video_path)
  

• 排查同名文件冲突
  检查项目目录下是否存在sightengine.py文件或同名文件夹，这会导致Python误导入本地文件而非官方SDK包。如果有，立即重命名该文件/文件夹。

修改后的完整代码片段示例：

from sightengine.client import SightengineClient
import os
import requests
import praw
import cv2
import subprocess

# Sightengine API credentials
sightengine_user = "CENSORED"
sightengine_secret = "CENSORED"

# Authenticate with Sightengine API
client = SightengineClient(sightengine_user, sightengine_secret)

#SKIPPED CODE ----------------------------#

# Check if the video is shorter than 60 seconds
if duration < 60:
    print("mp4_lenght=shorter-60s_[codecontinued]")

    # Apply workflow on the video
    workflow_id = "CENSORED"
    video_path = "video.mp4"
    response = client.video.check_workflow(workflow_id, video_path)
    print("started_for_searching_unwanted_content")

    # Check if the video contains unwanted content
    if response['status'] == 'success' and response['media']['status'] == 'success':
        if response['media']['warnings']:
            print("mp4_content=unwanted")
        else:
            print("mp4_content=wanted")
    else:
        print("Error: ", response['status'])

else:
    print("mp4_lenght=longer-60s_[codestopped]")
    exit()

内容的提问来源于stack exchange，提问作者Toni Storm