调用YouTube API获取频道字幕时触发KeyError及后续TypeError问题求助

几个月前开发的程序，用于获取指定YouTube频道所有视频的字幕并统计特定词汇（如firework）的出现次数，当时运行完全正常，如今重新使用时无法正常工作。

原代码 #

from apiclient.discovery import build
from youtube_transcript_api import YouTubeTranscriptApi
from pytube import YouTube
import os

# Your Google API Key
api_key = "Your Youtube API here"
# The Youtube channel you want to extract data from
channel_id = "Youtube channel here"
# Youtube Build
youtube = build('youtube', 'v3', developerKey=api_key)

def get_channel_videos(channel_id):
 
    # get Uploads playlist id
    res = youtube.channels().list(id=channel_id,
                                  part='contentDetails').execute()
    playlist_id = res['items'][0]['contentDetails']['relatedPlaylists']['uploads']
    videos = []
    next_page_token = None
 
    while 1:
        res = youtube.playlistItems().list(playlistId=playlist_id,
                                           part='snippet',
                                           maxResults=50,
                                           pageToken=next_page_token).execute()
        videos += res['items']
        next_page_token = res.get('nextPageToken')
 
        if next_page_token is None:
            break

    return videos
 
videos = get_channel_videos(channel_id)
video_ids = []  # list of all video_id of channel
video_titles = []

# Word to search for
word1 = "firework"

# Counter for the text files
counter = 1

for video in videos:
    video_ids.append(video['snippet']['resourceId']['videoId'])

# Loops through every video on the youtube channel
for video_id in video_ids:
    try:
        # Gets the transcripts of youtube videos. Only gets English transcripts.
        responses = YouTubeTranscriptApi.get_transcript(
            video_id, languages=['en'])

        # Get's the youtube video's link
        yt = YouTube("https://www.youtube.com/watch?v="+str(video_id))
        
        #print('\n'+"Video: "+"https://www.youtube.com/watch?v="+str(video_id)+'\n'+'\n'+"Captions:")
        print("\n"+"Video Number: "+str(counter)+'\n'+"Video: "+"https://www.youtube.com/watch?v="+str(video_id)+'\n'+'Title: '+yt.title+'\n'+"Firework/s Wordcount:")

        
        # This is for text file generation
        filename = "video #" + str(counter) + ".txt"

        # Generates text file
        with open(filename, "w") as file:
            # Writes the title of the Youtube video at the top of the text file
            file.write(yt.title)
            # Whitespace
            file.write("\n")
            # Writes all the data grabbed from the transcripts into a text file
            # in text form.
            for i in responses:
                file.write("{}\n".format(i))
            counter += 1
            file.close()

            # This reads data from all the generated files,
            # then converts all the data within the file to lowercase,
            # then it counts the amount of times that the word you are searching
            # for is used within each file that is generated.
            fileTest = open(filename, "r")
            read_data = fileTest.read()
            word_count = read_data.lower().count(word1)
            file.close()

        # Prints out the word count.
        print(word_count)

        for response in responses:
            text = response['text']
     
    except Exception as e:
            print(e)

首次运行错误（KeyError） #

Traceback (most recent call last):
  File "D:\Downloads\YouTube-Caption-Project-main\main.py", line 47, in <module>
    videos = get_channel_videos(channel_id)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Downloads\YouTube-Caption-Project-main\main.py", line 30, in get_channel_videos
    playlist_id = res['items'][0]['contentDetails']['relatedPlaylists']['uploads']
                  ~~~^^^^^^^^^
KeyError: 'items'

参考相关问题修改后，出现新错误（TypeError）：

Traceback (most recent call last):
  File "D:\Downloads\YouTube-Caption-Project-main\main.py", line 62, in <module>
    for video in videos:
TypeError: 'NoneType' object is not iterable

解决方案 #

1. 排查API调用失败根源 #

• 验证API密钥有效性：登录Google Cloud控制台，确认YouTube Data API v3已启用，API密钥未过期、配额未耗尽，且没有IP访问限制。
• 检查频道ID正确性：确保channel_id是YouTube频道的原始ID（可通过频道页面源码或第三方工具获取），而非自定义URL或用户名。
• 增加API错误处理：在调用API后先判断响应是否包含有效数据，避免直接索引不存在的字段。

2. 修复get_channel_videos函数 #

修改函数，确保无论成功或失败都返回可迭代的列表，避免NoneType报错：

def get_channel_videos(channel_id):
    try:
        # 获取频道上传列表ID
        res = youtube.channels().list(id=channel_id, part='contentDetails').execute()
        if not res.get('items'):
            print("无法获取频道信息，请检查频道ID或API密钥")
            return []
        
        playlist_id = res['items'][0]['contentDetails']['relatedPlaylists']['uploads']
        videos = []
        next_page_token = None
 
        while True:
            res = youtube.playlistItems().list(
                playlistId=playlist_id,
                part='snippet',
                maxResults=50,
                pageToken=next_page_token
            ).execute()
            
            if res.get('items'):
                videos += res['items']
            
            next_page_token = res.get('nextPageToken')
            if next_page_token is None:
                break
                
        return videos
    except Exception as e:
        print(f"获取视频列表时出错: {str(e)}")
        return []

3. 优化字幕统计逻辑（可选） #

无需生成文本文件再读取统计，直接在获取字幕时计算，提升效率：

# 替换原统计部分代码
full_text = ' '.join([item['text'].lower() for item in responses])
word_count = full_text.count(word1.lower())
print(word_count)

同时，with语句会自动关闭文件，无需手动调用file.close()，可删除代码中多余的关闭操作。

4. 更新依赖库 #

YouTube相关库可能因平台结构更新失效，执行以下命令更新：

pip install --upgrade pytube youtube-transcript-api google-api-python-client

5. 细化异常捕获 #

针对字幕相关异常单独处理，提升错误提示清晰度：

from youtube_transcript_api import NoTranscriptFound, TranscriptsDisabled

# 在循环中替换原try-except块
try:
    responses = YouTubeTranscriptApi.get_transcript(video_id, languages=['en'])
    # ... 其他代码 ...
except NoTranscriptFound:
    print(f"视频 {video_id} 无英文字幕")
except TranscriptsDisabled:
    print(f"视频 {video_id} 字幕已禁用")
except Exception as e:
    print(f"处理视频 {video_id} 时出错: {str(e)}")

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