我正在为自己制作一个小型 python 应用程序,用于从 YouTube 视频下载信息,为此我使用 YouTube api。
最近我看了这个视频,帮助我从Youtube视频中获取评论和他们的回复,并将它们导出到Excel文件,一切正常。然而,我现在遇到的问题是,我也想从 YouTube 视频中提取标题,但似乎无法让我的代码为其工作。
就像我之前提到的,我观看了链接的视频,并尝试向视频频道的作者发表评论以寻求帮助,不幸的是我没有得到回复。
我还尝试在 YouTube 上查找有关我的问题的任何其他有用视频,但找不到任何对我的问题有帮助的内容。
除了在那里询问和寻找其他视频之外,我还尝试查看文档并像这样迷惑代码,但这也不起作用。以下是我使用的代码:
# CODE FROM HERE: https://github.com/analyticswithadam/Python/blob/main/Pull_all_Comments_and_Replies_for_YouTube_Playlists.ipynb
from googleapiclient.discovery import build
import pandas as pd
import getpass
api_key = "API key here"
playlist_ids = ['Youtube Playlist Link Here']
# Build the YouTube client
youtube = build('youtube', 'v3', developerKey=api_key)
def get_all_video_ids_from_playlists(youtube, playlist_ids):
all_videos = [] # Initialize a single list to hold all video IDs
for playlist_id in playlist_ids:
next_page_token = None
# Fetch videos from the current playlist
while True:
playlist_request = youtube.playlistItems().list(
part='contentDetails',
playlistId=playlist_id,
maxResults=50,
pageToken=next_page_token)
playlist_response = playlist_request.execute()
all_videos += [item['contentDetails']['videoId'] for item in playlist_response['items']]
next_page_token = playlist_response.get('nextPageToken')
if next_page_token is None:
break
return all_videos
# Fetch all video IDs from the specified playlists
video_ids = get_all_video_ids_from_playlists(youtube, playlist_ids)
# Now you can pass video_ids to the next function
# next_function(video_ids)
'''
# Broken-ass title code
def get_vid_title(youtube, video_id): # Added video_id as an argument
all_titles = []
next_page_token = None
while True:
title_request = youtube.channels().list(
part="snippet",
videoId=video_id,
pageToken=next_page_token,
textFormat="plainText",
maxResults=100
)
title_response = title_request.execute()
for item in title_response['items']:
vid_title = item['snippet']['title']
all_titles.append({
'Title': vid_title['title']
})
print(vid_title['title'])
return all_titles
'''
# Function to get replies for a specific comment
def get_replies(youtube, parent_id, video_id): # Added video_id as an argument
replies = []
next_page_token = None
while True:
reply_request = youtube.comments().list(
part="snippet",
parentId=parent_id,
textFormat="plainText",
maxResults=100,
pageToken=next_page_token
)
reply_response = reply_request.execute()
for item in reply_response['items']:
comment = item['snippet']
replies.append({
'Timestamp': comment['publishedAt'],
'Username': comment['authorDisplayName'],
'VideoID': video_id,
'Comment': comment['textDisplay'],
'likeCount': comment['likeCount'],
'Date': comment['updatedAt'] if 'updatedAt' in comment else comment['publishedAt']
})
next_page_token = reply_response.get('nextPageToken')
if not next_page_token:
break
return replies
# Function to get all comments (including replies) for a single video
def get_comments_for_video(youtube, video_id):
all_comments = []
next_page_token = None
while True:
comment_request = youtube.commentThreads().list(
part="snippet",
videoId=video_id,
pageToken=next_page_token,
textFormat="plainText",
maxResults=100
)
comment_response = comment_request.execute()
for item in comment_response['items']:
top_comment = item['snippet']['topLevelComment']['snippet']
all_comments.append({
'Timestamp': top_comment['publishedAt'],
'Username': top_comment['authorDisplayName'],
'VideoID': video_id, # Directly using video_id from function parameter
'Comment': top_comment['textDisplay'],
'likeCount': top_comment['likeCount'],
'Date': top_comment['updatedAt'] if 'updatedAt' in top_comment else top_comment['publishedAt']
})
# Fetch replies if there are any
if item['snippet']['totalReplyCount'] > 0:
all_comments.extend(get_replies(youtube, item['snippet']['topLevelComment']['id'], video_id))
next_page_token = comment_response.get('nextPageToken')
if not next_page_token:
break
return all_comments
# List to hold all comments from all videos
all_comments = []
for video_id in video_ids:
video_comments = get_comments_for_video(youtube, video_id)
all_comments.extend(video_comments)
# Create DataFrame
comments_df = pd.DataFrame(all_comments)
# Export whole dataset to the local machine as CSV File
csv_file = 'comments_data.csv' # Name your file
comments_df.to_csv(csv_file, index=False)
任何帮助将不胜感激。
在
get_all_video_ids_from_playlists
使用:
part='snippet,contentDetails'
在这一行:
all_videos += [item['contentDetails']['videoId'] for item in playlist_response['items']]
更改如下:
for pl_item in playlist_response
all_videos.append({"Video_ID": item['contentDetails']['videoId'], {"Title" : {item['snippet']['title']}})
然后,在这一行:
for video_id in video_ids:
video_comments = get_comments_for_video(youtube, video_id)
您正在通过
item
- 即 {"Video_ID: "xxxx", "Title": "xxxx"}
:
最后,关于你的
get_comments_for_video
功能:
# Function to get all comments (including replies) for a single video
def get_comments_for_video(youtube, video_id):
用途:
videoId=video_id["Video_ID"]
和:
'VideoID': video_id["Video_ID"], # Directly using video_id from function parameter
'Title': video_id["Title"], # The title of the video.