I am making a small python application for myself that downloads the information from YouTube videos and for this I am using the YouTube api.
Recently I watched this video to help me get comments and their replies from a Youtube video and export them to an excel file, and everything works fine. However, the problem I have now, is that I want to extract the title as well from the YouTube video and can't seem to get my code working for it.
Like I mentioned previously, I watched the linked video and I also tried commenting on the author of the video's channel for help, unfortunately I did not get a reply.
I also tried looking around YouTube for any other helpful videos regarding my problem, but couldn't really find anything helpful to my problem.
Apart from asking there and looking for other videos, I also tried looking at the documentation and puzzling the code out like that, which also didn't work. Below is the code I am used:
# CODE FROM HERE: https://github.com/analyticswithadam/Python/blob/main/Pull_all_Comments_and_Replies_for_YouTube_Playlists.ipynbfrom googleapiclient.discovery import buildimport pandas as pdimport getpassapi_key = "API key here"playlist_ids = ['Youtube Playlist Link Here']# Build the YouTube clientyoutube = 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 playlistsvideo_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 codedef 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 commentdef 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 videodef 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 videosall_comments = []for video_id in video_ids: video_comments = get_comments_for_video(youtube, video_id) all_comments.extend(video_comments)# Create DataFramecomments_df = pd.DataFrame(all_comments)# Export whole dataset to the local machine as CSV Filecsv_file = 'comments_data.csv' # Name your filecomments_df.to_csv(csv_file, index=False)Any help would be greatly appreciated.
Edit
Thanks to user Mauricio Arias Olave, I got the code working to do exactly as I want.
Here is the completed code for anyone who is curious:
# CODE FROM HERE: https://github.com/analyticswithadam/Python/blob/main/Pull_all_Comments_and_Replies_for_YouTube_Playlists.ipynbfrom googleapiclient.discovery import buildimport pandas as pdimport getpassapi_key = "API Key Here"playlist_ids = ['Youtube playlist here']# Build the YouTube clientyoutube = 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='snippet,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']] for pl_item in playlist_response["items"]: all_videos.append({"Video_ID": pl_item['contentDetails']['videoId'], "Title" : pl_item['snippet']['title']}) next_page_token = playlist_response.get('nextPageToken') if next_page_token is None: break return all_videos# Fetch all video IDs from the specified playlistsvideo_ids = get_all_video_ids_from_playlists(youtube, playlist_ids)# Function to get replies for a specific commentdef 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 videodef 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["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["Video_ID"], # Directly using video_id from function parameter'Title': video_id["Title"], # The title of the video.'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 videosall_comments = []for video_id in video_ids: video_comments = get_comments_for_video(youtube, video_id) all_comments.extend(video_comments)# Create DataFramecomments_df = pd.DataFrame(all_comments)# Export whole dataset to the local machine as CSV Filecsv_file = 'comments_data.csv' # Name your filecomments_df.to_csv(csv_file, index=False)