| import csv |
| from googleapiclient.discovery import build |
| from urllib.parse import urlparse, parse_qs |
| from dotenv import load_dotenv |
| import os |
|
|
| load_dotenv() |
| key = os.getenv('YOUTUBE_API_KEY') |
| youtube = build('youtube','v3',developerKey=key) |
|
|
| vid_ids= ['0IZS7ucM9qQ','znefLMNyVE8','nxUcxHMEUGg','QedpcsA-GOU','BKOVzHcjEIo','NHk7scrb_9I','5XVoRGhrhZk','edx_qdEBAf0','g0CWxEuN2VI','RXeLbs_ogDU','uQoDmiMoXHg','8z_YVi-NycI','8YP81X1Jur4','xvvZATxs30M','vpNC1qM_cCE','zzwRbKI2pn4',"lFe5OA6uN8g","C3_mYPlvJeE","dpIJHTIEKC0",'ADllavPbUsg','WshNJefWEP8','WhvuONO1INQ','TNPtBnM4koE','QE6tkqvYW4'] |
|
|
| csv_exists = os.path.exists('yt_comment_dataset_v2.csv') |
|
|
| with open('yt_comment_dataset.csv', 'a' if csv_exists else 'w') as dataset: |
| writer = csv.writer(dataset) |
|
|
| for i in range(len(vid_ids)): |
|
|
| if not csv_exists: |
| writer.writerow(['video_id', 'comment', 'likes']) |
|
|
| vid_id = vid_ids[i] |
| print(f'scraping {vid_id}....') |
|
|
| count = 0 |
| next_page_token = None |
| try: |
| while True: |
| |
| |
| req_comments = youtube.commentThreads().list( |
| part='snippet', |
| videoId=vid_id, |
| textFormat='plainText', |
| pageToken=next_page_token, |
| order='relevance' |
| ) |
|
|
| res_comments = req_comments.execute() |
|
|
| for item in res_comments['items']: |
| comment = item['snippet']['topLevelComment']['snippet']['textDisplay'] |
| comment = comment.replace('\n','') |
| likes = int(item['snippet']['topLevelComment']['snippet']['likeCount']) |
| writer.writerow([vid_id, comment, likes]) |
| count+=1 |
|
|
|
|
| next_page_token = res_comments.get('nextPageToken') |
|
|
| if not next_page_token: |
| break |
|
|
| if count >= 1000: |
| break |
| |
| except Exception as e: |
| print(f"Error encountered at id {vid_id} : {e}") |
| continue |
|
|
| print('================DONE==================') |
