app/services/video_processing/scrape_hf.py

from bs4 import BeautifulSoup
import requests

def extract_video_urls(url):
    # Récupérer le contenu de la page
    response = requests.get(url)
    html_content = response.text
    
    # Créer un objet BeautifulSoup
    soup = BeautifulSoup(html_content, 'html.parser')
    
    # Trouver toutes les balises source
    video_sources = soup.find_all('source')
    
    # Extraire les URLs
    video_urls = []
    for source in video_sources:
        url = source.get('src')
        if url:
            # Nettoyer l'URL en retirant le #t=0.001
            clean_url = url.split('#')[0]
            video_urls.append(clean_url)
    
    return video_urls