Python script to download songs from SoundCloud

soundcloud_download.py

import os
import sys
import requests
import bs4
import subprocess
from pathvalidate import sanitize_filename

def load_urls(filename):
    with open(filename, 'r') as file:
        urls = file.read().strip().split('\n')
    return urls

def get_audio_codec(filename):
    command = f"ffprobe -v error -select_streams a:0 -show_entries stream=codec_name -of default=noprint_wrappers=1:nokey=1 '{filename}'"
    codec = subprocess.check_output(command, shell=True).decode().strip()
    return codec

def create_default_artwork_image(filename):
    # Base64 string for a black 512x512 JPEG image
    base64_image = (
        '''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'''
    )
    image_data = base64.b64decode(base64_image)
    with open(filename, 'wb') as f:
        f.write(image_data)

def download(url):
    print(f"Starting download for {url}")
    page = requests.get(url)
    soup = bs4.BeautifulSoup(page.content, 'html.parser')

    soundtrack_id = soup.find('meta', {'property': 'al:android:url'}).get('content').split(':')[-1]
    soundtrack_name = soup.find('a', {'itemprop': 'url'}).text
    soundtrack_filename = sanitize_filename(f'{soundtrack_name} [{soundtrack_id}].mp3').replace('\'', '')
    out_filename = f'{os.path.splitext(soundtrack_filename)[0]}_1.mp3'

    if os.path.exists(out_filename):
        print(f"File {out_filename} already exists. Skipping.")
        return

    artwork_img_tag = soup.find('meta', {'property': 'twitter:image'})
    if artwork_img_tag:
        artwork_img_url = artwork_img_tag.get('content')
        if artwork_img_url:
            artwork_filename = os.path.split(artwork_img_url)[-1]
            if not os.path.exists(artwork_filename):
                artwork_img = requests.get(artwork_img_url)
                open(artwork_filename, 'wb').write(artwork_img.content)
        else:
            artwork_filename = 'default_artwork.jpg'
    else:
        artwork_filename = 'default_artwork.jpg'

    # Check if default artwork file exists, if not create it
    if not os.path.exists(artwork_filename):
        create_default_artwork_image(artwork_filename)

    os.system(f'''yt-dlp -f 'bestvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best' {url} -o '{soundtrack_filename}' ''')

    codec = get_audio_codec(soundtrack_filename)

    if codec in ['opus', 'ogg']:
        temp_filename = f'converted_{soundtrack_filename}'
        os.system(f'''ffmpeg -i '{soundtrack_filename}' -codec:a libmp3lame -qscale:a 2 '{temp_filename}' ''')
        audio_file = temp_filename
    else:
        audio_file = soundtrack_filename

    os.system(f'''ffmpeg -i '{audio_file}' -i '{artwork_filename}' -map 0:0 -map 1:0 -c copy -id3v2_version 3 -metadata:s:v title='Album cover' -metadata:s:v comment='Cover (front)' '{out_filename}' ''')

    print(f"Finished download for {url}")

if __name__ == "__main__":
    urls = load_urls('extracted_links.txt')
    for url in urls:
        if url == '' or not url.startswith('http'):
            print(f'incorrect url: {url}')
            continue
        download(url)