Python script to download songs from SoundCloud

soundcloud_download.py

import os
import sys
import requests
import bs4
import subprocess
import base64
from pathvalidate import sanitize_filename

# Global proxy setting
PROXY = "http://localhost:18080"
PROXIES = {
    'http': PROXY,
    'https': PROXY,
}

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

def get_audio_codec(filename):
    command = ["ffprobe", "-v", "error", "-select_streams", "a:0", "-show_entries", "stream=codec_name", "-of", "default=noprint_wrappers=1:nokey=1", filename]
    try:
        codec = subprocess.check_output(command, stderr=subprocess.STDOUT).decode().strip()
        return codec
    except subprocess.CalledProcessError as e:
        print(f"Error getting codec for {filename}: {e.output.decode()}")
        return None

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}")
    try:
        page = requests.get(url, proxies=PROXIES, timeout=10)
        page.raise_for_status()
    except requests.exceptions.RequestException as e:
        print(f"Error fetching page {url}: {e}")
        return

    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
    base_filename = sanitize_filename(f"{soundtrack_name} [{soundtrack_id}]").replace('\'', '')

    # Define a temporary filename for the downloaded audio
    temp_download_filename = f"{base_filename}.tmp"

    artwork_img_tag = soup.find('meta', {'property': 'twitter:image'})
    artwork_filename = 'default_artwork.jpg'
    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):
                try:
                    artwork_img = requests.get(artwork_img_url, proxies=PROXIES, timeout=10)
                    artwork_img.raise_for_status()
                    with open(artwork_filename, 'wb') as f:
                        f.write(artwork_img.content)
                except requests.exceptions.RequestException as e:
                    print(f"Error fetching artwork from {artwork_img_url}: {e}")
                    artwork_filename = 'default_artwork.jpg'

    if not os.path.exists(artwork_filename):
        create_default_artwork_image(artwork_filename)

    # Use a specific format code to get the best quality AAC or MP3 and ensure the file extension is correct.
    # The --audio-format mp3 ensures that yt-dlp will convert to mp3 if the source is not.
    yt_dlp_command = [
        "yt-dlp",
        "--proxy", PROXY,
        "-f", "bestaudio/best",
        "--audio-format", "mp3",  # Force conversion to mp3
        "--extract-audio",        # Extract audio only
        url,
        "-o", f"{base_filename}.%(ext)s",
    ]

    print(f"Downloading and converting audio with yt-dlp...")
    try:
        subprocess.run(yt_dlp_command, check=True, timeout=300)
    except subprocess.CalledProcessError as e:
        print(f"yt-dlp command failed with exit code {e.returncode}: {e}")
        return
    except subprocess.TimeoutExpired:
        print("yt-dlp command timed out.")
        return
    except FileNotFoundError:
        print("yt-dlp command not found. Is it installed and in your PATH?")
        return

    # Check for the downloaded file with a .mp3 extension, as yt-dlp should have created it
    soundtrack_filename = f"{base_filename}.mp3"
    if not os.path.exists(soundtrack_filename):
        print(f"yt-dlp did not produce the expected MP3 file: {soundtrack_filename}")
        return

    out_filename = f"{base_filename}_1.mp3"

    ffmpeg_embed_command = [
        "ffmpeg",
        "-i", soundtrack_filename,
        "-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("Embedding artwork...")
    try:
        subprocess.run(ffmpeg_embed_command, check=True)
    except subprocess.CalledProcessError as e:
        print(f"ffmpeg embedding failed with exit code {e.returncode}: {e}")
        return

    # Cleanup temporary files
    os.remove(soundtrack_filename)
    if artwork_filename != 'default_artwork.jpg':
        os.remove(artwork_filename)

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

if __name__ == "__main__":
    urls = load_urls('extracted_links.txt')
    for url in urls:
        if not url or not url.startswith('http'):
            print(f'Incorrect or empty URL: {url}')
            continue
        try:
            download(url)
        except KeyboardInterrupt:
            print("\nDownload process interrupted by user.")
            sys.exit(0)
        except Exception as e:
            print(f"An unexpected error occurred for {url}: {e}")