spotify_resolver.py

import re
import requests

def resolve_spotify_track(url: str) -> dict:
    """
    Scrapes a public Spotify track page to extract the song title and artist.
    Returns: {"title": str, "artist": str} or None if resolution fails.
    """
    # Regex to extract track ID
    match = re.search(r"track/([a-zA-Z0-9]+)", url)
    if not match:
        return None
    
    headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
    }
    
    try:
        response = requests.get(url, headers=headers, timeout=10)
        if response.status_code != 200:
            return None
        
        html = response.text
        
        # Try og:title and og:description or title tag
        # og:title typically has the song title
        title_match = re.search(r'<meta\s+property="og:title"\s+content="([^"]+)"', html)
        # og:description or twitter:description usually has "Song · Artist · Year" or "Artist · Song"
        desc_match = re.search(r'<meta\s+(?:property|name)="(?:og|twitter):description"\s+content="([^"]+)"', html)
        
        title = None
        artist = None
        
        if title_match:
            title = title_match.group(1).strip()
            
        if desc_match:
            desc = desc_match.group(1).strip()
            # If desc looks like "Song · Artist · Year" or similar
            parts = [p.strip() for p in desc.split("·")]
            if len(parts) >= 2:
                # Often: "Song · Artist · Year" or "Artist · Song · [etc]"
                # Let's verify which one is the title.
                # If the first part matches the og:title, then the second part is the artist!
                if title and parts[0].lower() == title.lower():
                    artist = parts[1]
                elif title and parts[1].lower() == title.lower():
                    artist = parts[0]
                else:
                    artist = parts[1]
            else:
                # Fallback: check if description says "Listen to [Song] on Spotify. [Artist] · Song · [Year]"
                listen_match = re.search(r"Listen to\s+([^\s]+)\s+on Spotify\.\s+([^\·]+)", desc)
                if listen_match:
                    artist = listen_match.group(2).strip()
        
        # If still not found, try a simple title parse
        if not title:
            html_title = re.search(r"<title>([^<]+)</title>", html)
            if html_title:
                t_content = html_title.group(1)
                # Usually: "Song Name - song and artist by Artist | Spotify"
                t_parts = t_content.split(" | ")
                if len(t_parts) > 0:
                    title = t_parts[0].replace(" - song and artist by ", " - ")
        
        # Clean up common spotify suffix / prefix
        if title:
            title = re.sub(r"\s-\s.*$", "", title) # remove everything after " - "
        
        if not title:
            return None
            
        return {
            "title": title,
            "artist": artist or "Unknown Artist"
        }
    except Exception as e:
        print(f"Error resolving Spotify metadata: {e}")
        return None

# Simple testing hook
if __name__ == "__main__":
    test_url = "https://open.spotify.com/track/4PTG3Z6ehGkBF3zI7YSpA0"
    print(resolve_spotify_track(test_url))