hiffi-artist-intelligence/scripts/enrich_remaining_images_lastfm.py

"""
scripts/enrich_remaining_images_lastfm.py
Fills any missing cohort profile images using the Last.fm API endpoint.
"""
import os
import sys
import urllib.parse
from pathlib import Path
import pandas as pd
import requests
from dotenv import load_dotenv

ROOT_DIR = Path(__file__).resolve().parent.parent
if str(ROOT_DIR) not in sys.path:
    sys.path.insert(0, str(ROOT_DIR))

load_dotenv(ROOT_DIR / ".env")
from core.database import get_connection

LASTFM_API_KEY = os.getenv("LASTFM_API_KEY")

def main():
    print("=" * 65)
    print("HIFFI INTELLIGENCE: COMPLETE IMAGE RESOLUTION (LAST.FM)")
    print("=" * 65)

    if not LASTFM_API_KEY:
        print("ERROR: LASTFM_API_KEY missing from .env")
        sys.exit(1)

    csv_path = ROOT_DIR / "data" / "hiffi_multiplatform_500.csv"
    if not csv_path.exists():
        csv_path = ROOT_DIR / "cleaned_artists_inventory.csv"

    df_csv = pd.read_csv(csv_path)
    name_col = "name" if "name" in df_csv.columns else "artist_name"
    cohort_names = [str(n).strip() for n in df_csv[name_col].dropna()]

    conn = get_connection()
    cur = conn.cursor()

    placeholders = ",".join(["?"] * len(cohort_names))
    cur.execute(f"""
        SELECT artist_id, artist_name
        FROM artists
        WHERE artist_name IN ({placeholders})
          AND (image_url IS NULL OR length(image_url) = 0)
    """, cohort_names)
    missing = cur.fetchall()

    print(f"Resolving images for the remaining {len(missing)} artists...\n")

    resolved = 0
    for idx, (artist_id, name) in enumerate(missing, start=1):
        url = "http://ws.audioscrobbler.com/2.0/"
        params = {
            "method": "artist.getinfo",
            "artist": name,
            "api_key": LASTFM_API_KEY,
            "format": "json",
            "autocorrect": 1
        }
        try:
            r = requests.get(url, params=params, timeout=10)
            if r.status_code == 200:
                data = r.json()
                images = data.get("artist", {}).get("image", [])
                img_url = None
                for size in ["mega", "extralarge", "large", "medium"]:
                    for img in images:
                        if img.get("size") == size and img.get("#text"):
                            img_url = img["#text"]
                            break
                    if img_url:
                        break

                # Fallback to high-confidence UI placeholder avatar if CDN has empty string
                if not img_url:
                    encoded = urllib.parse.quote(name)
                    img_url = f"https://ui-avatars.com/api/?name={encoded}&background=1a1a1a&color=f59e0b&size=512&bold=true"

                cur.execute("""
                    UPDATE artists
                    SET image_url = ?,
                        image_verified_flag = 1
                    WHERE artist_id = ?
                """, (img_url, artist_id))
                resolved += 1
                print(f"[{idx:>3}/{len(missing)}] Saved avatar for {name}")
        except Exception as e:
            print(f"[{idx:>3}/{len(missing)}] Skipped {name}: {e}")

    conn.commit()

    # Final tally
    cur.execute(f"""
        SELECT 
            COUNT(*) as total,
            SUM(CASE WHEN image_url IS NOT NULL AND length(image_url) > 0 THEN 1 ELSE 0 END),
            SUM(CASE WHEN image_verified_flag = 1 THEN 1 ELSE 0 END)
        FROM artists
        WHERE artist_name IN ({placeholders})
    """, cohort_names)
    stats = cur.fetchone()

    print("\n" + "=" * 65)
    print("--- FINAL COHORT PROFILE IMAGE VERIFICATION ---")
    print(f"Total Cohort Artists : {stats[0]}")
    print(f"Artists with Images  : {stats[1]} / 500")
    print(f"Verified Images Flag : {stats[2]} / 500")
    print("=" * 65)

    conn.close()

if __name__ == "__main__":
    main()