hiffi-artist-intelligence/core/sales_ranking_engine.py

"""
core/sales_ranking_engine.py
Calculates dedicated commercial sales rank across all records in commercial_sales.
Completely isolated from 500-cohort telemetry XP ranking logic.
"""

import sqlite3
import time

DB_PATH = "data/hiffi_catalog.db"

def recalculate_17k_sales_ranks():
    conn = sqlite3.connect(DB_PATH)
    cur = conn.cursor()

    print("[*] Calculating global commercial sales rankings...")
    start_time = time.time()

    # Join with telemetry_metrics to cleanly break ties for 0-sales artists
    rank_query = """
        SELECT 
            cs.sales_id,
            cs.artist_id,
            cs.audited_sales_units,
            COALESCE(tm.total_views, 0) as total_views,
            COALESCE(tm.spotify_monthly_listeners, 0) as listeners
        FROM commercial_sales cs
        LEFT JOIN telemetry_metrics tm ON cs.artist_id = tm.artist_id
        ORDER BY 
            cs.audited_sales_units DESC,
            total_views DESC,
            listeners DESC,
            cs.artist_id ASC
    """
    rows = cur.execute(rank_query).fetchall()
    total = len(rows)

    updates = []
    for rank, row in enumerate(rows, start=1):
        sales_id = row[0]
        updates.append((rank, sales_id))

    cur.executemany("UPDATE commercial_sales SET sales_rank = ? WHERE sales_id = ?", updates)
    conn.commit()
    conn.close()

    elapsed = round(time.time() - start_time, 2)
    print(f"✓ Recalculated sales ranks for {total} artists in {elapsed}s.")

if __name__ == "__main__":
    recalculate_17k_sales_ranks()