import sqlite3
from pathlib import Path
import pandas as pd

PROJECT_ROOT = Path(__file__).resolve().parent.parent
DB_PATH = PROJECT_ROOT / "data" / "hiffi_catalog.db"

conn = sqlite3.connect(DB_PATH)

# Pull full telemetry alongside platform URLs
query = """
    SELECT 
        a.artist_id,
        a.artist_name,
        a.youtube_url,
        t.subscribers_count AS yt_subscribers,
        a.spotify_url,
        t.spotify_monthly_listeners AS sp_monthly_listeners,
        a.instagram_url,
        t.instagram_followers AS ig_followers
    FROM artists a
    JOIN telemetry_metrics t ON a.artist_id = t.artist_id
    WHERE a.image_verified_flag = 1
    ORDER BY t.spotify_monthly_listeners DESC
"""

df = pd.read_sql_query(query, conn)
conn.close()

# 1. Verification Integrity Checks
zero_yt = df[df["yt_subscribers"].fillna(0) == 0]
zero_sp = df[df["sp_monthly_listeners"].fillna(0) == 0]
zero_ig = df[df["ig_followers"].fillna(0) == 0]

print("=== 500-ARTIST TELEMETRY PROOF AUDIT ===")
print(f"Total Cohort Verified Artists: {len(df)}")
print(f"Artists with YouTube Subscribers > 0:    {len(df) - len(zero_yt)} / {len(df)}")
print(f"Artists with Spotify Monthly Listeners > 0: {len(df) - len(zero_sp)} / {len(df)}")
print(f"Artists with Instagram Followers > 0:   {len(df) - len(zero_ig)} / {len(df)}")

if not zero_yt.empty:
    print(f"\n[FLAG] {len(zero_yt)} artists still have 0 or null YouTube subscribers:")
    print(zero_yt[["artist_name", "youtube_url", "yt_subscribers"]].head(10).to_string(index=False))

# Export complete proof ledger to CSV
output_path = PROJECT_ROOT / "data" / "platform_telemetry_proof_500.csv"
df.to_csv(output_path, index=False, encoding="utf-8")
print(f"\nSaved complete proof ledger to: {output_path}")

print("\n--- SAMPLE HIGH-REACH ENTRIES (TOP 5) ---")
print(df[["artist_name", "yt_subscribers", "sp_monthly_listeners", "ig_followers"]].head(5).to_string(index=False))

print("\n--- SAMPLE LOWEST-REACH ENTRIES (TAIL 5) ---")
print(df[["artist_name", "yt_subscribers", "sp_monthly_listeners", "ig_followers"]].tail(5).to_string(index=False))