import sqlite3
import pandas as pd
import re

conn = sqlite3.connect('data/hiffi_catalog.db')
query = """
    SELECT 
        a.artist_id, a.artist_name, 
        a.youtube_url, a.spotify_url, a.instagram_url,
        t.subscribers_count, t.spotify_monthly_listeners, t.instagram_followers
    FROM artists a
    JOIN telemetry_metrics t ON a.artist_id = t.artist_id
    WHERE a.image_verified_flag = 1
"""
df = pd.read_sql_query(query, conn)
conn.close()

# Pattern checks
yt_search = df[df['youtube_url'].astype(str).str.contains('results\\?search_query', case=False, na=False)]
yt_invalid = df[~df['youtube_url'].astype(str).str.contains(r'youtube\.com/(channel/|c/|user/|@)', regex=True, na=False)]
sp_invalid = df[~df['spotify_url'].astype(str).str.contains(r'open\.spotify\.com/artist/[a-zA-Z0-9]{22}', regex=True, na=False)]
ig_invalid = df[~df['instagram_url'].astype(str).str.contains(r'instagram\.com/[a-zA-Z0-9_\.]+', regex=True, na=False)]

print('=== TOTAL COHORT PLATFORM URL HEALTH (500 ARTISTS) ===')
print(f'YouTube Search Fallback URLs: {len(yt_search)}')
print(f'YouTube Non-Canonical URLs:   {len(yt_invalid)}')
print(f'Spotify Malformed URLs:       {len(sp_invalid)}')
print(f'Instagram Malformed URLs:     {len(ig_invalid)}')

if len(yt_search) > 0:
    print('\nArtists with YouTube Search Fallbacks:')
    for _, r in yt_search.iterrows():
        print(f"  - {r['artist_name']} (ID: {r['artist_id']}) -> {r['youtube_url']}")

if len(sp_invalid) > 0:
    print('\nArtists with Malformed Spotify URLs:')
    for _, r in sp_invalid.iterrows():
        print(f"  - {r['artist_name']} (ID: {r['artist_id']}) -> {r['spotify_url']}")

if len(ig_invalid) > 0:
    print('\nArtists with Malformed Instagram URLs:')
    for _, r in ig_invalid.iterrows():
        print(f"  - {r['artist_name']} (ID: {r['artist_id']}) -> {r['instagram_url']}")