import os
import sqlite3
import pandas as pd
CSV_PATH = 'data/hiffi_multiplatform_500.csv'
DB_PATH = 'data/hiffi_catalog.db'
HTML_OUT = 'audit_top500_profiles.html'
df = pd.read_csv(CSV_PATH)
conn = sqlite3.connect(DB_PATH)
conn.row_factory = sqlite3.Row
cur = conn.cursor()
db_meta = {
str(r['artist_id']): dict(r)
for r in cur.execute('SELECT artist_id, country_origin FROM artists').fetchall()
}
missing_origin = []
cards_html = []
for idx, row in df.iterrows():
aid = str(row.get('artist_id'))
name = row.get('name') or row.get('artist_name')
origin = row.get('country_origin') or db_meta.get(aid, {}).get('country_origin') or 'Unknown'
if origin == 'Unknown' or pd.isna(origin):
missing_origin.append((aid, name))
img = row.get('image_url')
img_name = os.path.basename(str(img).strip())
rel_img_path = f'data/artist_avatars/{img_name}'
yt = row.get('youtube_url') if pd.notna(row.get('youtube_url')) else '#'
sp = row.get('spotify_url') if pd.notna(row.get('spotify_url')) else '#'
ig = row.get('instagram_url') if pd.notna(row.get('instagram_url')) else '#'
card = f"""
#{idx+1} {name}
{origin}
"""
cards_html.append(card)
full_html = f"""
HIFFI Top 500 - Visual Portrait & Metadata Audit
HIFFI Top 500 - Visual Portrait & Metadata Audit
Review real human face presence, 1:1 square framing, and canonical platform routing.
{''.join(cards_html)}
"""
with open(HTML_OUT, 'w', encoding='utf-8') as f:
f.write(full_html)
print('=== METADATA COMPLETENESS AUDIT ===')
print(f'Total cohort evaluated: {len(df)}')
print(f'Missing Country/City Origin: {len(missing_origin)}')
if missing_origin:
print('Sample missing origins:', missing_origin[:10])
print(f'\nVisual gallery successfully generated at: {HTML_OUT}')