#!/usr/bin/env python3
"""One-time script to find duplicate partners by CUI (bare number, ignoring RO prefix)."""
import sys, os, csv
sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..'))

# Setup Oracle env same as start.sh
from api.app import database


def scan_duplicates():
    database.init_oracle()
    conn = database.get_oracle_connection()
    try:
        with conn.cursor() as cur:
            cur.execute("""
                SELECT bare_cui, COUNT(*) as cnt,
                       LISTAGG(id_part||':'||denumire, ', ') WITHIN GROUP (ORDER BY id_part) as partners
                FROM (SELECT id_part, denumire,
                      TRIM(REGEXP_REPLACE(UPPER(TRIM(cod_fiscal)), '^RO\\s*', '')) as bare_cui
                      FROM nom_parteneri WHERE NVL(sters,0)=0
                        AND cod_fiscal IS NOT NULL AND LENGTH(TRIM(cod_fiscal)) >= 3)
                GROUP BY bare_cui HAVING COUNT(*) > 1
                ORDER BY cnt DESC
            """)
            rows = cur.fetchall()
    finally:
        database.pool.release(conn)
        database.close_oracle()

    # Output markdown + CSV
    print(f"\n## Duplicate Partners Report\n")
    print(f"Found {len(rows)} CUIs with duplicate partners.\n")
    print("| CUI | Count | Partners |")
    print("|-----|-------|----------|")
    for row in rows:
        print(f"| {row[0]} | {row[1]} | {row[2][:100]} |")

    # CSV output
    csv_path = os.path.join(os.path.dirname(__file__), 'duplicate_partners.csv')
    with open(csv_path, 'w', newline='') as f:
        writer = csv.writer(f)
        writer.writerow(['bare_cui', 'count', 'partners'])
        for row in rows:
            writer.writerow(row)
    print(f"\nCSV saved: {csv_path}")


if __name__ == '__main__':
    scan_duplicates()