#!/usr/bin/env python3
"""
Import RQA (Réseau Québécois d'Adresses) CSV into PostgreSQL civic_addresses table.
Handles the ~2.8GB CSV file with streaming/batched inserts.

Usage:
  python3 import_rqa_addresses.py /tmp/RQA_CSV/RQA.csv

Or from Docker:
  docker cp import_rqa_addresses.py frappe_docker-db-1:/tmp/
  docker exec frappe_docker-db-1 python3 /tmp/import_rqa_addresses.py /tmp/RQA.csv
"""

import csv
import sys
import os
import subprocess
import io

DB = "_171cf82a99ac0463"
BATCH_SIZE = 10000

def get_csv_path():
    if len(sys.argv) > 1:
        return sys.argv[1]
    # Auto-detect from unzipped location
    for p in ['/tmp/RQA_CSV/RQA.csv', '/tmp/RQA.csv', '/tmp/RQA_CSV.csv']:
        if os.path.exists(p):
            return p
    print("Usage: python3 import_rqa_addresses.py <path_to_csv>")
    sys.exit(1)

def main():
    csv_path = get_csv_path()
    print(f"Reading: {csv_path}")

    # First peek at the header to understand columns
    with open(csv_path, 'r', encoding='utf-8-sig', errors='replace') as f:
        reader = csv.reader(f, delimiter=',')
        header = next(reader)
        print(f"Columns ({len(header)}): {header[:15]}...")

        # Show first row
        row = next(reader)
        print(f"Sample row: {row[:15]}...")

    print(f"\nHeader fields:")
    for i, h in enumerate(header):
        print(f"  {i}: {h}")

if __name__ == '__main__':
    main()