scripts/build_rxnorm_db.py

#!/usr/bin/env python3
import sqlite3, csv, argparse
from metaphone import doublemetaphone
import sys, os
sys.path.append(os.getcwd())
from config import DB_PATH, RXNORM_CSV

def build_db(db_path: str, csv_path: str):
    conn = sqlite3.connect(db_path)
    c = conn.cursor()
    c.execute("DROP TABLE IF EXISTS drugs")
    c.execute("DROP TABLE IF EXISTS metaphone")
    c.execute("CREATE TABLE drugs(name TEXT PRIMARY KEY, cui TEXT)")
    c.execute("CREATE TABLE metaphone(meta TEXT, name TEXT, cui TEXT)")

    with open(csv_path, newline='', encoding='utf-8') as f:
        reader = csv.DictReader(f)
        for row in reader:
            nm, cui = row['name'], row['cui']
            c.execute("INSERT OR IGNORE INTO drugs VALUES(?,?)", (nm, cui))
            meta = doublemetaphone(nm)[0]
            c.execute("INSERT INTO metaphone VALUES(?,?,?)", (meta, nm, cui))
    conn.commit()
    conn.close()

if __name__ == "__main__":
    build_db(DB_PATH, RXNORM_CSV)
    print(f"✅ RxNorm DB built at {DB_PATH}")