API reference
Public exports and the connection methods in react-native-nitro-sqlite.
open() is the main entry point. It opens a named database and returns a connection whose methods no longer need the database name.
import { open } from 'react-native-nitro-sqlite'
const db = open({ name: 'app.sqlite' })
const { rows } = await db.executeAsync<{ id: number; title: string }>(
'SELECT id, title FROM notes WHERE id = ?',
[1],
)
console.log(rows.item(0))
db.close()| Task | Connection method | Reference |
|---|---|---|
| Open, close, delete, attach, or detach a database | open, close, delete, attach, detach | Connections |
| Run SQL and read rows or column metadata | execute, executeAsync | Queries and results |
| Group statements | transaction, executeBatch, executeBatchAsync | Transactions and batches |
| Import a SQL file | loadFile, loadFileAsync | Connections |
| Look up exported types and error behavior | QueryResult, NitroSQLiteError, and other types | Types and errors |
| Call the underlying database-name-based hybrid object | NitroSQLite.native | Native API |
The package also exports NitroSQLite, NitroSQLiteError, typeORMDriver, and the connection, query, batch, and transaction types. NitroSQLite.open(options) and the named open(options) export are the same connection helper. The other top-level NitroSQLite methods are listed on the global and native API page. The typeORMDriver export is for TypeORM; vector helpers come from the separate sqlite-vec package.
Synchronous methods run on the JavaScript thread. Their async counterparts run the native work off that thread. Calls made through one opened connection are coordinated by a per-database JavaScript queue: async calls run in submission order, and a conflicting synchronous call throws a busy error.