# Working with Query Runner
# What is QueryRunner
Each new QueryRunner
instance takes a single connection from connection pool, if RDBMS supports connection pooling.
For databases not supporting connection pools, it uses the same connection across data source.
# Creating a new QueryRunner
instance
Use createQueryRunner
method to create a new QueryRunner
:
const queryRunner = dataSource.createQueryRunner()
# Using QueryRunner
After you create a new instance of QueryRunner
use connect
method to actually obtain a connection from the connection pool:
const queryRunner = dataSource.createQueryRunner()
await queryRunner.connect()
2
Important: make sure to release it when it is not needed anymore to make it available to the connection pool again:
await queryRunner.release()
After connection is released it is not possible to use the query runner methods.
QueryRunner
has bunch of methods you can use, it also has it's own EntityManager
instance,
which you can use through manager
property in order to run EntityManager
methods on a particular database connection
used by QueryRunner
instance:
const queryRunner = connection.createQueryRunner()
// take a connection from the connection pool
await queryRunner.connect()
// use this particular connection to execut queries
const users = await queryRunner.manager.find(User)
// don't forget to release connection after you are done using it
await queryRunner.release()
2
3
4
5
6
7
8
9
10