Skip to content

LoggingQueryRunner

A general-purpose query runner that intercepts all queries and delegates their execution to another query runner while allowing you to log query details. Useful for debugging, performance monitoring, and auditing executed statements.

Supported databases

Tip

LoggingQueryRunner supports synchronous query execution. See the Synchronous query runners for more information.

Do not share connections between requests

A ts-sql-query connection object — along with the query runner instances passed to its constructor — represents a dedicated connection to the database.

Therefore, you must not share the same connection object between concurrent HTTP requests. Instead, create a new connection object for each request, along with its own query runners.

Even if the query runner internally uses a connection pool, the ts-sql-query connection still represents a single active connection, acquired from the pool. It must be treated as such and never reused across requests.

Usage Example

import { LoggingQueryRunner } from "ts-sql-query/queryRunners/LoggingQueryRunner";

async function main() {
    const connection = new DBConnection(new LoggingQueryRunner({
        onQuery(queryType, query, params) {
            console.log('onQuery', queryType, query, params, { startedAt })
        },
        onQueryResult(queryType, query, params, result) {
            console.log('onQueryResult', queryType, query, params, result, { startedAt, endedAt })
        },
        onQueryError(queryType, query, params, error) {
            console.log('onQueryError', queryType, query, params, error, { startedAt, endedAt })
        }
    }, otherQueryRunner));
    // Do your queries here
}

API Overview

The LoggingQueryRunner receives an object as first argument of the constructor that can define the following functions:

  • onQuery: Executed before the query.
  • onQueryResult: Executed after the successful execution of the query.
  • onQueryError: Executed after the query in case of error.

All these functions receive as argument:

  • type: QueryType: type of the query to be executed. The QueryType is defined as:
type QueryType = 'selectOneRow' | 'selectManyRows' | 'selectOneColumnOneRow' | 'selectOneColumnManyRows' |
    'insert' | 'insertReturningLastInsertedId' | 'insertReturningMultipleLastInsertedId' |
    'insertReturningOneRow' | 'insertReturningManyRows' | 'insertReturningOneColumnOneRow' | 'insertReturningOneColumnManyRows' |
    'update' | 'updateReturningOneRow' | 'updateReturningManyRows' | 'updateReturningOneColumnOneRow' | 'updateReturningOneColumnManyRows' |
    'delete' | 'deleteReturningOneRow' | 'deleteReturningManyRows' | 'deleteReturningOneColumnOneRow' | 'deleteReturningOneColumnManyRows' |
    'executeProcedure' | 'executeFunction' | 'beginTransaction' | 'commit' | 'rollback' | 
    'executeDatabaseSchemaModification' | 'executeConnectionConfiguration'
  • query: string: query required to be executed, empty in the case of beginTransaction, commit or rollback
  • params: any[]: parameters received by the query.
  • result: any: (only in onQueryResult) result of the execution of the query.
  • error: any: (only in onQueryError) error that happens executing the query.
  • startedAt: elapsed time value in nanoseconds before the query execution.
  • endedAt: (only in onQueryResult or onQueryError) elapsed time value in nanoseconds after the query execution.

Info

  • onQuery, onQueryResult, and onQueryError are optional; you can define only the methods that you need.
  • In case the provided query runner doesn't support low-level transaction management, fake beginTransaction, commit, and rollback operations will be emitted to allow you to see them in the log.