Database-agnostic connection pooling, querying, and result sets

  • any-db

Downloads in past


2.2.16 years ago11 years agoMinified + gzip package size for any-db in KB


Build Status
The less-opinionated Node.js database abstraction layer


Establish a connection:
// Takes an optional callback
var conn = anyDB.createConnection('driver://user:pass@hostname/database')

Make queries:
var sql = 'SELECT * FROM questions'

// query() returns a readable stream
conn.query(sql).on('data', function (row) {})

// pass a callback to collect results
conn.query(sql, function (error, result) {})

Use bound parameters:
sql += ' WHERE answer = ?'
conn.query(sql, [42], function (err, res) {})

Manage database transactions with any-db-transactionTransaction:
var begin = require('any-db-transaction')

var tx = begin(conn)              // Can also take a callback
tx.on('error', function (err) {}) // Emitted for unhandled query errors
tx.query(...)                     // same interface as connections, plus...
tx.rollback()                     // this too
tx.commit()                       // takes an optional callback for errors

Create a connection poolConnectionPool that maintains 2-20 connections:
var pool = anyDB.createPool(dbURL, {min: 2, max: 20})
pool.query(...)       // perform a single query, same API as connection
var tx = begin(pool)  // create a transaction with the first available connection
pool.close()          // close the pool (call when your app should exit)


The purpose of this library is to provide a consistent API for the commonly used functionality of SQL database drivers, while avoiding altering driver behaviour as much as possible.


For Applications

npm install --save any-db-{postgres,mysql,sqlite3,mssql}
All of the adapter libraries have any-db as a peerDependency, which means that require('any-db') will work even though you don't install it directly or add it to your package.json.

For Libraries

Add any-db to peerDependencies in package.json. This allows users of your library to satisfy the any-db dependency by installing the adapter of their choice.


module.exports := {
  createConnection: (Url, Continuation<Connection>?) => Connection
  createPool: (Url, PoolConfig) => ConnectionPool

Url := String | { adapter: String }

PoolConfig := {
  min: Number,
  max: Number,
  onConnect: (Connection, ((Error) => void) => void
  reset: (Connection, ((Error) => void) => void

Continuation := (Maybe<Error>, Any) => void

The API of Connection and Query objects is fully described in the adapter-spec, while Transaction and ConnectionPool objects have their own documentation. Connections, transactions and pools all have a query method that behaves consistently between drivers.
Both exported functions require an Url as their first parameter. This can either be a string of the form adapter://user:password@host/database (which will be parsed by parse-db-url) or an object. When an object is used, it must have an adapter property, and any other properties required by the specified adapters createConnection method.
See also: README for your chosen adapter (MS SQL, MySQL, Postgres, and SQLite3)