A Seneca.js data storage plugin

seneca-postgres-store

| | This open source module is sponsored and supported by Voxgig. | |---|---|

Description

seneca-postgres-store is a PostgreSQL database plugin for the Seneca MVP toolkit. The plugin is using the node-postgres driver. For query generation it uses internally the seneca-standard-query plugin and the standard functionality can be extended by using the seneca-store-query plugin.

Usage:

var Seneca = require('seneca');
var store = require('seneca-postgres-store');

var DBConfig = {
  name: 'senecatest',
  host: 'localhost',
  username: 'senecatest',
  password: 'senecatest',
  port: 5432
}
...

var si = Seneca(DBConfig)
si.use(require('seneca-postgres-store'), DBConfig)
si.ready(function() {
  var product = si.make('product')
  ...
})
...

Seneca compatibility

Supports Seneca versions 1.x - 3.x

Supported functionality

All Seneca data store supported functionality is implemented in seneca-store-test as a test suite. The tests represent the store functionality specifications.

Usage

You don't use this module directly. It provides an underlying data storage engine for the Seneca entity API:

var entity = seneca.make$('typename')
entity.someproperty = "something"
entity.anotherproperty = 100

entity.save$(function (err, entity) { ... })
entity.load$({id: ...}, function (err, entity) { ... })
entity.list$({property: ...}, function (err, entity) { ... })
entity.remove$({id: ...}, function (err, entity) { ... })

Query Support

The standard Seneca query format is supported. See the seneca-standard-query plugin for more details.

Extended Query Support

By using the seneca-store-query plugin its query capabilities can be extended. See the plugin page for more details.

Column name transformation, backward compatibility

In seneca-postgres-store 2.0 the internal CamelCase to snake_case column names conversion was removed.

To update from seneca-postgres-store 1.x to 2.x on systems built with seneca-postgres-store 1.x you must provide to the plugin through its options the functions that do the CamelCase to snake_case conversion and back. Any other name transformations to and from database column name can be also made with these. Example:

var DefaultConfig = {
...
  fromColumnName: function (attr) {
    // apply some conversion on column names
    return attr.toUpperCase()
  },
  toColumnName: function (attr) {
    // convert back column names
    return attr.toLowerCase()
  }
}
seneca.use(require('seneca-postgres-store'), DefaultConfig)

For a fully functional CamelCase to snake_case implementation sample please look in the postgres.test.js at the 'Column Names conversions' test code.

Limits

By default queries are limited to 20 values. This can be bypassed by passing the nolimit option, which if set to true will not limit any queries.

Fields

To filter the fields returned from the list operation, pass a fields$ array of column names to return. If no fields$ are passed, all fields are returned (i.e. select * is used). e.g.

query.fields$ = ['id', 'name']

Note: The implicit id that is generated on save$ has uuid value. To override this you must provide entity.id$ with a desired value.

Custom ID generator

To generate custom IDs it is exposed a seneca action pattern hook that can be overwritten:

seneca.add({role: 'sql', hook: 'generate_id', target: <store name>}, function (args, done) {
  return done(null, {id: idPrefix + Uuid()})
})

Native Driver

As with all seneca stores, you can access the native driver, in this case, the pg connection object using entity.native$(function (err, connectionPool, release) {...}). Please make sure that you release the connection after using it.

entity.native$( function (err, client, releaseConnection){
  // ... you can use client
  // ... then release connection
  releaseConnection()
} )

Contributing

The Senecajs org encourages open participation. If you feel you can help in any way, be it with documentation, examples, extra testing, or new features please get in touch.

To run tests with Docker

Build the PostgreSQL Docker image:

npm run build

Start the PostgreSQL container:

npm run start

Stop the PostgreSQL container:

npm run stop

While the container is running you can run the tests into another terminal:

npm run test

Testing for Mac users

Before the tests can be run you must run docker-machine env default and copy the docker host address (example: '192.168.99.100'). This address must be inserted into the test/default_config.json file as the value for the host variable. The tests can now be run.

License

Copyright (c) 2012 - 2016, Marian Radulescu and other contributors. Licensed under MIT.