Manage Azure Data Lake Analytics using Azure SDK for Node.js

This article describes how to manage Azure Data Lake Analytics accounts, data sources, users, and jobs using an app written using the Azure SDK for Node.js.

The following versions are supported:

  • Node.js version: 0.10.0 or higher
  • REST API version for Account: 2015-10-01-preview
  • REST API version for Catalog: 2015-10-01-preview
  • REST API version for Job: 2016-03-20-preview

Features

  • Account management: create, get, list, update, and delete.
  • Job management: submit, get, list, and cancel.
  • Catalog management: get and list.

How to Install

npm install azure-arm-datalake-analytics

Authenticate using Azure Active Directory

var msrestAzure = require('ms-rest-azure');
//user authentication
var credentials = new msRestAzure.UserTokenCredentials('your-client-id', 'your-domain', 'your-username', 'your-password', 'your-redirect-uri');
//service principal authentication
var credentials = new msRestAzure.ApplicationTokenCredentials('your-client-id', 'your-domain', 'your-secret');

Create the Data Lake Analytics client

var adlaManagement = require("azure-arm-datalake-analytics");
var accountClient = new adlaManagement.DataLakeAnalyticsAccountClient(credentials, 'your-subscription-id');
var jobClient = new adlaManagement.DataLakeAnalyticsJobClient(credentials, 'azuredatalakeanalytics.net');
var catalogClient = new adlaManagement.DataLakeAnalyticsCatalogClient(credentials, 'azuredatalakeanalytics.net');

Create a Data Lake Analytics account

var util = require('util');
var resourceGroupName = 'testrg';
var accountName = 'testadlaacct';
var location = 'eastus2';

// A Data Lake Store account must already have been created to create
// a Data Lake Analytics account. See the Data Lake Store readme for
// information on doing so. For now, we assume one exists already.
var datalakeStoreAccountName = 'existingadlsaccount';

// account object to create
var accountToCreate = {
  tags: {
    testtag1: 'testvalue1',
    testtag2: 'testvalue2'
  },
  name: accountName,
  location: location,
  properties: {
    defaultDataLakeStoreAccount: datalakeStoreAccountName,
    dataLakeStoreAccounts: [
      {
        name: datalakeStoreAccountName
      }
    ]
  }
};

client.account.create(resourceGroupName, accountName, accountToCreate, function (err, result, request, response) {
  if (err) {
    console.log(err);
    /*err has reference to the actual request and response, so you can see what was sent and received on the wire.
      The structure of err looks like this:
      err: {
        code: 'Error Code',
        message: 'Error Message',
        body: 'The response body if any',
        request: reference to a stripped version of http request
        response: reference to a stripped version of the response
      }
    */
  } else {
    console.log('result is: ' + util.inspect(result, {depth: null}));
  }
});

Get a list of jobs

var util = require('util');
var accountName = 'testadlaacct';
jobClient.job.list(accountName, function (err, result, request, response) {
  if (err) {
    console.log(err);
  } else {
    console.log('result is: ' + util.inspect(result, {depth: null}));
  }
});

Get a list of databases in the Data Lake Analytics Catalog

var util = require('util');
var accountName = 'testadlaacct';
catalogClient.catalog.listDatabases(accountName, function (err, result, request, response) {
  if (err) {
    console.log(err);
  } else {
    console.log('result is: ' + util.inspect(result, {depth: null}));
  }
});

See also