Integrate Real-Time Access to Databricks in SAPUI5 MVC Apps

In this article we show how to use the CData API Server and with the ADO.NET Provider for Databricks (or any of 200+ other ADO.NET Providers) to write SAPUI5 apps that leverage the capabilities of the Databricks API, without writing to a back-end database. The API Server is a lightweight Web application that runs on your server and produces OData feeds of Databricks data. OData is the standard for real-time data access over the Web and has built-in support in SAPUI5 and OpenUI5.

Getting Started

---

Set Up the API Server

Follow the steps below to begin producing secure Databricks OData services:

Deploy

The API Server runs on your own server. On Windows, you can deploy using the stand-alone server or IIS. On a Java servlet container, drop in the API Server WAR file. See the help documentation for more information and how-tos.

The API Server is also easy to deploy on Microsoft Azure, Amazon EC2, and Heroku.

Connect to Databricks

After you deploy the API Server and the ADO.NET Provider for Databricks, provide authentication values and other connection properties needed to connect to Databricks by clicking Settings -> Connections and adding a new connection in the API Server administration console.

To connect to a Databricks cluster, set the properties as described below.

Note: The needed values can be found in your Databricks instance by navigating to Clusters, and selecting the desired cluster, and selecting the JDBC/ODBC tab under Advanced Options.

• Server: Set to the Server Hostname of your Databricks cluster.
• HTTPPath: Set to the HTTP Path of your Databricks cluster.
• Token: Set to your personal access token (this value can be obtained by navigating to the User Settings page of your Databricks instance and selecting the Access Tokens tab).

You can then choose the Databricks entities you want to allow the API Server access to by clicking Settings -> Resources. This article uses Customers as an example.

Authorize API Server Users

After determining the OData services you want to produce, authorize users by clicking Settings -> Users. The API Server uses authtoken-based authentication and supports the major authentication schemes. You can authenticate as well as encrypt connections with SSL. Access can also be restricted by IP address; Access is restricted to only the local machine by default.

Create the View

In this article the user views and interacts with Databricks data through an SAPUI5 table control. Table columns will be automatically detected from the metadata retrieved from the API Server's API endpoint. We define the following table in a separate View.view.xml file:

<mvc:View
  controllerName="sap.ui.table.sample.OData2.Controller"
  xmlns="sap.ui.table"
  xmlns:mvc="sap.ui.core.mvc"
  xmlns:u="sap.ui.unified"
  xmlns:c="sap.ui.core"
  xmlns:m="sap.m">
  <m:Page
    showHeader="false"
    enableScrolling="false"
    class="sapUiContentPadding">
    <m:content>
      <Table
        id="table"
        selectionMode="MultiToggle"
        visibleRowCount="10"
        enableSelectAll="false"
        rows="{/Customers}"
        threshold="15"
        enableBusyIndicator="true"
        columns="{
          path: 'meta>/dataServices/schema/[${namespace}===\'CData\']/entityType/[${name}===\'Customers\']/property',
          factory: '.columnFactory'
        }">
        <toolbar>
          <m:Toolbar>
            <m:Title text="Databricks Customers"></m:Title>
          </m:Toolbar>
        </toolbar>
        <noData>
          <m:BusyIndicator class="sapUiMediumMargin"/>
        </noData>
      </Table>
    </m:content>
  </m:Page>
</mvc:View>

Create the Model and Controller

In SAPUI5, you do not need to write any OData queries; an ODataModel instance handles the application's data access commands. The API Server then translates the queries into Databricks API calls.

The controller processes user input and represents information to the user through a view. Define the controller in a new file, Controller.controller.js. Instantiate the model in the onInit function -- you will need to replace the placeholder values for the URL to the API Server, a user allowed to access the OData endpoint of the API Server, and the authtoken for the user.

sap.ui.define([
  "sap/ui/core/mvc/Controller",
  "sap/ui/model/odata/v2/ODataModel",
  "sap/ui/model/json/JSONModel",
  "sap/ui/table/Column",
  "sap/m/Text",
], function(Controller, ODataModel, JSONModel, Column, Text ) {
  "use strict";
  

  return Controller.extend("sap.ui.table.sample.OData2.Controller", {
    
    onInit : function () {
      
      var oView = this.getView();
      var oDataModel = new ODataModel("http://myserver/api.rsc/",{user: "MyUser", password: "MyAuthToken"});
      
      oDataModel.getMetaModel().loaded().then(function(){
        oView.setModel(oDataModel.getMetaModel(), "meta");
      });
      oView.setModel(oDataModel);
      
      var oTable = oView.byId("table");
      var oBinding = oTable.getBinding("rows");
      var oBusyIndicator = oTable.getNoData();
      oBinding.attachDataRequested(function(){
      oTable.setNoData(oBusyIndicator);
      });
      oBinding.attachDataReceived(function(){
        oTable.setNoData(null); //use default again ("no data" in case no data is available)
      });
    },
    
    onExit : function () {
    },
    
    columnFactory : function(sId, oContext) {
      var oModel = this.getView().getModel();
      var sName = oContext.getProperty("name");
      var sType = oContext.getProperty("type");
      var iLen = oContext.getProperty("maxLength");
      iLen = iLen ? parseInt(iLen, 10) : 10;
      
      return new Column(sId, {
        sortProperty: sName, 
        filterProperty: sName,
        width: (iLen > 9 ? (iLen > 50 ? 15 : 10) : 5) + "rem",
        label: new sap.m.Label({text: "{/#Customers/" + sName + "/@name}"}),
        hAlign: sType && sType.indexOf("Decimal") >= 0 ? "End" : "Begin",
        template: new Text({text: {path: sName}})
      });
    }
    
  });

});

Describe Application Logic

Create a component that contains the resources of your application. Define the following in Component.js:

sap.ui.define([
  'sap/ui/core/UIComponent'
], function(UIComponent) {
  "use strict";

  return UIComponent.extend("sap.ui.table.sample.OData2.Component", {
    metadata : {
      rootView : "sap.ui.table.sample.OData2.View",
      dependencies : {
        libs : [
          "sap.ui.table",
          "sap.ui.unified",
          "sap.m"
        ]
      },

      config : {
        sample : {
          stretch : true,
          files : [
            "View.view.xml",
            "Controller.controller.js"
          ]
        }
      }
    }
  });

});

Bootstrap OpenUI5 and Launch

To complete the MVC application, simply add the bootstrap and initialization code. Add these directly to index.html:

<!DOCTYPE HTML>

<html>
<head>
  <meta http-equiv="x-ua-compatible" content="ie=edge">
  <meta charset="utf-8"> 
  <title>Databricks Customers</title>
  
  <script id="sap-ui-bootstrap"
    src="https://openui5.hana.ondemand.com/resources/sap-ui-core.js"
    data-sap-ui-libs="sap.m"
    data-sap-ui-theme="sap_bluecrystal"
    data-sap-ui-xx-bindingSyntax="complex"
    data-sap-ui-preload="async"
    data-sap-ui-compatVersion="edge" 
    data-sap-ui-resourceroots='{"sap.ui.table.sample.OData2": "./", "sap.ui.demo.mock": "mockdata"}'>
  </script>
  
  <!-- application launch configuration -->
  <script>
  
      sap.ui.getCore().attachInit(function() {
        new sap.m.App ({
          pages: [
                new sap.m.Page({
                    title: "Databricks Customers", 
                    enableScrolling : false,
                  content: [ new sap.ui.core.ComponentContainer({
                  height : "100%", name : "sap.ui.table.sample.OData2"
                })]
            })
          ]
      }).placeAt("content");
    });

  </script>
</head> 
  <!-- UI Content -->
<body class="sapUiBody" id="content" role="application">
</body> 
</html>

The resulting SAPUI5 table control reflects any changes to a table in the remote Databricks data. You can now browse and search current Databricks data.