Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 14 Next »

Unknown macro: {scrollbar}

How to create a parameterize report that uses data from a collection in MongoDB. By the end of this guide you should understand how data can be read from MongoDB and used in a report. The data we are going to use contains data about the flow of visitors to a web site. This guide shows how to create a report that shows the most popular landing pages for the sample web site.

Intro Video

Prerequisites

In order follow along with this how-to guide you will need the following:

MongoDB

A single-node local cluster is sufficient for these exercises but a larger and/or remote configuration will work as well. You will need to know the address and port that MongoDB is running on and have a user id and password for the server (if applicable).
These guides were developed using the MongoDB version 2.0.2. You can find MongoDB downloads here: http://www.mongodb.org/downloads

Kettle

A desktop installation of the Kettle design tool called 'Spoon'. Download here.

Pentaho Report Designer

Pentaho Report Designer (PRD) is a desktop tool for creating highly formatted reports that can be exported to many popular formats. Reports created with PRD can be published to a Pentaho BI Server so they can be accessed using a browser. Download here.

Data

To follow this guide you need to have a populated MongoDB collection. If you do not have any data in MongoDB yet you can use the Write Data To MongoDB guide to add some data to your MongoDB installation. You will also need the transformation created in Create a Report with MongoDB. The instructions in this guide assume that the demo data set is available in a collection called PageSuccessions in a database called Demo and that you have access to the transform called 'top_landing_pages.ktr'.

Step-By-Step Instructions

We will create the report using two tools. First we will use Spoon to modify the data transformation that selects data from MongoDB and sorts it into descending order. Then we will use PRD to create a report using the data transformation as its data source.

Setup

Start MongoDB if is not running.

Modify the Data Transformation

Start Spoon on your desktop.

  1. Open the Transformation: Choose 'File' -> 'Open' from the menu system. Select the 'top_landing_pages.ktr' file.
  2. Edit the MongoDb Input Step: Double-click on the 'MongoDb Input' step to edit its properties. Enter this information:
    1. Host name, Port, Authentication user and password: the connection information for your MongoDB installation.
    2. Database: 'Demo' or another database if you want.
    3. Collection: 'PageSuccessions'
    4. Query expression: { "$query" : { "url" : "${page}" }, "$orderby" : { "Count" : -1 } }
      The ${} syntax is used to specify that a parameter should be inserted. In this case the parameter is called "page".
      The window should look like this:
  3. Define the Transformation Parameter: In order the for parameter in the JSON query to work we need to define the parameter at the transformation level. Choose 'Edit' -> 'Settings...' from the menu system. Click on the 'Parameters' tab. Enter 'page' for the parameter and '—firstpage—' for the default value.
  4. Preview the Data: With the 'Json Input' step selected click on the Preview toolbar button (the green arrow with the magnifying glass ) or right-click on the step and choose 'Preview'. The 'Transformation debug dialog' will open. Click on 'Quick Launch'. You will should see the data returned by the MongoDB query with the extracted fields.
  5. Change the parameter value: Now we will preview the step with a different parameter. Open the 'Transformation debug dialog' window again. Click on 'Configure', the 'Execute a transformation' window will open. In the parameters section enter '/about' for the value of the page parameter.

    Click on the 'Launch' button.

    The data displayed is now for the '/about' url, . Try parameter values of '/demo' or '/events' to see how the data changes.
  6. Save the Transformation: Choose 'File' -> 'Save as...' from the menu system. Save the transformation as 'page_successions' into a folder of your choice.

Start Pentaho Report Designer

When PRD starts click on the 'Report Wizard' button or choose 'File' -> 'Report Wizard...' from the menu.

  1. Select a Template: On the 'Look and Feel' stage of the wizard select a report template and click on 'Next'
  2. Add a Data Source: On the 'Data Source' stage click on the '+' icon in the top right to add a new data source.
  3. Choose the Data Source Type: From the 'Choose Type' list click on 'Pentaho Data Integration'.
  4. Add a Query: In the 'Pentaho Data Integration Data Source' window click on the '+' icon to add a new query. A default query, called 'Query 1' is added. Change the name of the query to 'Top Landing Pages'. Then click on the 'Browse' button and select the 'page_successions.ktr' file created above. Finally select the 'Json Input' step.

    If you want you can click the 'Preview' button to see the data generated by the 'Json Input' step. Click on 'OK' to close the 'Pentaho Data Integration Data Source' window.
  5. Select the Query: In the 'Report Design Wizard' click on the 'Top Landing Pages' query to select it and then click on the 'Next' button.
  6. Layout the Fields: In the 'Layout Step' of the wizard click on 'URL' and then click the button to add 'URL' to the 'Group Items By' box. Add 'NextURL' and 'Count' to the 'Selected Items' box. This will position these two fields as two columns in the report.

    Click on the 'Next' button.
  7. Format the Fields: In this step you can change the formatting of the fields. Click on the 'URL' field to highlight it, then change the 'Group Header Label' to 'Source Page: '. Click on the 'NextURL' field to highlight it, then change the Display Name to 'Destination Page'. Click on 'Count' to highlight it, then change the data format to '#,###;(#,###)' and select 'Sum' from the 'Aggregation' list.
  8. Finish the Wizard: Click on 'Finish'. The wizard will close and you will see your report in design mode.
  9. Change the Titles: Double-click on the report title and change it to 'Pages Successions'. Double-click on the first subtitle and change it to 'In descending order'. Double-click on the second subtitle and remove the text. Notice that there are lot of style properties you can set for these report elements.
  10. Add a Parameter Data Source: Next, we need to create a data source that will be used to populate a UI control so that users can select a valid value for the report's parameter. On the right panel click on the 'Data' tab. Click on the 'Add Datasources' toolbar button and choose the 'XML' option.
  11. Define the Parameter XML Data Source: In the 'XML Data Source Editor' window click on the 'Browse' button and select the 'pages.xml' file. Click on the green '+' button to add a new Available Query. For the 'Query Name' enter 'PageList'. For the 'Query' enter '/pages/page'.

    Click on the 'Preview...' button and you will see the values read from the 'pages.xml' file.
  12. Add a Parameter: Next we will create a report parameter and UI control. On the data tab click on the 'Add a new Master-report Parameter' toolbar button . The 'Add Parameter...' window opens. Double-click on the 'XML' DataSource folder to open it, then click on the 'PageList'. Fill in these fields:
    1. Name: 'Page'
    2. Label: 'Page'
    3. Value Type: 'String'
    4. Display Type: 'Drop Down'
    5. Query: 'PageList'
    6. Value and Display Name: 'url'
  13. Connect the Parameter: Finally we need to associate the report parameter with the transformation parameter. On the right panel, double-click on the 'Page Successions' data source to edit it. Click on the 'Edit Parameter' button, the 'Transformation Parameters' window will open. Click on the green + button to add a row to the table. For 'DataRow Column' choose 'Page' from the dropdown list. For 'Transformation Parameter' choose 'page' from the dropdown list.

    Click on the 'OK' button to close the 'Transformation Parameters' window and then click on the 'OK' button in the 'Pentaho Data Integration Data Source' window.
  14. Preview the Report: Click on the preview icon (the eye towards the top left) to preview your report. Notice that there is a dropdown list above the report. You can use this control to change the parameter for the report. You can alter the size of the parameter panel using the grab bar between it.
  15. Run or Export the Report: Click on the run button (green arrow on the toolbar) or choose 'File' -> 'Export' from the menu system. Select the output format for your report. If you used the run button the report will run and the appropriate application will be opened to view the report. If you chose the export option you will be prompted for the location and name of the file that will be exported. With either method you will be given the opportunity to select the parameter value before the report runs.

Check The Results

  1. Using the run or export options you will be able to create and view PDF, Excel, HTML, and other file types. Try these options and check that the exported files contain the expected data.

Summary

During this guide you learned how to read data from a MongoDB collection and use it as the data source for a parameterized report.
Other guides in this series cover to sort and group MongoDB data, and combine data from MongoDB with data from other sources.

Unknown macro: {scrollbar}
  • No labels