Learning Pentaho Data Integration 8 CE - Third Edition

Most of the Pentaho engines, including the engines mentioned earlier, were created as community projects and later adopted by Pentaho. The PDI engine is not an exception; Pentaho Data Integration is the new denomination for the business intelligence tool born as Kettle.

In April 2006, the Kettle project was acquired by the Pentaho Corporation, and Matt Casters, the Kettle founder, also joined the Pentaho team as a data integration architect.

When Pentaho announced the acquisition, James Dixon, the Chief Technology Officer, said:

By joining forces with Pentaho, Kettle benefited from a huge developer community, as well as from a company that would support the future of the project.

From that moment, the tool has grown with no pause. Every few months a new release is available, bringing to the user's improvements in performance and existing functionality, new functionality, and ease of use, along with great changes in look and feel. The following is a timeline of the major events related to PDI since its acquisition by Pentaho:

Paying attention to its name, Pentaho Data Integration, you could think of PDI as a tool to integrate data.

In fact, PDI does not only serve as a data integrator or an ETL tool. PDI is such a powerful tool that it is common to see it being used for these and for many other purposes. Here you have some examples.

Loading data warehouses or data marts

The loading of a data warehouse or a data mart involves many steps, and there are many variants depending on business area or business rules.

However, in every case, with no exception, the process involves the following steps:

1. Extracting information from one or more databases, text files, XML files, and other sources. The extract process may include the task of validating and discarding data that doesn't match expected patterns or rules.
2. Transforming the obtained data to meet the business and technical needs required on the target. Transforming includes such tasks such as converting data types, doing some calculations, filtering irrelevant data, and summarizing.
3. Loading the transformed data into the target database or file store. Depending on the requirements, the loading may overwrite the existing information or may add new information each time it is executed.

Kettle comes ready to do every stage of this loading process. The following screenshot shows a simple ETL designed with the tool:

ETL process

The previous examples show typical uses of PDI as a standalone application. However, Kettle may be used embedded as part of a process or a data flow. Some examples are preprocessing data for an online report, sending emails in a scheduled fashion, generating spreadsheet reports, feeding a dashboard with data coming from web services, and so on.

Spoon is PDI's desktop design tool. With Spoon, you design, preview, and test all your work, that is, transformations and jobs. When you see PDI screenshots, what you are really seeing are Spoon screenshots. The other PDI components, which you will learn about in the following chapters, are executed from Terminal windows.

Here is how you launch the tool:

Welcome page

These simple steps would be enough to start working, but before that, it's advisable to customize Spoon to your needs. For doing that:

Kettle options

Look and Feel options

French as preferred language

As you can see, the Options window has a lot of settings. We changed only a few, just to show the feature. Feel free to change the settings according to your needs or preferences.

In particular, take note of the following tip about the selected language.

One of the settings that you changed was the appearance of the Welcome! window at startup. The Welcome! page is full of links to web resources, blogs, forums, books on PDI, and more. Following those links, you will be able to learn more and become active in the Pentaho community. You can reach that window anytime by navigating to the Help | Welcome Screen option.

Exploring the Spoon interface

As explained earlier, Spoon is the tool with which you create, preview, and run transformations. The following screenshot shows you the basic work areas: Main Menu, Main Toolbar, Steps Tree, Transformation Toolbar, and Canvas (Work Area). Note that there is a sample Transformation opened; it allows you to see how the tool looks when you are working with it:

Spoon interface

Note

The terms Canvas and work area will be used interchangeably throughout the book.

The Steps Tree option is only available in Design view. There is also an area named View that shows the structure of the Transformation currently being edited. You can see that area by clicking on the View tab at the upper-left corner of the screen:

Spoon view area

Pentaho Data Integration is built on a pluggable architecture. This means that it can be extended to fulfill needs not included out of the box. The Marketplace—a plugin itself—emerged as a straightforward way for browsing and installing available plugins, developed by the community or even by Pentaho.

In PDI, you will find plugins for connecting to a particular database engine, for executing scripts, for transforming data in new ways, and more. According to the purpose, the plugins are classified into several types: big data, connectivity, and statistics, among others. In particular, there is a type named Experimental, which you will not use except for playing around. There is another type named Deprecated, which we don't recommend you use unless you need it for back compatibility.

An important point to highlight about plugins is the maturity stage. The maturity classification model consists of two parallel lanes:

There are four stages in each lane. To put it simply, stage 1 means that the plugin is under development (it is usually a lab experiment), while stage 4 indicates a mature state; a plugin in stage 4 is successfully adopted and could be used in production environments. Stages 2 and 3 are stages in between these two.

That said, let's go back to Spoon. You can access the Marketplace page by clicking on Marketplace from the Tools menu. The page is quite simple, as shown in the following screenshot:

Marketplace

By default, you see the list of all the Available/Installed plugins. There is a secondary tab where you can filter just the installed ones.

Also, you can filter by plugin Type and by maturity Stage. And if you are looking for a particular plugin, there is also a Search textbox available.

Once in the Marketplace page, for every plugin you can see:

If you click on the plugin name, a pop-up window shows up displaying the full description for the selected plugin, as shown in the following example:

Sample plugin in Marketplace

Besides browsing the list of plugins, you can install or uninstall them:

It's premature to decide if you need to install a plugin for your work. So let's put this subject aside for a while; we will get back to this feature later in the book.

A Transformation is an entity made of steps linked by hops. These steps and hops build paths through which data flows: the data enters or is created in a step, the step applies some kind of Transformation to it, and finally, the data leaves that step. Therefore, it's said that a Transformation is data flow oriented. Graphically, steps are represented with small boxes, while hops are represented by directional arrows, as depicted in the following sample:

Steps and hops

A Transformation itself is neither a program nor an executable file. It is just plain XML. The Transformation contains metadata, which tells the Kettle engine what to do.

A step is a minimal unit inside a Transformation. A big set of steps is available, either out of the box or the Marketplace, as explained before. These steps are grouped in categories, as, for example, input, output, or transform. Each step is conceived to accomplish a specific function, going from a simple task as reading a parameter to normalizing a dataset.

A hop is a graphical representation of data flowing between two steps: an origin and a destination. The data that flows through that hop constitutes the output data of the origin step and the input data of the destination step.

That's enough theory for now. Let's see it in practice.

In this section, we will design, preview, and run a simple Hello World! Transformation; simple, but good enough for our first practical example.

Designing a Transformation

Here are the steps to start working on our very first Transformation. All you need for starting is to have PDI installed:

1. Open Spoon.From the main menu and navigate to File | New | Transformation.
2. On the left of the screen, under the Design tab, you'll see a tree of Steps. Expand the Input branch by double-clicking on it.

Note

Note that if you work in Mac OS, a single click is enough.

3. Then, left-click on the Data Grid icon and without releasing the button, drag and drop the selected icon to the main canvas. The screen will look like the following screenshot:

Dragging and dropping a step

Note

The dotted grid appeared as a consequence of the changes we made in the options window. Also, note that we changed the preferred language back to English.

4. Double-click on the Data Grid step you just put on the canvas, and fill the Meta tab as follows:

Configuring a metadata tab

5. Now select the Data tab and fill the grid with some names, as in the following screenshot. Then click on OK to close the window:

Filling a Data tab

6. From the Steps tree, double-click on the Scripting branch, click on the User Defined Java Expression icon, and drag and drop it to the main canvas.
7. Put the mouse cursor over the Data Grid step and wait until a tiny toolbar shows up succeeding the Data Grid icon, as shown next:

Mouseover assistance toolbar

8. Click on the output connector (the icon highlighted in the preceding image) and drag it towards the User Defined Java Expression (UDJE) step. A greyed hop is displayed.

9. When the mouse cursor is over the UDJE step, release the button. A link—a hop from now on is created from the Data Grid step to the UDJE step. The screen should look like this:

Connecting steps with a hop

10. Double-click the UDJE icon and fill the grid as shown. Then close the window:

Configuring a UDJE step

Done! We have a draft for our first Transformation. A Data Grid with the names of a list of people, and a script step that builds the hello_message.

Before continuing, let's just add some color note to our work. This is totally optional, but as your work gets more complicated, it's highly recommended that you comment your transformations:

1. Right-click anywhere on the canvas to bring a contextual menu.
2. In the menu, select the New note option. A note editor appears.
3. Type some description, such as Hello, World!. Select the Font style tab and choose some nice font and colors for your note, and then click on OK. This should be the final result:

Hello World Transformation

The final step is to save the work:

1. From the main menu, navigate to Edit | Settings.... A window appears to specify Transformation properties. Fill the Transformation name textbox with a simple name, such as hello world. Fill the Description textbox with a short description, such as My first transformation. Finally, provide a more clear explanation in the Extended description textbox, and then click on OK.

2. From the main menu, navigate to File | Save and save the Transformation in a folder of your choice with the name hello_world.

Next step is to preview the data produced and run the Transformation.

Previewing and running a Transformation

Now we will preview and run the Transformation created earlier. Note the difference between both:

• The Preview functionality allows you to see a sample of the data produced for selected steps
• The Run option effectively runs the whole Transformation

In our Transformation, we will preview the output of the User Defined Java Expression step:

1. Select the User Defined Java Expression step by left-clicking on it.
2. Click on the Preview icon in the bar menu preceding in the main canvas: 

Preview icon in the Transformation toolbar

3. The Transformation debug dialog window will appear. Click on the Quick Launch button.

4. A window will appear to preview the data generated by the Transformation, as shown in the following screenshot:

Previewing the Hello World Transformation

5. Close the preview window.

Note

You can preview the output of any step in the Transformation at any time of your designing process. You can also preview the data even if you haven't yet saved the work.

Once we have the Transformation ready, we can run it:

1. Click on the Run icon:

Run icon in the Transformation toolbar

2. A window named Run Options appears. Click on Run.

Note

You need to save the Transformation before you run it. If you have modified the Transformation without saving it, you will be prompted to do so.

3. At the bottom of the screen, you should see a log with the result of the execution:

Sample execution results window

Whether you preview or run a Transformation, you'll get an Execution Results window showing what happened. You will learn more about this in Chapter 2, Getting Started with Transformations.