Book Image

Learning Pentaho Data Integration 8 CE - Third Edition

Book Image

Learning Pentaho Data Integration 8 CE - Third Edition

Overview of this book

Pentaho Data Integration(PDI) is an intuitive and graphical environment packed with drag-and-drop design and powerful Extract-Tranform-Load (ETL) capabilities. This book shows and explains the new interactive features of Spoon, the revamped look and feel, and the newest features of the tool including transformations and jobs Executors and the invaluable Metadata Injection capability. We begin with the installation of PDI software and then move on to cover all the key PDI concepts. Each of the chapter introduces new features, enabling you to gradually get practicing with the tool. First, you will learn to do all kind of data manipulation and work with simple plain files. Then, the book teaches you how you can work with relational databases inside PDI. Moreover, you will be given a primer on data warehouse concepts and you will learn how to load data in a data warehouse. During the course of this book, you will be familiarized with its intuitive, graphical and drag-and-drop design environment. By the end of this book, you will learn everything you need to know in order to meet your data manipulation requirements. Besides, your will be given best practices and advises for designing and deploying your projects.
Table of Contents (17 chapters)

Going forward and backward across rows


Besides the common use cases explained in the previous sections, there are other use cases that work with groups of rows, looking for rows before or after the current one within each group.

Some examples of this are as follows:

  • You have a dataset with monthly sales, group by product line. For each product line, you want to calculate the variation of sales from one month to the next.
  • You have daily sales and want to infer the number of days without sales. (This is the gap in days between a date and the next in your dataset.)
  • You have a dataset with a list of sales amounts and sales commissions. The fields in your dataset are sales_amount_from, sales_amount_to, and commission_%. You detected that there are overlaps in the data:
sales_amount_from, sales_amount_to, commission_%
0, 1000, %5
1001, 5000, %15
4500, 9999, %15

You want to automatically fix these overlaps. In this case, you want to change the second row to the following:

1001, 4499, %15

In all these examples...