Drill-through Functionality of the Profiling Step Disabled

Reported for version 10

Drill-through is a functionality of the Profiling step in Ataccama products which enables inspection of the individual records that include the generated statistics and other measures. In other words, the drill-through functionality provides an insight into the data from a specific point of view. These may be found under the Profile viewer. As it stores additional fields, a database connection is required. To allow drill-through in the generated profile, click Enable Drill-through of the Drill-through tab in the Profiling step.  

When you open the Profile viewer after running the Profiling step, you see the different statistics. To actually drill through them, right click on the type of the statistic to see this option. The whole Drill-through View shows the representative values of you data according to your drill-through parameters. However, only one record is highlighted and shows the chosen type with its value. 

You may notice that not all the statistic types support the drill-through option. This is because, even though there are some instances of data to drill through, the engine is not able to pick a specific record. Take a look at the following example:

This basic analysis example of the profiled data is from 7.03 Party advanced tutorial. The drill-through option of the Duplicate records is greyed-out. The reason for this is that the engine cannot choose one representative value from the dataset of the duplicates. The same applies to the Distinct, Non-unique and Unique datasets of the Counts table list. Generally said, you can drill through datasets with a provided frequency along their values or a given percentage of the counts and values of the datasets. You cannot drill through sets of data where it is unclear which individual value should be chosen from the set.

When overviewing the statistics of your data and doing drill-through, remember that performing such task may not be accomplished at all times as the basis of the functionality lies in choosing representative values of a dataset.