The Row Warning tool is very useful for automated testing and productionizing models. It enables you to create custom warnings based on conditions so you can highlight if something seems strange.

The Row Warning tool is typically preceded by a Filter tool to create your condition.

You can create multiple warnings in a single model

Model warnings are shown explicitly in your folders

Note that all logs messages of Models can be extracted as a table with the Less Log Connector. That means that all your custom Warnings (using the Row Warning tool) will also show up in the logs which means you can extract all your warnings at scale for monitoring performance in Less with the Less Logs Connector.

​
Configuration

There are two required inputs of the Row Warning tool.

1

Set Row Trigger

Input the number of incoming rows that should trigger a warning. If you input 1 and your incoming dataset is contains more than 1 row, then the warning will be triggered.

2

Input Warning

Input your custom warning message.

​
When To Use

The Row Warning tool is typically used for flag shortcomings of your data or to flag certain scenarios:

  • It is very useful to set up warnings that inform you of data quality issues. It could be invoices with an amount lower than zero.
  • The Row Warning tool takes precedent over errors. That means you can use it to overwrite errors, for example after an API tool to test if the dataset is empty. That will overwrite errors of downstream tools that except more than zero rows.
  • Use the Row Warning tool to flag if certain columns are NULL

​
Examples