PDI Version 5.1 Minor Functionality Changes
Overview
Lists minor functionality changes for PDI version 5.1. Backward compatibility provisions and related Jira cases are listed.
The following table describes minor functionality changes for PDI version 5.1. Provisions to preserve backward compatibility and related Jira cases are also listed.
Description | Related Jira Cases |
In 5.0, the Checks if files exist, Check if a folder is empty, and Evaluate rows number in a table job entries fail when they are scheduled to run because the number of errors were set internally. In 5.1, the error flag is no longer set, but if you need it to be, set KETTLE_COMPATIBILITY_SET_ERROR_ON_SPECIFIC_JOB_ENTRIES to Y. | |
In 5.0, the strict checking of numbers was enabled by default when numbers were converted from a string. While this was not enforced for integer parsing in 5.0, but is enforced for integer parsing in 5.1. To be backward compatible, set KETTLE_LENIENT_STRING_TO_NUMBER_CONVERSION to Y. | |
When using the aggregate functions MIN and SUM in the Group by, Memory Group by or Row Denormaliser steps, the behavior of including NULL values into the aggregate has changed. To support backward compatibility, there are two new variables.
| |
In 5.0, the time zone for the DATE data type was ignored. This has been addressed in 5.1. But, if you want the time zone for the DATE data type to continue to be ignored, set KETTLE_COMPATIBILITY_DB_IGNORE_TIMEZONE to Y. | |
The default conversion format for Timestamps is yyyy/MM/dd HH:mm:ss.SSSSSSSSS. Modify the value of the KETTLE_DEFAULT_TIMESTAMP_FORMAT variable if you need to change the default conversion format. | |
New Database Connections set the option Preserve case of reserved words to True by default. By setting this option by default, reserved words are no longer changed to upper or lower case. |