Oracle HCM Cloud test automation with Selenium and Python

Today I would like to talk about one crucial point in the Oracle Cloud Applications lifecycle.
There are thousands of customers around the world that use Oracle HCM/ERP/EPM/CX Cloud Applications. A SAAS delivery model gives them many more benefits over the on-premise deployment but they also face some inconveniences. The main pain point is the quarterly updates. All customer PODs are updated at least 4 times a year and those are mandatory. You can’t skip or reschedule them.

Non-production environments: First Friday of your Quarterly Update month. The exception is the Middle East, where the update is applied on Thursday.
Production environments: Third Friday of your Quarterly Update
month. The exception is the Middle East, where the update is applied on Thursday.
You can review the Oracle Applications Cloud – Standard Update Policy document here:
It means that customers have only 2 weeks’ time slot to test all the processes to make sure there are no issues after an update.
This very time and resource-consuming process because the number of transactions to be tested may reach hundreds. It is a real challenge for IT support or shared service departments.
But there is a solution that intended to extremely simplify testing of all UI transactions every quarter and it is based on test automation framework.

Let me briefly explain how it works:

  • Use Selenium IDE – to record a test script
  • Export to Python
  • Create a TestCase based on the generated python code adopting a Python unittest data driven framework

More information about a unittest in google search like this

Concluding the post by a short video as usual:

I hope you find it helpful.




Generate HCM Data Loader Files for Data That Failed to Load

Hi community,

Long time no see. It’s time to post something interesting.
Recently I have been reviewing Oracle HCM Cloud 21C new features document and bump into the feature I found it quite useful. It is already available in our 21A instance, so it seems I missed it earlier.
The feature is related to HCM Data Loader and allows generating an HDL file for the failed lines.
This is quite useful during data migration steps when you run large HDL files and if it is only a few record failed you can just extract those failed lines into a new HDL, fix it (normally errors are related to incorrect referential data) and then load only this part. It saves a lot of time.

Well let’s get through it.

  • You have to enable this feature in the FSM task: Configure HCM Data Loader:
  • Load something with HDL to get failed lines:
  • Press an “Extract Failed Lines” button and submit a process
  • Once it is finished you can download an HDL file generated

That’s it!

I hope you find it helpful.