Branches for Agent processing
How can I leverage branches for agent development? My application relies heavily on agents for automation. We are constantly having to check-in rules to our main ruleset in order to unit test changes. This can be very disruptive to other developers. I'm currently running on 7.2.2 but planning to upgrade to PRPC 8 early in 2020. I would appreciate any tips regardless of which version that you are using.
Pegasystems Inc.
IN
Hi,
I am not sure is there any way to achieve the functionality, but others have tried and it didn't go well for them
You can refer below discussion
https://collaborate.pega.com/question/agents-doesnt-work-if-configured-branch-rulesets
https://collaborate.pega.com/question/what-are-disadvantages-branch-rulesets
Thank You
Computershare
US
That's what I expected but I wanted to ask just to be sure. When I'm concerned that my changes are too risky I tend to create a test activity that calls the agent's activity and any code in my branch or checked out rules. Typically I have to read a queue record first then call the activity. I usually keep this activity for debugging later too. Agent tracing can be a pain to use and it can be hard to know if the queue record being processed is relevant to the bug at hand. By using my activity I can be sure that I'm debugging the right use case, and with the right data.