Hi Jakub.
Thank you very much for these insights.
So, as far as I understand, you are triggering some kind of integration based on activity logs.
Therefore you need the triggering activity log available in your PAM process.
Unfortunately, the macros available for PAM Integration integration only provide the persistent_id of the corresponding cr object and not the alg persistent_id when starting PAM processes.
My first idea was, to write the alg persid of the latest alg into the cr_object, so it would be available in the PAM process.
But what would happen, if more than one alg gets written in the same timeframe, let say a field update and a transfer.
What is about the following approach:
As you are already writing you own spel code, you might be able to start a PAM process with the alg persid as a start parameter instead of the cr persid. I have no idea, what consequences this might have with the rest of the PAM integration, like displaying PAM tasks and other integration points.
This would skip the necessity to write a file, and to start any external script.
For sure this would be a non supported coding. It's just a send_wait of a specific method passing appropriate arguments. Doing the same as before but passing the alg persid instead of the cr persid
Would that help ? And do you want to give it a try?
Regards
............Michael