I did a bit of tests with FEEL-Scala extension. Looks awesome and runs fast. One question related to handling complex data types. I am looking for a good example of implementing a case of list/context based rules.
So we have a table of manual journal entries (lines) that needs to be validated against a collection of rules. At the moment we implemented Array pattern and send lines one by one to be validated. In this case for each line we return a list of failed validations with error messages. This is not fast and require “process wrapper” around DMN.
diagram_2 - Copy.dmn (1.6 KB)
validations_example.bpmn (11.1 KB)
We would like to implement rules on the whole dataset, so an output would be slightly different. Each failed rule will return a list of input lines that failed validation test with corresponding error message. So we will need to dynamically populate output variables from input expressions. Ideally, we are looking for a way to do it without bpmn orchestrator, but rather running REST calls to DMN engine directly.
Is it something achievable with FEEL?
Thank you in advance.