Result variables shouldn't be merged into the process instance when output mappings exist

[abdul99ahad]

Describe the bug

When a script task has output mappings defined, only the variables specified in the output mappings should be merged into the process instance. However, the current behavior merges all result variables into the process instance, even when output mappings exist. This leads to unintended variables being accessible at the process level, violating the expected scoping rules.

Here, we shouldn't get foo in the suggestions as output mappings already exists and foo is a result variable.

See Camunda docs
If one or more output mappings are defined, the results variables are set as local variables in the scope where the mapping is defined. Then, the output mappings are applied to the variables and create new variables in this scope. The new variables are merged into the parent scope. If there is no mapping for a job/message variable, the variable is not merged.

If no output mappings are defined, all results variables are merged into the process instance.

this issue first discussed in #4614

Steps to reproduce

1. Create a script task
2. Create script variables and output variables
3. Create a user task
4. In the user task, see in output or input variables; script variable is suggested

Expected behavior

• If output mappings exists, the script variable shouldn't be merged in process instance or parent scope
• If output mappings doesn't exist, the script variable can be merged in process instance or parent scope

Environment

• OS: MacOS 14.7
• Camunda Modeler Version: v5.28.0
• Execution Platform: Camunda 8.6
• Installed plug-ins: none

Additional context

Related to #4614

[barmac]

I've added "related" part to the additional context part.

I can reproduce the issue. I've also confirmed that the runtime behaviour is as described in the doc.

[barmac]

Perhaps it makes sense to tackle this together with #4614? What do you think @abdul99ahad ?

[abdul99ahad]

Perhaps it makes sense to tackle this together with #4614? What do you think @abdul99ahad ?

I'm not sure if we could tackle this together. To my understanding, issue #4614 is about parsing the result variables and this is about handling the scope of variables. Let me know if you think otherwise.

[philippfromme]

@abdul99ahad Should we put this in backlog or ready?

[abdul99ahad]

Added in ready

[barmac]

To clarify: this is NOT a part of the v5.31.0 release 👍