bedrock.cube.data.copy

[balooox]

🐞 Bug Report: bedrock.cube.data.copy incorrectly aborts on partial dimension matches

Description

The process bedrock.cube.data.copy aborts when the parameter pZeroTarget = 1 is set and at least one of the mapped dimensions has the same source and target element.

This behavior is incorrect.
The process should only abort if all mapped dimensions have identical source and target elements.

Currently, the process stops even when only one of several mapped dimensions matches.

---

Current Behavior

• The process aborts if any single mapped dimension has identical source and target elements.

Expected Behavior

• The process should abort only if all mapped dimensions have identical source and target elements.
• If only some dimensions match but at least one differs, the process should not abort.

---

Example Mappings

Mapping (pEleMapping)	Current Behavior	Expected Behavior	Mismatch
Scenario: A → A	Abort	Abort	–
Scenario: A → A & Year: 2026 → 2027	Abort	Do not abort	❌
Scenario: A → A & Year: 2026 → 2026	Abort	Abort	–

---

Summary

The abort logic for pZeroTarget = 1 is currently too strict.
It aborts on partial dimension matches, but should only abort when all mapped dimensions match exactly.

[balooox]

🔍 Additional Findings: pFilter as a Partial Workaround

The process also supports an additional parameter: pFilter.
This parameter can be used to restrict the copy operation to specific, allowed source elements.

With pFilter, the previously described issue can be partially avoided.
For example:

Previous (problematic) setup:

• pEleMapping:
  • Scenario: A → A & Year: 2026 → 2027
    → This triggers an unwanted abort.

Workable alternative using pFilter:

• pEleMapping:
  • Year: 2026 → 2027
• pFilter:
  • Scenario: A

This prevents the process from aborting, because the Scenario dimension is no longer mapped directly.

---

⚠️ Why pFilter is not a sufficient solution

In some of our processes, the element mapping is generated dynamically and iteratively.

For example:
Scenario: A → B & Year: 2026 → 2026 !!! Process break with current behavior
Scenario: A → B & Year: 2026 → 2027
Scenario: A → B & Year: 2026 → 2028

In such cases:

• The set of mapped dimensions changes per iteration
• The logic to maintain and split mappings across pEleMapping and pFilter becomes highly complex

---

Given the above, I still strongly recommend adjusting the process behavior