#1038 S1: IC4 return-CASE repro + fixture lock

[lmeyerov]

Summary

Execution slice S1 for meta #1261 and source residual #1038.

Pin the IC4/new-topics return-side CASE ... END shape in deterministic test coverage and record current behavior against origin/master.

Parent/Related

• Meta: #1261
• Source residual: #1038
• Queue meta: #992

Scope

• Confirm canonical IC4-shaped query text/shape used for #1038
• Keep regression lock in graphistry/tests/compute/gfql/cypher/test_lowering.py
• Record whether current behavior is pass or fail and update linked issues

Acceptance Criteria

• Deterministic test fixture covers the #1038 return-side CASE shape
• Expected behavior is explicitly documented (current state pass/fail)
• Targeted pytest command + result captured in issue comment
• Links back to #1261 and #1038

[lmeyerov]

Superseded: see corrected S1 receipt comment at #issuecomment-4367530953.

[lmeyerov]

Correction: previous comment had shell-expansion artifacts.

Initial S1 progress on branch issue-1261-case-return-decomp:

• Added regression lock: test_issue_1038_ic4_return_side_case_expression_regression_lock in graphistry/tests/compute/gfql/cypher/test_lowering.py
• Targeted validation:
  • PYTHONPATH=. python -m pytest graphistry/tests/compute/gfql/cypher/test_lowering.py -k "issue_1038_ic4_return_side_case_expression_regression_lock" -q
  • result: 1 passed

Observation on current origin/master: IC4-shaped return-side CASE repro executes successfully in local checks, so this lock currently protects against regression rather than pinning an active failure.

[lmeyerov]

Closed by merged PR #1271 (commit 74efb35): scope delivered and regression receipts are in the PR thread and linked parent issues.