From d89c72e0183d9981350ed7e70c5e95501517e160 Mon Sep 17 00:00:00 2001 From: weichen Date: Tue, 13 Oct 2026 11:10:33 +0000 Subject: [PATCH] Reserve a distinct one-shot A01 evaluation label Why: Reusing an exhausted authorization label would overwrite or misattribute the next controlled quality sample. Evidence: New label/output and 30000-token guarded-run tests failed before implementation and passed after; full Backend Go suite passed. Impact: The r0.2 A01 harness can be used only after separate resource approval, with an independent output name and existing two-run guard. Refs: U3 quality implementation plan r0.2, Task 6 preparation --- eval/live/work_checkpoints_test.go | 4 ++++ internal/server/work_checkpoint_eval_budget_test.go | 1 + internal/server/work_checkpoint_eval_test.go | 2 +- 3 files changed, 6 insertions(+), 1 deletion(-) diff --git a/eval/live/work_checkpoints_test.go b/eval/live/work_checkpoints_test.go index e7959fc..cb27b9e 100644 --- a/eval/live/work_checkpoints_test.go +++ b/eval/live/work_checkpoints_test.go @@ -33,6 +33,9 @@ func checkpointEvalOutputName(filter, authorization string) (string, bool) { if filter == "A01" && authorization == "approved-followup-20261013" { return "candidate-followup-a01", true } + if filter == "A01" && authorization == "approved-r02-20261013" { + return "candidate-r02-a01", true + } if filter == "A02,A04" { return "candidate-remaining", true } @@ -82,6 +85,7 @@ func TestFollowupA01KeepsItsOwnOneShotResult(t *testing.T) { }{ {"A01", "approved-20261013", "candidate-extra-a01", true}, {"A01", "approved-followup-20261013", "candidate-followup-a01", true}, + {"A01", "approved-r02-20261013", "candidate-r02-a01", true}, {"A01", "", "", false}, {"A01", "unknown", "", false}, {"A02,A04", "", "candidate-remaining", true}, diff --git a/internal/server/work_checkpoint_eval_budget_test.go b/internal/server/work_checkpoint_eval_budget_test.go index 46e8f6c..76f2eb4 100644 --- a/internal/server/work_checkpoint_eval_budget_test.go +++ b/internal/server/work_checkpoint_eval_budget_test.go @@ -102,6 +102,7 @@ func TestFollowupA01UsesNewBoundedAuthorization(t *testing.T) { {"", 20000, false}, {"approved-20261013", 20000, true}, {"approved-followup-20261013", 30000, true}, + {"approved-r02-20261013", 30000, true}, {"unapproved", 20000, false}, } { tokens, guard := evalA01Budget(tc.label) diff --git a/internal/server/work_checkpoint_eval_test.go b/internal/server/work_checkpoint_eval_test.go index 927b7b2..1457e92 100644 --- a/internal/server/work_checkpoint_eval_test.go +++ b/internal/server/work_checkpoint_eval_test.go @@ -47,7 +47,7 @@ func evalA01Budget(label string) (int, bool) { switch label { case "approved-20261013": return 20000, true - case "approved-followup-20261013": + case "approved-followup-20261013", "approved-r02-20261013": return 30000, true default: return 20000, false