fix the other 4: close all OPEN-list items in one wave

Substantial wave addressing all 4 prior OPEN items. Three closed in full, one partially (the speculative half deliberately deferred). OPEN #1 — Periodic fresh→main index merge (FULL): - POST /v1/vectors/index/{src}/merge with {dest, clear_source} - Idempotent on re-runs (existing-in-dest items skipped) - internal/vectord/index.go: new Index.IDs() snapshot method + i.ids tracker field as canonical ID set, independent of meta map's nil-vs-{} sparseness (was a real bug — IDs() backed by meta alone missed items added with nil metadata) - 4 cmd-level integration tests (happy path drain+clear, dim mismatch, dest not found, self-merge rejection) + 1 unit test - DecodeIndex backward-compat: old envelopes restore i.ids from meta keys (best effort; new items going forward use the tracker) OPEN #2 — Distillation SFT export (SUBSTRATE): - internal/distillation/sft_export.go ports the load-bearing half: IsSftNever predicate + ListScoredRunFiles (data/scored-runs/YYYY/ MM/DD walk) + LoadScoredRunsFromFile + partial ExportSft. - Synthesis (instruction/input/response generation) deferred to a separate wave — too big for this session, but the substrate makes the next wave a port-not-design exercise. - TestSftNever_PinsExpectedSet locks the contamination firewall set: if a future commit adds/removes from SftNever, this test fails — forcing the change through review. - 5 new tests; firewall fires end-to-end through the partial port. OPEN #3 — Distribution drift via PSI (FULL): - internal/drift/drift.go: ComputeDistributionDrift via Population Stability Index. Standard finance/risk metric, well-defined verdict tiers (stable < 0.10, minor 0.10–0.25, major ≥ 0.25). - Equal-width bucketing over combined min/max so neither dist falls outside; epsilon-clamping for empty buckets so log doesn't blow up. Per-bucket breakdown for drilldown. - Pairs with the existing ComputeScorerDrift: scorer drift is categorical, distribution drift is continuous. Different shapes, same package. - 7 new tests covering identical-is-stable, hard-shift-is-major, moderate-detected-not-stable, empty-inputs-safe, all-identical- safe, bucket-counts-conserved, num-buckets-clamping. OPEN #4 — Ops nice-to-haves (PARTIAL — wall-clock done, others deferred): - (a) Real-time wall-clock for stress harness: per-phase elapsed time logged to stdout as it runs (`[stress] phase NAME starting (T+12.3s)` + `[stress] phase NAME done — 8.5s (T+20.8s)`). Output.PhaseTimings + Output.TotalElapsedMs in JSON. - (b) chatd fixture-mode S3 mock + (c) liberal-paraphrase calibration: not actioned — no fired trigger, would be speculative. Documented as deferred-until-need rather than ignored. Per the project's discipline ("don't add features beyond what the task requires"). OPEN list now empty / steady-state. Future items will land as production triggers fire. Build + vet + tests green; 18 new tests across the 4 closures. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-30 23:42:11 -05:00
parent 356d76b4b0
commit b216b7e5b6
10 changed files with 1139 additions and 16 deletions
@@ -102,6 +102,9 @@ type ResultRef struct {
 type Event struct {
 	Phase             string         `json:"phase"`
 	Hour              int            `json:"hour"` // operational-narrative time label, not real wall clock
+	// Real wall-clock time of the event is in TimestampUnixNano below
+	// (already present pre-OPEN-#4). Consumers can derive
+	// "since run start" from event.TimestampUnixNano - Output.GeneratedAt.
 	Coordinator       string         `json:"coordinator"`
 	Contract          string         `json:"contract"`
 	Role              string         `json:"role"`
@@ -127,13 +130,26 @@ type Event struct {
 }

 type Output struct {
-	Coordinators []string  `json:"coordinators"`
-	Contracts    []string  `json:"contracts"`
-	Events       []Event   `json:"events"`
-	Diversity    Diversity `json:"diversity"`
-	Determinism  Determ    `json:"determinism"`
-	Learning     Learning  `json:"learning"`
-	GeneratedAt  time.Time `json:"generated_at"`
+	Coordinators []string      `json:"coordinators"`
+	Contracts    []string      `json:"contracts"`
+	Events       []Event       `json:"events"`
+	Diversity    Diversity     `json:"diversity"`
+	Determinism  Determ        `json:"determinism"`
+	Learning     Learning      `json:"learning"`
+	PhaseTimings []PhaseTiming `json:"phase_timings,omitempty"` // OPEN #4: real wall-clock per phase
+	TotalElapsedMs int64       `json:"total_elapsed_ms,omitempty"`
+	GeneratedAt  time.Time     `json:"generated_at"`
+}
+
+// PhaseTiming records the real wall-clock duration of a single phase.
+// Closes OPEN #4's "real-time wall-clock for the stress harness" —
+// operators reading the JSON now see actual phase durations alongside
+// the operational-narrative Hour labels. Hour is a fictional time
+// (the simulated 48-hour mock); ElapsedMs is the real one.
+type PhaseTiming struct {
+	Phase     string    `json:"phase"`
+	StartUTC  time.Time `json:"start_utc"`
+	ElapsedMs int64     `json:"elapsed_ms"`
 }

 // Diversity = how distinct are top-K worker sets across (coord,
@@ -223,6 +239,10 @@ func main() {
 	ctx := context.Background()
 	_ = ctx

+	// runStart pinned BEFORE startPhase is defined so the closure
+	// can reference it; output.GeneratedAt picks it up later.
+	runStart := time.Now().UTC()
+
 	// Optional Langfuse client. Best-effort: missing env file or
 	// unreachable Langfuse just means traces don't go anywhere; the
 	// run still proceeds.
@@ -255,7 +275,33 @@ func main() {
 	// Subsequent emitSpan calls nest under it. Idempotent — returns
 	// "" when Langfuse isn't configured so callers don't need nil
 	// checks.
+	// Per-phase wall-clock tracker (OPEN #4 closure). Operators see
+	// "phase 2: surge took 8.3s" as the run progresses, not only in
+	// the final JSON. Used by both the stdout log line and the
+	// per-phase EndTime in the Langfuse span (close-on-next-startPhase).
+	var (
+		currentPhaseName  string
+		currentPhaseStart time.Time
+		phaseTimings      []PhaseTiming
+	)
+	closePhase := func() {
+		if currentPhaseName == "" {
+			return
+		}
+		elapsed := time.Since(currentPhaseStart)
+		phaseTimings = append(phaseTimings, PhaseTiming{
+			Phase:     currentPhaseName,
+			StartUTC:  currentPhaseStart.UTC(),
+			ElapsedMs: elapsed.Milliseconds(),
+		})
+		log.Printf("[stress] phase %s done — %.2fs (T+%.1fs)", currentPhaseName, elapsed.Seconds(), time.Since(runStart).Seconds())
+		currentPhaseName = ""
+	}
 	startPhase := func(name string, hour int, meta map[string]any) {
+		closePhase()
+		currentPhaseName = name
+		currentPhaseStart = time.Now()
+		log.Printf("[stress] phase %s starting (T+%.1fs)", name, time.Since(runStart).Seconds())
 		if lf == nil {
 			return
 		}
@@ -267,7 +313,7 @@ func main() {
 			TraceID:   runTraceID,
 			Name:      name,
 			Metadata:  spanMeta,
-			StartTime: time.Now(),
+			StartTime: currentPhaseStart,
 		})
 	}
 	// emitSpan records one span as a child of the current phase span.
@@ -320,7 +366,7 @@ func main() {
 	output := Output{
 		Coordinators: []string{"alice", "bob", "carol"},
 		Contracts:    []string{contracts[0].Name, contracts[1].Name, contracts[2].Name},
-		GeneratedAt:  time.Now().UTC(),
+		GeneratedAt:  runStart,
 	}

 	log.Printf("[stress] 3 coords, 3 contracts, k=%d, corpora=%v", *k, corpora)
@@ -753,6 +799,11 @@ func main() {
 	log.Printf("[stress] phase 7: diversity analysis")
 	output.Diversity = computeDiversity(output.Events)

+	// Close the final phase + record total elapsed (OPEN #4 closure).
+	closePhase()
+	output.PhaseTimings = phaseTimings
+	output.TotalElapsedMs = time.Since(runStart).Milliseconds()
+
 	// ── write ───────────────────────────────────────────────────
 	if err := os.MkdirAll(filepath.Dir(*out), 0o755); err != nil {
 		log.Fatalf("mkdir: %v", err)