Preserve cancellation in aggregate smoke tests

This commit is contained in:
2026-11-07 15:36:23 +00:00
parent a806a9a62e
commit b66d265032
4 changed files with 98 additions and 7 deletions

View File

@@ -1,6 +1,6 @@
# Consolidated audit catalog # Consolidated audit catalog
Status: **27 distinct open findings: five P2 and 22 P3**, consolidated from 30 Status: **26 distinct open findings: four P2 and 22 P3**, consolidated from 30
numbered entries in seven audit documents. Reviewed audit state: numbered entries in seven audit documents. Reviewed audit state:
`db732e3a1de55c844f77870e1c6db683c2685d5d`. `db732e3a1de55c844f77870e1c6db683c2685d5d`.
@@ -9,7 +9,7 @@ decisions. The linked original entries retain code locations, evidence,
contract analysis, and detailed contract analysis, and detailed
remediation options. Consolidation alone does not resolve a finding or select a remediation options. Consolidation alone does not resolve a finding or select a
proposed contract change. Explicit decisions recorded below supersede the proposed contract change. Explicit decisions recorded below supersede the
original entries' alternatives; all findings remain open until remediated. original entries' alternatives; findings remain open until remediated.
Original priorities are retained. Within those priorities, incorrect outcomes Original priorities are retained. Within those priorities, incorrect outcomes
and request isolation lead; blocked validation, and request isolation lead; blocked validation,
resource retention, and consumer contracts precede scaling and maintainability resource retention, and consumer contracts precede scaling and maintainability
@@ -19,7 +19,7 @@ The [remediation analysis](remediation-analysis.md) records the completed design
review and its rationale. All choices identified for discussion have been review and its rationale. All choices identified for discussion have been
resolved; no further product or architectural decision is required before resolved; no further product or architectural decision is required before
implementation planning. Accepted conclusions are recorded in this catalog; implementation planning. Accepted conclusions are recorded in this catalog;
all 27 findings still require remediation and validation. the remaining findings still require remediation and validation.
Three pairs describe the same underlying issue: correctness C2/concurrency N2, Three pairs describe the same underlying issue: correctness C2/concurrency N2,
correctness C4/concurrency N3, and complexity M6/responsibility R1. Other correctness C4/concurrency N3, and complexity M6/responsibility R1. Other
@@ -38,6 +38,12 @@ these boundaries.
## F01 — P2: Aggregate smoke testing can stop on cancellation and return success ## F01 — P2: Aggregate smoke testing can stop on cancellation and return success
**Complete.** The later cancellation checkpoint retains a cause when work is
omitted, without duplicating a cause already recorded. Focused aggregate and
cancellation tests pass in the aligned workspace; the new progress-delivery
regression failed against the original behavior for both successful and failed
provider calls, and also covers cancellation after the last completed check.
In ckprompt's `runSmokeAggregate`, cancellation triggered during progress In ckprompt's `runSmokeAggregate`, cancellation triggered during progress
delivery can stop the remaining smoke kinds without entering the operation delivery can stop the remaining smoke kinds without entering the operation
cause. A partial aggregate can consequently return status 0; cancellation can cause. A partial aggregate can consequently return status 0; cancellation can

View File

@@ -1,10 +1,10 @@
# Audit remediation implementation plan # Audit remediation implementation plan
Status: planned; no stage implemented by this document. Status: in progress; Stage 1 complete.
Baseline: `d492c7747175c01a53261206c2b4356e53453621`. Source remains the audited Baseline: `d492c7747175c01a53261206c2b4356e53453621`. Before this sprint,
implementation; intervening changes recorded decisions and retired the completed intervening changes recorded decisions and retired the completed older
older remediation. The [roadmap](audit-remediation.md) owns the target and scope, remediation. The [roadmap](audit-remediation.md) owns the target and scope,
the [catalog](audit/catalog.md) owns accepted decisions and finding status, and the [catalog](audit/catalog.md) owns accepted decisions and finding status, and
its original audit links own defect evidence. The its original audit links own defect evidence. The
[analysis](audit/remediation-analysis.md) explains the selected designs. [analysis](audit/remediation-analysis.md) explains the selected designs.
@@ -107,6 +107,13 @@ not converted into a passed check or a request for per-stage approval.
## Stage 1 — Retain aggregate smoke cancellation ## Stage 1 — Retain aggregate smoke cancellation
**Complete.** The progress-delivery cancellation regression reproduced both
the false success and missing joined cause before the fix. The focused aligned
workspace checks passed: `go test -mod=readonly -run 'TestSmokeAggregate|TestSmokeTestCancellation' ./integrations/ckprompt`
with Go 1.27.0, `GOTOOLCHAIN=local`, and `GOPROXY=off`. Tests cover partial
results with and without a provider error, retained diagnostics and explanation,
omitted callbacks, and cancellation after the final check.
Findings: F01. Primary source: `integrations/ckprompt/smoke_command.go`. Findings: F01. Primary source: `integrations/ckprompt/smoke_command.go`.
1. In runSmokeAggregate, retain cancellation observed at the later checkpoint 1. In runSmokeAggregate, retain cancellation observed at the later checkpoint

View File

@@ -3,9 +3,11 @@ package ckprompt
import ( import (
"bytes" "bytes"
"context" "context"
"encoding/json"
"errors" "errors"
"io" "io"
"reflect" "reflect"
"strings"
"testing" "testing"
"go.maximumdirect.net/ckconfig" "go.maximumdirect.net/ckconfig"
@@ -109,6 +111,79 @@ func TestSmokeAggregateCoordinatesConcreteRequestsAndRetainsErrors(t *testing.T)
} }
} }
type cancelProgressWriter struct {
bytes.Buffer
cancel context.CancelFunc
}
func (writer *cancelProgressWriter) Write(p []byte) (int, error) {
writer.cancel()
return writer.Buffer.Write(p)
}
func TestSmokeAggregateCancellationDuringProgress(t *testing.T) {
providerErr := errors.New("provider failed")
for _, test := range []struct {
name string
err error
cancelAt int
}{
{name: "partial success", cancelAt: 1},
{name: "partial provider failure", err: providerErr, cancelAt: 1},
{name: "last completed test", cancelAt: 4},
} {
t.Run(test.name, func(t *testing.T) {
ctx, cancel := context.WithCancel(context.Background())
defer cancel()
var kinds []promptkit.SmokeTestKind
app := newPromptTestApp(t, Command(Suite[struct{}, struct{}]{
SmokeTest: func(_ context.Context, request SmokeRequest, _, _ struct{}) (commandkit.Acquired[*promptkit.SmokeTestReport], error) {
kinds = append(kinds, request.Request.Kind)
acquired := commandkit.Acquired[*promptkit.SmokeTestReport]{Value: &promptkit.SmokeTestReport{Kind: request.Request.Kind, Completed: true}}
if len(kinds) == test.cancelAt {
acquired.Diagnostics = []commandkit.Diagnostic{{Severity: commandkit.SeverityProgress, Write: func(w io.Writer) error { _, err := io.WriteString(w, "progress evidence\n"); return err }}}
acquired.Explanation = commandkit.Explanation{Message: "provider explanation"}
return acquired, test.err
}
return acquired, nil
},
}))
var output bytes.Buffer
diagnostics := &cancelProgressWriter{cancel: cancel}
outcome := app.Execute(ctx, []string{"promptkit", "test", "all", "--progress", "--format=json"}, commandkit.Streams{Out: &output, Err: diagnostics})
if len(kinds) != test.cancelAt || !strings.Contains(diagnostics.String(), "progress evidence") {
t.Fatalf("kinds=%v diagnostics=%q", kinds, diagnostics.String())
}
if test.cancelAt < 4 {
if outcome.Kind != commandkit.OutcomeOperational || !errors.Is(outcome.Cause, context.Canceled) || (test.err != nil && (!errors.Is(outcome.Cause, test.err) || !strings.Contains(diagnostics.String(), "provider explanation"))) {
t.Fatalf("partial outcome=%#v diagnostics=%q", outcome, diagnostics.String())
}
} else if outcome.Kind != commandkit.OutcomeSuccess || outcome.Cause != nil {
t.Fatalf("completed outcome=%#v", outcome)
}
var report struct {
Tests []struct {
Kind string `json:"kind"`
Report *struct {
Completed bool `json:"completed"`
} `json:"report"`
} `json:"tests"`
}
if err := json.Unmarshal(output.Bytes(), &report); err != nil {
t.Fatal(err)
}
if len(report.Tests) != len(kinds) {
t.Fatalf("report=%s", output.String())
}
for i, entry := range report.Tests {
if entry.Kind != string(kinds[i]) || entry.Report == nil || !entry.Report.Completed {
t.Fatalf("report entry=%#v", entry)
}
}
})
}
}
func TestSmokeAggregateCancellationRetainsPartialResults(t *testing.T) { func TestSmokeAggregateCancellationRetainsPartialResults(t *testing.T) {
ctx, cancel := context.WithCancel(context.Background()) ctx, cancel := context.WithCancel(context.Background())
t.Cleanup(cancel) t.Cleanup(cancel)

View File

@@ -173,6 +173,9 @@ func runSmokeAggregate[I, S any](invocation commandkit.Invocation, suite Suite[I
} }
} }
if err := promptKitContextError(invocation.Context); err != nil { if err := promptKitContextError(invocation.Context); err != nil {
if len(results) < len(kinds) && !errors.Is(errors.Join(operationErrors...), err) {
operationErrors = append(operationErrors, err)
}
break break
} }
} }