E-2: a heard event keeps its own time, and a refused wait is never a hot loop (issue 1966)

HearFirst reads the event's x-time as Published, notes a subscription the bus
refuses while the others go on waiting, and answers a wait refused whole only
after its timeout, in words naming the grants part's grant of every event
subject; mesh_wait lists the refused subjects in its answer.

Builds: design 34 every-event §12, §16, §18, §19 (bus/wait_test.go)

Claude-Session: https://claude.ai/code/session_01NcbgfGKaxw984XyAPfLm4h
This commit is contained in:
2026-11-12 05:03:13 +01:00
parent 3f89c96d20
commit cb47de8517
3 changed files with 119 additions and 58 deletions
+47 -27
View File
@@ -4,6 +4,7 @@ import (
"context"
"encoding/json"
"fmt"
"slices"
"strings"
"time"
@@ -23,31 +24,42 @@ type WaitIssued struct {
Subject string `json:"subject"`
}
// Heard is one event a wait heard: the subject it arrived on, its body and when it arrived.
// Heard is one event a wait heard: the subject it arrived on, its body, when it arrived, and when it was published, the
// event's own header `x-time` as its emitter wrote it, empty for an event that carries none (design 34 every-event §12).
type Heard struct {
Subject string
Body json.RawMessage
At time.Time
Subject string
Body json.RawMessage
At time.Time
Published string
}
// HearFirst subscribes the subjects for this call alone and answers the first message whose body match
// accepts (every one, when match is nil), or ok false once `within` has passed. Every subscription is
// closed before it returns. A subscription the bus refuses is said as an error naming the subject, at once;
// a caller gone — ctx done — ends the wait at once with ctx's error, so nothing waits on for nobody.
func (c *Conn) HearFirst(ctx context.Context, subjects []string, within time.Duration, match func(json.RawMessage) bool) (heard Heard, ok bool, err error) {
// closed before it returns. A subscription the bus refuses is noted and the others go on waiting; refused lists
// those noted by the time it answers (design 34 every-event §16). When the bus refused every subscription of the
// call, it still waits out `within` and then answers the refusal, so a caller looping on it asks once a timeout and
// never at once again. A caller gone — ctx done — ends the wait at once with ctx's error, so nothing waits on for
// nobody.
func (c *Conn) HearFirst(ctx context.Context, subjects []string, within time.Duration, match func(json.RawMessage) bool) (heard Heard, ok bool, refused []string, err error) {
if len(subjects) == 0 {
return Heard{}, false, fmt.Errorf("a wait names at least one subject")
return Heard{}, false, nil, fmt.Errorf("a wait names at least one subject")
}
var unique []string
for _, subject := range subjects {
if !slices.Contains(unique, subject) {
unique = append(unique, subject)
}
}
got := make(chan Heard, 1)
refused := make(chan string, len(subjects))
refusals := make(chan string, len(unique))
var subs []*nats.Subscription
defer func() {
for _, s := range subs {
_ = s.Unsubscribe()
}
c.mu.Lock()
for _, subject := range subjects {
delete(c.waits[subject], refused)
for _, subject := range unique {
delete(c.waits[subject], refusals)
if len(c.waits[subject]) == 0 {
delete(c.waits, subject)
}
@@ -58,14 +70,14 @@ func (c *Conn) HearFirst(ctx context.Context, subjects []string, within time.Dur
if c.waits == nil {
c.waits = map[string]map[chan string]bool{}
}
for _, subject := range subjects {
for _, subject := range unique {
if c.waits[subject] == nil {
c.waits[subject] = map[chan string]bool{}
}
c.waits[subject][refused] = true
c.waits[subject][refusals] = true
}
c.mu.Unlock()
for _, subject := range subjects {
for _, subject := range unique {
sub, err := c.nc.Subscribe(subject, func(msg *nats.Msg) {
body := json.RawMessage(msg.Data)
if !json.Valid(body) {
@@ -76,31 +88,39 @@ func (c *Conn) HearFirst(ctx context.Context, subjects []string, within time.Dur
return
}
select {
case got <- Heard{Subject: msg.Subject, Body: body, At: time.Now().UTC()}:
case got <- Heard{Subject: msg.Subject, Body: body, At: time.Now().UTC(), Published: msg.Header.Get("x-time")}:
default:
}
})
if err != nil {
return Heard{}, false, fmt.Errorf("could not wait on %s: %w", subject, err)
return Heard{}, false, nil, fmt.Errorf("could not wait on %s: %w", subject, err)
}
subs = append(subs, sub)
}
// The bus has every subscription — or has said which it refused — once this returns.
if err := c.nc.FlushTimeout(5 * time.Second); err != nil {
return Heard{}, false, fmt.Errorf("the bus did not take the wait: %w", err)
return Heard{}, false, nil, fmt.Errorf("the bus did not take the wait: %w", err)
}
timer := time.NewTimer(within)
defer timer.Stop()
select {
case h := <-got:
return h, true, nil
case subject := <-refused:
return Heard{}, false, fmt.Errorf("the bus refused this account a subscription on %s: it is not in the "+
"account's wait grant (novox/hq ADR 0367 §2)", subject)
case <-timer.C:
return Heard{}, false, nil
case <-ctx.Done():
return Heard{}, false, ctx.Err()
for {
select {
case h := <-got:
return h, true, refused, nil
case subject := <-refusals:
if !slices.Contains(refused, subject) {
refused = append(refused, subject)
}
case <-timer.C:
if len(refused) == len(unique) {
return Heard{}, false, refused, fmt.Errorf("the bus refused this account a subscription on %s: the "+
"grants part grants this machine's tool runner every event subject (novox/hq ADR 0670 §1), so its "+
"grant is not composed yet; read the state from its owner meanwhile", strings.Join(refused, ", "))
}
return Heard{}, false, refused, nil
case <-ctx.Done():
return Heard{}, false, refused, ctx.Err()
}
}
}
+58 -24
View File
@@ -2,22 +2,24 @@ package bus_test
import (
"context"
"encoding/json"
"errors"
"strings"
"testing"
"time"
"github.com/nats-io/nats-server/v2/server"
"github.com/nats-io/nats.go"
"github.com/novox/mesh-tools/node-tools/internal/bus"
mt "github.com/novox/mesh-tools/node-tools/internal/meshtest"
)
// novox/hq ADR 0367 §2, §5: the bus refuses the tool runner's account any subject its wait grant does not
// name — another asker's `decided` among them — and a wait on one says the bus refused it, at once.
// novox/hq ADR 0670 §1 (design 34 every-event §12, §16): a subscription the bus refuses is noted and the others go on
// waiting, a heard event carries its own time; a wait whose every subscription is refused answers only after its
// timeout, in words naming the grants part's grant of every event subject, never a wait grant (E-2).
func TestTheBusRefusesAWaitOutsideTheGrant(t *testing.T) {
granted := "mesh.seat.operator-channel.event.decided.claude-code"
other := "mesh.seat.operator-channel.event.decided.shop"
s, err := server.NewServer(&server.Options{Host: "127.0.0.1", Port: server.RANDOM_PORT, NoLog: true, NoSigs: true,
Users: []*server.User{{Username: "desk.node-tools", Password: "pw", Permissions: &server.Permissions{
Publish: &server.SubjectPermission{Allow: []string{"$JS.API.DIRECT.GET.>", "mesh.>"}},
@@ -33,33 +35,65 @@ func TestTheBusRefusesAWaitOutsideTheGrant(t *testing.T) {
t.Cleanup(s.Shutdown)
c := dial(t, s.ClientURL(), "node-tools", "desk", "desk.node-tools")
// Every subscription refused: answered after the timeout, never at once, in §16's words.
start := time.Now()
_, ok, err := c.HearFirst(context.Background(), []string{"mesh.seat.operator-channel.event.decided.shop"}, 5*time.Second, nil)
if ok || err == nil || !strings.Contains(err.Error(), "refused") {
t.Errorf("a wait on another asker's decided: heard %v, %v", ok, err)
_, ok, refused, err := c.HearFirst(context.Background(), []string{other}, 300*time.Millisecond, nil)
if took := time.Since(start); took < 300*time.Millisecond || took > 3*time.Second {
t.Errorf("a wait refused whole answered after %s, want its timeout of 300ms", took)
}
if time.Since(start) > 3*time.Second {
t.Errorf("the refusal took %s to be said", time.Since(start))
if ok || err == nil || !strings.Contains(err.Error(), "the bus refused this account a subscription on "+other) ||
!strings.Contains(err.Error(), "every event subject (novox/hq ADR 0670 §1)") ||
strings.Contains(err.Error(), "wait grant") || strings.Contains(err.Error(), "0367") ||
len(refused) != 1 || refused[0] != other {
t.Errorf("a wait on another asker's decided: heard %v, refused %v, %v", ok, refused, err)
}
done := make(chan bus.Heard, 1)
go func() {
h, ok, err := c.HearFirst(context.Background(), []string{granted}, 5*time.Second, func(body json.RawMessage) bool { return true })
if ok && err == nil {
done <- h
}
}()
time.Sleep(300 * time.Millisecond)
if err := c.Say(granted, []byte(`{"ask":"a1"}`)); err != nil {
// One refused beside one granted: the granted one hears, with the event's own time, and the refused one is listed.
pub, err := nats.Connect(s.ClientURL(), nats.UserInfo("desk.node-tools", "pw"))
if err != nil {
t.Fatal(err)
}
select {
case h := <-done:
if h.Subject != granted || string(h.Body) != `{"ask":"a1"}` {
t.Errorf("heard %+v", h)
t.Cleanup(pub.Close)
for _, published := range []string{"2026-11-12T03:00:00.5Z", ""} {
type result struct {
h bus.Heard
refused []string
err error
}
done := make(chan result, 1)
go func() {
h, ok, refused, err := c.HearFirst(context.Background(), []string{other, granted}, 5*time.Second, nil)
if !ok && err == nil {
err = errors.New("timed out")
}
done <- result{h, refused, err}
}()
// The event is said again until the wait, subscribed in its own time, hears it.
say := time.NewTicker(20 * time.Millisecond)
var r result
heard:
for {
select {
case r = <-done:
break heard
case <-say.C:
msg := nats.NewMsg(granted)
msg.Data = []byte(`{"ask":"a1"}`)
if published != "" {
msg.Header.Set("x-time", published)
}
if err := pub.PublishMsg(msg); err != nil {
t.Fatal(err)
}
}
}
say.Stop()
if r.err != nil || r.h.Subject != granted || string(r.h.Body) != `{"ask":"a1"}` || r.h.Published != published {
t.Errorf("heard %+v, %v; want published %q", r.h, r.err, published)
}
if len(r.refused) > 1 || len(r.refused) == 1 && r.refused[0] != other {
t.Errorf("refused %v, want at most %s", r.refused, other)
}
case <-time.After(5 * time.Second):
t.Fatal("a wait on the granted subject heard nothing")
}
}
@@ -70,7 +104,7 @@ func TestAWaitEndsWhenItsCallerIsGone(t *testing.T) {
ctx, cancel := context.WithCancel(context.Background())
time.AfterFunc(200*time.Millisecond, cancel)
start := time.Now()
_, ok, err := c.HearFirst(ctx, []string{"mesh.mod.mesh-knowledge.event.reviewed"}, time.Minute, nil)
_, ok, _, err := c.HearFirst(ctx, []string{"mesh.mod.mesh-knowledge.event.reviewed"}, time.Minute, nil)
if ok || !errors.Is(err, context.Canceled) {
t.Errorf("a cancelled wait answered %v, %v", ok, err)
}
+14 -7
View File
@@ -86,18 +86,25 @@ func (s *Surface) waitAnswer(ctx context.Context, args map[string]any) map[strin
family[subject] = name
}
heard, ok, err := s.conn.HearFirst(ctx, subjects, timeout, func(body json.RawMessage) bool { return bodyMatches(body, filter) })
heard, ok, refused, err := s.conn.HearFirst(ctx, subjects, timeout, func(body json.RawMessage) bool { return bodyMatches(body, filter) })
if err != nil {
return failure("mesh_wait could not wait: " + err.Error())
}
var answer map[string]any
if !ok {
return answerText(map[string]any{"timed-out": true, "waited": timeout.String(), "subjects": names,
"read": "nothing matching was heard; read the state once from the verb that owns it, then wait again"})
answer = map[string]any{"timed-out": true, "waited": timeout.String(), "subjects": names,
"read": "nothing matching was heard; read the state once from the verb that owns it, then wait again"}
} else {
answer = map[string]any{
"heard": map[string]any{"subject": family[heard.Subject], "body": heard.Body, "at": heard.At.Format(time.RFC3339Nano)},
"read": "an event only wakes you: read the state from the verb that owns it before acting (novox/hq ADR 0367 §4)",
}
}
return answerText(map[string]any{
"heard": map[string]any{"subject": family[heard.Subject], "body": heard.Body, "at": heard.At.Format(time.RFC3339Nano)},
"read": "an event only wakes you: read the state from the verb that owns it before acting (novox/hq ADR 0367 §4)",
})
// A subscription the bus refused while others stood is said (design 34 every-event §16).
if len(refused) > 0 {
answer["refused"] = refused
}
return answerText(answer)
}
// waitNames is the families a call names: a list of names, or one name.