mesh-host/internal/link/session_test.go

package link

import (
	"context"
	"crypto/ed25519"
	"errors"
	"fmt"
	"os"
	"strings"
	"sync"
	"testing"
	"time"

	"git.novox.be/novox/mesh-contracts/go/seats/nodeengine"
	"git.novox.be/novox/mesh-delivery/modules/mesh-delivery-builder/builds"
	"github.com/nats-io/nats.go"
)

// quietLink is a link with no broker behind it.
type quietLink struct {
	lost chan error
}

func newQuietLink() *quietLink { return &quietLink{lost: make(chan error, 1)} }

func (l *quietLink) Lost() <-chan error { return l.lost }
func (l *quietLink) Close()             {}

// runningLink hears the node seat's running events: the last of each subject first, then the durable consumer's.
type runningLink struct {
	*quietLink
	first, after [][]byte
}

func (l *runningLink) FirstRunning(_ context.Context, first func([]byte)) error {
	for _, raw := range l.first {
		first(raw)
	}
	return nil
}

func (l *runningLink) HearRunning(_ string, heard func([]byte)) (func(), error) {
	for _, raw := range l.after {
		heard(raw)
	}
	return func() {}, nil
}

func (l *runningLink) HearBuildEnds(_ context.Context, heard func([]byte), readEnded func()) error {
	for _, raw := range l.first {
		heard(raw)
	}
	readEnded()
	for _, raw := range l.after {
		heard(raw)
	}
	return nil
}

func aMember(t *testing.T) (Membership, ed25519.PrivateKey) {
	t.Helper()
	public, private, err := ed25519.GenerateKey(nil)
	if err != nil {
		t.Fatal(err)
	}
	return Membership{Node: "n1", Signer: public}, private
}

// Facts made with no link open are not kept, and nothing is said: the next link says them as they are then.
func TestFactsWithNoLinkOpenAreNotSaid(t *testing.T) {
	m, _ := aMember(t)
	l := newQuietLink()
	s := &Session{Membership: m, Timeout: time.Second}
	if s.SayFacts(context.Background(), Report{}) {
		t.Fatal("facts were said with no link")
	}
	s.Attach(context.Background(), l)
	s.Detach(l)
	if s.Linked() || s.SayFacts(context.Background(), Report{}) {
		t.Fatal("facts were said on a link that was let go")
	}

	// Design 47 G §199, §556: with no link the running events are not followed; with one, the last message of each subject
	// comes first, the read ends, then the durable consumer's, and a body that is no running event is passed over.
	if err := s.FollowRunning(context.Background(), func(nodeengine.Running) {}, func() {}); !errors.Is(err, ErrNoLink) {
		t.Fatalf("followed with no link: %v", err)
	}
	rl := &runningLink{quietLink: newQuietLink(),
		first: [][]byte{[]byte(`{"node":"a","module":"x","fingerprint":"f1"}`), []byte(`not json`), []byte(`{"module":"x"}`)},
		after: [][]byte{[]byte(`{"node":"b","module":"y","stopped":true}`)}}
	s.Attach(context.Background(), rl)
	var order []string
	ctx, stop := context.WithCancel(context.Background())
	err := s.FollowRunning(ctx, func(ev nodeengine.Running) { order = append(order, ev.Node+"/"+ev.Module) },
		func() { order = append(order, "read"); stop() })
	if err != nil || strings.Join(order, " ") != "a/x b/y read" {
		t.Fatalf("followed %v: %v, want a/x b/y read", order, err)
	}

	// Design 47 G A-293: the builder's build ends likewise, the last of each subject first, then the read's end, then the
	// rest, a body that is no end of a module at a fingerprint passed over.
	if err := (&Session{Membership: m}).FollowBuildEnds(context.Background(), func(builds.BuildEnded) {}, func() {}); !errors.Is(err, ErrNoLink) {
		t.Fatalf("build ends followed with no link: %v", err)
	}
	rl.first = [][]byte{[]byte(`{"module":"x","fingerprint":"f1","outcome":"failed","error":"no go"}`), []byte(`not json`), []byte(`{"module":"x"}`)}
	rl.after = [][]byte{[]byte(`{"module":"y","fingerprint":"f2","outcome":"done","branch":"b"}`)}
	order = nil
	err = s.FollowBuildEnds(context.Background(), func(e builds.BuildEnded) {
		order = append(order, e.Module+"@"+e.Fingerprint+":"+e.Outcome+":"+e.Error+":"+e.Branch)
	}, func() { order = append(order, "read") })
	if err != nil || strings.Join(order, " ") != "x@f1:failed:no go: read y@f2:done::b" {
		t.Fatalf("build ends followed %v: %v", order, err)
	}
}

// A link opening says the node's facts, off the link's own loop.
func TestALinkOpeningSaysTheFacts(t *testing.T) {
	m, _ := aMember(t)
	l := newQuietLink()
	opened := make(chan struct{}, 1)
	s := &Session{Membership: m, Timeout: time.Second, OnLink: func(context.Context) { opened <- struct{}{} }}
	s.Attach(context.Background(), l)
	select {
	case <-opened:
	case <-time.After(2 * time.Second):
		t.Fatal("a link opening did not ask for the facts")
	}
}

// aBus is a connection to a real server, because some claims here are about one: set MESH_TEST_NATS to a server with
// JetStream, or the test is skipped.
func aBus(t *testing.T) (*nats.Conn, nats.JetStreamContext) {
	t.Helper()
	url := os.Getenv("MESH_TEST_NATS")
	if url == "" {
		t.Skip("MESH_TEST_NATS unset")
	}
	conn, err := nats.Connect(url)
	if err != nil {
		t.Fatal(err)
	}
	t.Cleanup(conn.Close)
	js, err := conn.JetStream()
	if err != nil {
		t.Fatal(err)
	}
	return conn, js
}

// eventuallyHeld waits for a condition the bus reaches on its own time.
func eventuallyHeld(t *testing.T, what string, ok func() bool) {
	t.Helper()
	deadline := time.Now().Add(5 * time.Second)
	for !ok() {
		if time.Now().After(deadline) {
			t.Fatal(what)
		}
		time.Sleep(5 * time.Millisecond)
	}
}

// refusingLink is a quiet link the bus refused subscriptions to.
type refusingLink struct {
	*quietLink
	refused chan string
}

func (l *refusingLink) Refused() <-chan string { return l.refused }

// The sweep of 2026-10-30: every engine was refused its node seat's subscriptions at 12:10, by a user list composed
// without them, and nats asks a refused subscription again only on a new connection; the seat answered nobody all
// afternoon. A refused subscription is read from the server's words, said, and the link opened again after a pause
// that grows while it is refused again; a refused publish is not one.
func TestASubscriptionTheBusRefusedIsAskedAgainOnANewLink(t *testing.T) {
	// The server's words, as the client hands them on (journal of nox-mesh-host on novox, 2026-10-30 12:10:44).
	for _, row := range []struct {
		err  error
		want string
	}{
		{fmt.Errorf("%w: %s", nats.ErrPermissionViolation,
			`Permissions Violation for Subscription to "mesh.seat.node-engine.tool.root-free.novox"`),
			"mesh.seat.node-engine.tool.root-free.novox"},
		{errors.Join(fmt.Errorf("%w: %s", nats.ErrPermissionViolation,
			`Permissions Violation for Subscription to "$SRV.PING.node-engine"`), errors.New("stopping service: x")),
			"$SRV.PING.node-engine"},
		{fmt.Errorf("%w: %s", nats.ErrPermissionViolation,
			`Permissions Violation for Publish to "$JS.API.STREAM.INFO.KV_mesh-host_health"`), ""},
		{errors.New("nats: timeout"), ""},
	} {
		got, ok := RefusedSubscription(row.err)
		if got != row.want || ok != (row.want != "") {
			t.Errorf("%v reads as %q, %v", row.err, got, ok)
		}
		l := &natsLink{refused: make(chan string, 1)}
		l.asyncError(nil, nil, row.err)
		if heard := len(l.refused); heard != map[bool]int{true: 1}[row.want != ""] {
			t.Errorf("%v handed %d refusals on", row.err, heard)
		}
	}

	m, _ := aMember(t)
	q := &Session{Membership: m, Timeout: time.Second, RefusedPause: 50 * time.Millisecond}
	var said []string
	var mu sync.Mutex
	say := func(s string) { mu.Lock(); said = append(said, s); mu.Unlock() }
	for _, want := range []time.Duration{50 * time.Millisecond, 100 * time.Millisecond} {
		l := &refusingLink{quietLink: newQuietLink(), refused: make(chan string, 2)}
		l.refused <- "mesh.seat.node-engine.tool.root-free.n1"
		l.refused <- "$SRV.PING"
		began := time.Now()
		done := make(chan error, 1)
		go func() { done <- serve(context.Background(), l, m, q, say, time.Second) }()
		select {
		case err := <-done:
			if err == nil || !strings.Contains(err.Error(), "refused 2 of this node's subscriptions, "+
				"mesh.seat.node-engine.tool.root-free.n1 first") {
				t.Fatalf("the link ended with %v", err)
			}
			if took := time.Since(began); took < want {
				t.Errorf("the link was dropped after %s, before its pause of %s", took, want)
			}
		case <-time.After(5 * time.Second):
			t.Fatal("a link the bus refused a subscription was kept, and the subscription never asked again")
		}
	}
	mu.Lock()
	defer mu.Unlock()
	if n := strings.Count(strings.Join(said, "\n"), "the bus refused this node a subscription to "+
		"mesh.seat.node-engine.tool.root-free.n1"); n != 2 {
		t.Errorf("the refusal was said %d times in two links: %q", n, said)
	}
	if q.refusalsHeld(); q.refusedPause() != 50*time.Millisecond {
		t.Error("a link that held its subscriptions did not start from the first pause again")
	}
}