feat: panadapter spots that read, a CI-V link that survives, no auto-call

Panadapter spots are now worth reading. The comment carries the spotter, the
entity and the status in DXHunter's own shape — "CQ up 2 [F4BPO] [Franz Josef
Land] [New Slot]" — which needed two things nothing documents: SmartSDR splits
its command line on SPACES, so the words ran together until every space became
non-breaking; and it truncates past ~60 characters, so the cluster's own words
are trimmed first and the three brackets always survive. RBN column padding is
collapsed on the way in, or a preserved run of spaces opened a gap wide enough
to push the rest off screen.

"Already worked" means the CALLSIGN is in the log, not the entity: saying it of
a station never contacted was simply wrong. Each status can also be kept off the
panadapter entirely, and the WSJT-X decode spots obey the same switches — the
palette governs the panadapter, not one of the two things that feed it.

And the radio is no longer hammered: a spot whose frequency, colour and comment
are unchanged is not removed and redrawn. A busy skimmer feed re-spots the same
station every few seconds; one two-minute session sent 2128 adds, 88 of them for
a single callsign, and the display did not move a pixel for any of them.

CI-V, from an IC-7850 that kept killing JTDX: a reply the rig sent to another
controller on the same bus is no longer taken for ours, and a set_ptt, set_freq
or set_mode whose acknowledgement goes missing is verified by reading the rig
back instead of being reported as a failure. WSJT-X and JTDX answer a failed
command with a Rig Control Error and drop the link mid-over — 98 keyings, 6 lost
acknowledgements, 2 dropped connections in one session. The check waits 700 ms,
not the poll's 150: the rig has just failed to answer twice because it was
retuning, and a short probe would fail for the same reason.

Auto-call is withdrawn — it duplicated DXHunter, which already answers decodes,
and two programs deciding that from one shack key over each other. The library
is kept whole and dormant; a guard in App.tsx makes sure a stored preference
cannot key a transmitter whose switch no longer exists.

Also:
  - the log rotates while running, not only at startup: the CI-V trace left on
    wrote 416 MB and nothing would have stopped it before the disk did. Closing
    it now releases the crash file too — the runtime keeps its own duplicate.
  - the interface zoom announces itself, with a badge, a click back to 100% and
    a View menu; Ctrl+wheel and Ctrl+0 always worked and nothing said so.
  - no more elastic bounce, and no swipe-to-navigate out of the app.
  - Edit QSO: your own TX power and the contacted station's extended locator
    were saved and written back with no box to set them.
  - FT decodes: continents are a multiple choice; a compound MSHV message that
    answers two stations in one line is recognised as addressed to you.

internal/applog/applog.go

@@ -20,9 +20,15 @@ var (
 	mu        sync.Mutex
 	file      *os.File
 	path      string
+	written   int64    // bytes in the CURRENT file, for the size check in Printf
 	crashFile *os.File // kept open so the runtime can write a crash traceback to it
 )
 
+// maxLogBytes is where the file rolls over, at startup AND while running.
+// A variable, not a constant, so the rotation can be exercised in a test without
+// writing ten megabytes to do it.
+var maxLogBytes int64 = 10 * 1024 * 1024
+
 // Init opens (creates) the log file in dataDir. On rotation we truncate
 // at startup if the file is too big; for now it's a single file, no
 // rolling — the volume is low (a few KB per session).
@@ -47,12 +53,12 @@ func Init(dataDir string) (string, error) {
 			_ = os.Rename(oldLog, logPath)
 		}
 	}
-	// Rotate (don't delete) once the file grows past ~10MB: rename it to
+	// Rotate (don't delete) once the file grows past maxLogBytes: rename it to
 	// opslog.log.1 so the PREVIOUS session's log survives. Deleting it outright used
 	// to erase exactly the diagnostics we needed when a user reported an issue from
 	// the session that just ended. One generation kept — enough, without unbounded growth.
-	if fi, err := os.Stat(logPath); err == nil && fi.Size() > 10*1024*1024 {
-		_ = os.Remove(logPath + ".1")     // drop the older generation
+	if fi, err := os.Stat(logPath); err == nil && fi.Size() > maxLogBytes {
+		_ = os.Remove(logPath + ".1") // drop the older generation
 		if err := os.Rename(logPath, logPath+".1"); err != nil {
 			_ = os.Remove(logPath) // rename failed (locked?) → fall back to the old behaviour
 		}
@@ -96,13 +102,57 @@ func Printf(format string, args ...any) {
 		msg = msg[:len(msg)-1]
 	}
 	if file != nil {
-		fmt.Fprintf(file, "%s %s\n", stamp, msg)
+		n, _ := fmt.Fprintf(file, "%s %s\n", stamp, msg)
+		written += int64(n)
+		rotateIfBigLocked()
 	}
 	// Also dump to stderr in case the binary was launched with a console
 	// attached (wails dev, custom build).
 	fmt.Fprintf(os.Stderr, "%s %s\n", stamp, msg)
 }
 
+// rotateIfBigLocked rolls the file over mid-session once it passes maxLogBytes.
+//
+// The check used to happen at STARTUP only, on the reasonable belief that a
+// session writes a few kilobytes. The CI-V wire trace disproved it: left on for
+// a day it wrote 416 MB, and nothing would have stopped it before the disk did.
+// A log that has to be rotated by quitting the program is not a bound.
+//
+// Same one-generation scheme as the startup rotation, so the previous stretch
+// survives — which is the half an operator usually needs, the part just before
+// the thing they are reporting.
+//
+// Caller holds mu.
+func rotateIfBigLocked() {
+	if written < maxLogBytes || file == nil || path == "" {
+		return
+	}
+	fmt.Fprintf(file, "%s ── log reached %d MB, rotating to %s.1 ──\n",
+		time.Now().Format("15:04:05.000"), maxLogBytes/(1024*1024), filepath.Base(path))
+	_ = file.Close()
+	file = nil
+	_ = os.Remove(path + ".1")
+	if err := os.Rename(path, path+".1"); err != nil {
+		// Rename refused (the file is held open elsewhere, a virus scanner): keep
+		// writing to the old handle rather than losing the log entirely, and try
+		// again in another maxLogBytes rather than on every single line.
+		f, oerr := os.OpenFile(path, os.O_CREATE|os.O_WRONLY|os.O_APPEND, 0o644)
+		if oerr == nil {
+			file = f
+		}
+		written = 0
+		return
+	}
+	f, err := os.OpenFile(path, os.O_CREATE|os.O_WRONLY|os.O_APPEND, 0o644)
+	if err != nil {
+		return // nothing more to be done; the next Printf simply writes nowhere
+	}
+	file = f
+	written = 0
+	fmt.Fprintf(file, "%s ── continued from %s.1 ──\n",
+		time.Now().Format("15:04:05.000"), filepath.Base(path))
+}
+
 // Path returns where the file is so the UI can surface it.
 func Path() string {
 	mu.Lock()
@@ -118,4 +168,19 @@ func Close() {
 		_ = file.Close()
 		file = nil
 	}
+	// The crash file is held open for the whole run so the runtime can dump a
+	// traceback into it without allocating; on the way out it has to be released
+	// like any other handle, or the data directory cannot be moved or removed.
+	if crashFile != nil {
+		// The RUNTIME holds its own duplicate of this handle (SetCrashOutput), so
+		// closing ours releases nothing on Windows — the data directory stays
+		// locked. Hand the runtime a nil first. Anything fatal in the last moments
+		// of shutdown is no longer captured, which is the right trade at a point
+		// where the databases are already closed.
+		_ = debug.SetCrashOutput(nil, debug.CrashOptions{})
+		_ = crashFile.Close()
+		crashFile = nil
+	}
+	path = ""
+	written = 0
 }

internal/applog/rotate_test.go

@@ -0,0 +1,49 @@
+package applog
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+)
+
+// The log has to be bounded WHILE RUNNING, not only at startup. The CI-V wire
+// trace left on for a day wrote 416 MB, and nothing would have stopped it
+// before the disk did.
+func TestRotatesMidSession(t *testing.T) {
+	orig := maxLogBytes
+	maxLogBytes = 8 * 1024 // small, so the test costs milliseconds
+	defer func() { maxLogBytes = orig }()
+
+	dir := t.TempDir()
+	p, err := Init(dir)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer Close()
+
+	line := strings.Repeat("x", 256)
+	for i := 0; i < 100; i++ { // ~25 KB: several times the limit
+		Printf("%s", line)
+	}
+
+	if _, err := os.Stat(p + ".1"); err != nil {
+		t.Fatalf("no previous generation kept: %v", err)
+	}
+	fi, err := os.Stat(p)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if fi.Size() > maxLogBytes {
+		t.Fatalf("current log is %d bytes, over the %d limit — it did not roll over", fi.Size(), maxLogBytes)
+	}
+	// The new file says where the rest went: a log that starts mid-sentence with
+	// no explanation is how an operator concludes the app lost its history.
+	b, err := os.ReadFile(p)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !strings.Contains(string(b), "continued from "+filepath.Base(p)+".1") {
+		t.Error("the new file does not point at the one it continues")
+	}
+}

internal/cat/flex.go

@@ -74,6 +74,15 @@ type Flex struct {
 	// end up outside the new window — see ZoomPan.
 	panWindow map[string]panView
 
+	// spotSig is what each live spot LOOKS like, and when it was drawn. A
+	// callsign spotted again with the same frequency, colour and comment does not
+	// need to be removed and re-added: the panadapter would not change by one
+	// pixel, and a busy RBN feed re-spots the same station every few seconds. One
+	// two-minute session sent 2 128 adds and 1 634 removes, 88 of them for a
+	// single callsign — see SendSpot.
+	spotSig  map[string]string
+	spotSent map[string]time.Time
+
 	spotByCall map[string]int // callsign → live spot index, so re-spotting a call replaces its old spot (WSJT decodes re-fire every cycle)
 	sentCmds   map[int]string // seq → command text, so an R<seq> error names the command
 
@@ -205,7 +214,7 @@ func NewFlex(host string, port int, spotsEnabled bool) *Flex {
 	return &Flex{
 		host: strings.TrimSpace(host), port: port,
 		slices: map[int]*flexSlice{}, spotsEnabled: spotsEnabled,
-		spotIdx: map[int]bool{}, pendingSpot: map[int]string{}, pendingSpotMode: map[int]string{}, spotCall: map[int]string{}, spotMode: map[int]string{}, spotFreq: map[int]int64{}, pendingSpotFreq: map[int]int64{}, panWindow: map[string]panView{}, spotByCall: map[string]int{}, pendingSplit: map[int]bool{},
+		spotIdx: map[int]bool{}, pendingSpot: map[int]string{}, pendingSpotMode: map[int]string{}, spotCall: map[int]string{}, spotMode: map[int]string{}, spotFreq: map[int]int64{}, pendingSpotFreq: map[int]int64{}, panWindow: map[string]panView{}, spotSig: map[string]string{}, spotSent: map[string]time.Time{}, spotByCall: map[string]int{}, pendingSplit: map[int]bool{},
 		meterMeta: map[int]meterInfo{}, meterVal: map[int]float64{}, meterSub: map[int]bool{},
 		sentCmds: map[int]string{}, txSetAt: map[string]time.Time{},
 		pinnedSlice: -1,
@@ -859,6 +868,13 @@ func (f *Flex) handleStatus(payload string) {
 				}
 				f.mu.Lock()
 				if removed {
+					// The radio expired or dropped this spot: forget what it looked
+					// like, or the next identical spot would be skipped as "already
+					// drawn" and the station would never come back.
+					if c, ok := f.spotCall[idx]; ok {
+						delete(f.spotSig, strings.ToUpper(c))
+						delete(f.spotSent, strings.ToUpper(c))
+					}
 					delete(f.spotIdx, idx)
 					delete(f.spotCall, idx)
 					delete(f.spotMode, idx)
@@ -1390,7 +1406,22 @@ func (f *Flex) SendSpot(s SpotInfo) error {
 	// and Go mutexes aren't reentrant (calling send while locked deadlocks the
 	// whole Flex goroutine → the radio drops OFFLINE).
 	upperCall := strings.ToUpper(s.Callsign)
+	// Unchanged and still fresh → leave it alone.
+	//
+	// Refreshed once past half its lifetime, so a long-lived spot still gets its
+	// timer wound on and never quietly expires off the panadapter; anything that
+	// actually differs (a new frequency, a status that has changed colour, a
+	// different comment) is redrawn at once.
+	sig := fmt.Sprintf("%d|%s|%s|%s|%s", s.FreqHz, s.Mode, s.Color, s.BackgroundColor, s.Comment)
 	f.mu.Lock()
+	if prev, ok := f.spotSig[upperCall]; ok && prev == sig {
+		if at, ok := f.spotSent[upperCall]; ok && time.Since(at) < time.Duration(life)*time.Second/2 {
+			f.mu.Unlock()
+			return nil
+		}
+	}
+	f.spotSig[upperCall] = sig
+	f.spotSent[upperCall] = time.Now()
 	old, hadOld := f.spotByCall[upperCall]
 	if hadOld {
 		delete(f.spotByCall, upperCall)
@@ -1464,6 +1495,8 @@ func (f *Flex) ClearSpots() error {
 	f.spotCall = map[int]string{}
 	f.spotMode = map[int]string{}
 	f.spotFreq = map[int]int64{}
+	f.spotSig = map[string]string{}
+	f.spotSent = map[string]time.Time{}
 	f.spotByCall = map[string]int{}
 	connected := f.conn != nil
 	f.mu.Unlock()

internal/cat/icom_foreign_test.go

@@ -0,0 +1,43 @@
+package cat
+
+import (
+	"testing"
+
+	"hamlog/internal/cat/civ"
+)
+
+// A CI-V bus can carry a second controller, and the rig answers each of them on
+// the same wire. A reply addressed to somebody else must not be taken for ours —
+// on an IC-7850 one arrived between a PTT command and its acknowledgement, the
+// acknowledgement was never matched, and the failure was handed to JTDX as a rig
+// control error.
+func TestForeignCIVRepliesAreIgnored(t *testing.T) {
+	const rig = 0x8E
+	frames, _ := civ.Scan([]byte{
+		0xFE, 0xFE, 0x01, rig, 0x1C, 0x00, 0x01, 0xFD, // to controller 01 — not ours
+		0xFE, 0xFE, civ.AddrController, rig, 0xFB, 0xFD, // to us: the acknowledgement
+		0xFE, 0xFE, 0x00, rig, 0x00, 0x00, 0x25, 0x09, 0x14, 0x00, 0xFD, // transceive broadcast
+	})
+	if len(frames) != 3 {
+		t.Fatalf("scanned %d frames, want 3", len(frames))
+	}
+	var kept []civ.Decoded
+	for _, f := range frames {
+		if f.From != rig {
+			continue
+		}
+		if f.To != civ.AddrController && f.To != 0x00 {
+			continue
+		}
+		kept = append(kept, f)
+	}
+	if len(kept) != 2 {
+		t.Fatalf("kept %d frames, want 2 (ours + the broadcast)", len(kept))
+	}
+	if kept[0].Cmd != 0xFB {
+		t.Errorf("the acknowledgement was not the first kept frame: % X", kept[0].Cmd)
+	}
+	if kept[1].To != 0x00 {
+		t.Errorf("the transceive broadcast was dropped")
+	}
+}

internal/cat/icomserial.go

@@ -514,8 +514,21 @@ func (b *IcomSerial) SetFrequency(hz int64) error {
 		return fmt.Errorf("invalid frequency")
 	}
 	b.lastSetFreq, b.lastSetFreqAt = hz, time.Now()
-	return b.execIdempotent(fmt.Sprintf("set frequency %d Hz", hz),
+	err := b.execIdempotent(fmt.Sprintf("set frequency %d Hz", hz),
 		append([]byte{civ.CmdSetFreq}, civ.FreqToBCD(hz)...)...)
+	// Same reasoning as SetPTT: a lost acknowledgement is not a failed command,
+	// and WSJT-X or JTDX answer a failed set_freq by dropping the link — an
+	// IC-7850 log shows exactly that at 15:16:15, the disconnect four seconds
+	// later. The rig is asked what frequency it is on rather than being called a
+	// failure on the strength of a missing "FB".
+	if err == nil || !errors.Is(err, errIcomAckLost) {
+		return err
+	}
+	if got, rerr := b.readFreq(); rerr == nil && got == hz {
+		applog.Printf("icom: frequency %d Hz not acknowledged, but the rig is on it — treating as done", hz)
+		return nil
+	}
+	return err
 }
 
 func (b *IcomSerial) SetMode(mode string) error {
@@ -526,7 +539,16 @@ func (b *IcomSerial) SetMode(mode string) error {
 	// Set the base mode (keeping the rig's current filter by sending only the
 	// mode byte), then set the data-mode flag for digital modes.
 	if err := b.execIdempotent("set mode "+mode, civ.CmdSetMode, code); err != nil {
-		return err
+		// A lost acknowledgement is not a refused mode — see SetPTT. The mode is
+		// readable, so ask rather than report a failure that would cost the
+		// digital application its connection.
+		if !errors.Is(err, errIcomAckLost) {
+			return err
+		}
+		if got, ok := b.readMode(); !ok || got != code {
+			return err
+		}
+		applog.Printf("icom: mode %s not acknowledged, but the rig reports it — treating as done", mode)
 	}
 	dataByte := byte(0)
 	if data {
@@ -569,12 +591,43 @@ func (b *IcomSerial) execIdempotent(what string, payload ...byte) error {
 }
 
 // SetPTT keys or unkeys the transmitter (CI-V 0x1C 0x00).
+// A lost acknowledgement is NOT proof the rig ignored us, and treating it as one
+// is expensive: OpsLog fronts a Hamlib rigctl port, so the failure is handed
+// straight to WSJT-X or JTDX, which answer a failed set_ptt with a Rig Control
+// Error and drop the link mid-over. An IC-7850 log showed exactly that, three
+// times in a session, on a CI-V bus with a second controller corrupting frames.
+//
+// So when the acknowledgement does not arrive, ASK the rig what it is doing. PTT
+// is readable (0x1C 0x00), the answer is unambiguous, and a rig that is keyed
+// received the command whatever became of its reply.
 func (b *IcomSerial) SetPTT(on bool) error {
 	state := byte(0)
 	if on {
 		state = 1
 	}
-	return b.execIdempotent(fmt.Sprintf("PTT %v", on), civ.CmdPTT, civ.SubPTT, state)
+	err := b.execIdempotent(fmt.Sprintf("PTT %v", on), civ.CmdPTT, civ.SubPTT, state)
+	if err == nil || !errors.Is(err, errIcomAckLost) {
+		return err
+	}
+	// A GENEROUS window for the verification, not the poll's 150 ms.
+	//
+	// The rig has just failed to answer twice in 800 ms, and the reason seen on a
+	// real IC-7850 is that it was retuning: JTDX moves the transmit frequency and
+	// keys in the same breath. Asking it again with the short probe timeout would
+	// fail for exactly the same reason and prove nothing.
+	keyed, ok := b.readTXWithin(icomPTTVerifyTimeout)
+	if ok && keyed == on {
+		applog.Printf("icom: PTT %v not acknowledged, but the rig reports it — treating as done", on)
+		return nil
+	}
+	// Logged either way, because "the fix did not work" and "the rig really did
+	// not key" look identical from outside and need telling apart.
+	if ok {
+		applog.Printf("icom: PTT %v not acknowledged, and the rig reports %v — reporting the failure", on, keyed)
+	} else {
+		applog.Printf("icom: PTT %v not acknowledged, and the rig did not answer the check either", on)
+	}
+	return err
 }
 
 // SetPower turns the transceiver on or off (CI-V 0x18). Power-ON is prefixed with
@@ -686,6 +739,19 @@ func (b *IcomSerial) reader(port civTransport, done chan struct{}) {
 			if f.From != b.rigAddr {
 				continue // echo of our own command
 			}
+			// Addressed to US, or broadcast.
+			//
+			// A CI-V bus can carry a second controller — RS-BA1, wfview, another
+			// logger — and the rig answers each of them in turn on the same wire.
+			// Those replies come FROM the rig, so the check above lets them through,
+			// and one can be taken for the answer to our own command: an IC-7850 log
+			// showed "FE FE 01 8E …" (to controller 01) arriving between our PTT
+			// command and its acknowledgement, with bus collisions around it.
+			// Transceive broadcasts (to 0x00) stay welcome — that is how the rig
+			// announces a knob turn.
+			if f.To != civ.AddrController && f.To != 0x00 {
+				continue
+			}
 			if f.Cmd == civ.CmdScope {
 				b.route(b.specCh, f)
 				continue
@@ -1273,16 +1339,30 @@ func (b *IcomSerial) readTXFreq() (int64, bool) {
 
 // readTX reads the transmit state (CI-V 0x1C 0x00): non-zero data = keyed.
 func (b *IcomSerial) readTX() bool {
+	on, _ := b.readTXWithin(icomDSPTimeout)
+	return on
+}
+
+// icomPTTVerifyTimeout is how long the PTT check waits. Long, deliberately: it
+// runs only after a command was already given up on, so it costs nothing in the
+// normal case and everything depends on it in the abnormal one.
+const icomPTTVerifyTimeout = 700 * time.Millisecond
+
+// readTXWithin reads the transmit state, reporting whether the rig ANSWERED.
+//
+// The two are different questions and the caller usually needs both: "not
+// transmitting" and "did not say" are the same bool but not the same fact.
+func (b *IcomSerial) readTXWithin(timeout time.Duration) (on bool, answered bool) {
 	if err := b.write(civ.CmdPTT, civ.SubPTT); err != nil {
-		return false
+		return false, false
 	}
-	f, err := b.recv(icomDSPTimeout, func(d civ.Decoded) bool {
+	f, err := b.recv(timeout, func(d civ.Decoded) bool {
 		return d.Cmd == civ.CmdPTT && len(d.Data) >= 2 && d.Data[0] == civ.SubPTT
 	})
 	if err != nil {
-		return false
+		return false, false
 	}
-	return f.Data[1] != 0
+	return f.Data[1] != 0, true
 }
 
 // readMeter reads a meter (CI-V 0x15) and returns it scaled to 0-100.