…mments

Addresses non-blocking review notes 1 and 2 from aethersdr#3473:

1. ax25DemodDescription() now appends ", N lane(s)" via a new
   ax25DemodLaneCount() free function computed from config — restores
   the suffix that was lost when the free function was extracted from
   the member demodDescription() during the threading refactor.

2. Restore two explanatory comments dropped during
   ax25BuildTransmitAudioFromFrame extraction: the minimum-frame-size
   invariant (14 addr + 1 ctrl = 15 bytes, no FCS) and the
   "host omits FCS; TNC appends it" note.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

…mments

Addresses non-blocking review notes 1 and 2 from aethersdr#3473:

1. ax25DemodDescription() now appends ", N lane(s)" via a new
   ax25DemodLaneCount() free function computed from config — restores
   the suffix that was lost when the free function was extracted from
   the member demodDescription() during the threading refactor.

2. Restore two explanatory comments dropped during
   ax25BuildTransmitAudioFromFrame extraction: the minimum-frame-size
   invariant (14 addr + 1 ctrl = 15 bytes, no FCS) and the
   "host omits FCS; TNC appends it" note.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- Drop dead selectedTonePolarity() declaration from Ax25HfPacketDecodeDialog.h
  — leftover from conflict resolution during rebase onto main after aethersdr#3473 merged
- handleMqttMessage in Ax25HfPacketDecodeDialog now guards with
  isMqttTopicEnabled, consistent with the CW transmit handler in MainWindow
- cw/decode TX payloads (rx:false) now pull pitch_hz/speed_wpm from
  transmitModel().cwPitch()/cwSpeed() rather than the last RX measurement
- radio/state publish coalesced through a 150 ms single-shot timer to prevent
  flooding on VFO drag/scan (frequencyChanged fires on every intermediate step)
- setMqttTopicEnabled no longer saves to disk per call; saveSettings() does
  one AppSettings::save() after the full loop
- Add comment near kCwDecodeTopic/kCwTransmitTopic noting relay scripts should
  filter on the aethersdr/ namespace to avoid cw/decode→cw/transmit feedback
  loops — the full topic path serves as the origin identifier.
  @aethersdr-agent: was an explicit origin field your intent, or is the
  namespace sufficient?

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

… HDLC framer (#3475)

## Summary

Second of three focused PRs extracted from #3466 per @jensenpat's review guidance, implementing part of the design laid out in #2701 (Native Soft-TNC + APRS Integration). Introduces `HdlcCodec`, a self-contained pure-C++ HDLC framer with CRC-16-CCITT, and wires it into the VHF 1200-baud decode lane in place of libmodem's `bitstream_state`. libmodem is retained for HF demodulation and AX.25 frame parsing.

Also addresses non-blocking review notes 1 and 2 from #3473: lane count restored in `ax25DemodDescription()` via `ax25DemodLaneCount()`, and two explanatory comments restored in `ax25BuildTransmitAudioFromFrame`.

## Motivation

libmodem's `bitstream_state` is an opaque internal struct — not unit-testable in isolation and not replaceable without forking libmodem. Extracting HDLC framing into `HdlcCodec` makes the state machine independently testable, and establishes the boundary where the Direwolf demodulator (PR 3) will plug in.

## Changes

- **New:** `src/core/tnc/HdlcCodec.h` / `HdlcCodec.cpp` — pure C++ HDLC bit-unstuffing and CRC-16-CCITT framer
- **New:** `tests/hdlc_codec_test.cpp` — 27 tests covering state transitions, FCS validation cross-checked against libmodem, two-frame decode, long payload, and reset
- `CMakeLists.txt`: added `HdlcCodec.cpp` to main build and shim test; added `hdlc_codec_test` target
- `AetherAx25LibmodemShim`: `DecodeLane` replaces `bitstreamState` with `HdlcCodec hdlcCodec`
- `ax25DemodLaneCount()` free function added; `ax25DemodDescription()` now includes lane count suffix
- Two explanatory comments restored in `ax25BuildTransmitAudioFromFrame` (minimum-frame-size invariant; host-omits-FCS note)

Squashed-from: #3475

Co-authored-by: Paul <274291579+K5PTB@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: jensenpat <84298137+jensenpat@users.noreply.github.com>

## Release prep for v26.6.3

Bumps the version and refreshes documentation for the **58 commits since
v26.6.2** (2026-06-07). No tag is created — tagging is the release
trigger (the #3567 MS Store automation fires on `v*` tags), so that step
is left to the maintainer after this merges.

### Version
- `CMakeLists.txt`: `VERSION 26.6.2 → 26.6.3` (drives the MSIX
`Identity.Version`).
- README.md + AGENTS.md "current version" strings updated to match.

### CHANGELOG.md — v26.6.3 entry
Covers: WFM software demodulator, APRS client, PSK Reporter reception
map + reusable Qt mapping engine (QGeoView), Direwolf-derived VHF AFSK
demodulator, DAX-IQ made fully usable, profile-load recovery hardening,
DEXP protocol fix, Multi-Flex ping-watchdog grace, MQTT enhancements,
the #3351 MainWindow decomposition + `RadioSession`, the MS Store CI
auto-stage, and the fix roll-up.

### Doc refresh (reviewed all repo-root + `docs/` files)
- **README.md** — Highlights: added WFM, AetherModem packet/APRS, PSK
Reporter map.
- **ROADMAP.md** — cycle → `post-v26.6.3`; moved **AetherModem Phase 1**
from Queued to **Recently shipped** alongside WFM / AFSK / PSK-Reporter
map / DAX-IQ / the decomposition.
- **AGENTS.md** — version, added `RadioSession` to Key classes, threads
**11 → 12**.
- **CONTRIBUTING.md** — thread count 11 → 12.
- **docs/architecture/pipelines.md** — added the **AX.25 TNC thread**
(#3473) to the list + table; **11 → 12**.
- **docs/MODEM.md** — VHF 1200 baud now documents the **Direwolf-derived
Profile A+ demodulator (9 slicers)** that replaced the libmodem lane
bank (#3527); the APRS client + map are no longer "out of scope" (HF
300-baud section unchanged).

Reviewed (via two parallel doc-review agents) and found **no changes
needed** in: CONSTITUTION, GOVERNANCE, SUPPORT, SECURITY,
CODE_OF_CONDUCT, CLAUDE/GEMINI (pointer files), and the rest of `docs/`
(audio-pipeline, vita49-format, tci-*, style/, theming/, qa/, etc.).

### Verification
- Builds clean with `VERSION 26.6.3`; full suite **38/38 pass**.

### To roll the release (maintainer)
After merge: `git tag -s v26.6.3 && git push origin v26.6.3` → triggers
the Windows installer / release workflows (and the MS Store draft-stage
once its credentials are set).

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

## Summary

First of three focused PRs extracted from aethersdr#3466 per @jensenpat's review
guidance. Lays the threading groundwork for aethersdr#2701 (Native Soft-TNC +
APRS Integration) — the shim thread boundary is required before a more
capable demodulator can be dropped in without blocking the GUI.

## Motivation

`AetherAx25LibmodemShim` ran on the GUI thread. Audio callbacks, HDLC
decode, and profile configuration all competed with paint events and
user interaction, making it impossible to later add a CPU-heavier
demodulator without degrading responsiveness.

## Changes

- `AetherAx25LibmodemShim` moved to a dedicated `QThread`
(`m_shimThread`)
- All GUI→shim calls use `QMetaObject::invokeMethod(...,
Qt::QueuedConnection)`
- `m_shimConfig` cache added to dialog — GUI thread never reads
`m_impl->config` directly
- Free functions extracted: `ax25DemodDescription`,
`ax25BuildTransmitAudio`, `ax25BuildTransmitAudioFromFrame`
- `startTransmit(const QString&)` extracted from `startTransmitFromUi`
- Destructor: `m_shimThread.quit(); m_shimThread.wait()`
- **ASAN fix (`b495dac9`):** `MainWindow::~MainWindow()` now explicitly
deletes `m_ax25HfPacketDecodeDialog` before AudioEngine teardown.
Pre-existing UAF: dialog destructor called
`m_audio->setTncRxTapEnabled(false)` via raw pointer after AudioEngine
was freed via `deleteLater()`. Same pattern as the existing
`m_networkDiagnosticsDialog` fix in that destructor. Found on RPi 5 with
ASAN; confirmed clean after fix.

## Test plan

Tested on all three target platforms.

1. **Basic decode** — enabled modem on 144.390 MHz APRS; frames appeared
in the log within seconds; disable/re-enable confirmed clean reset with
no stale state.
2. **Profile switching under load** — switched HF↔VHF while audio was
flowing, including rapid successive switches; no crash, new profile
description appeared correctly each time.
3. **TX path** — transmitted a UI frame; confirmed key, send, and unkey
with no crash or hang. KISS frame sent via `kiss_test.py` also keyed
cleanly, exercising `ax25BuildTransmitAudioFromFrame` as a free
function. **Hardware note:** transverter attenuated to prevent RF output
— full software TX path exercised but no RF generated; iGate/APRS-IS
verification deferred pending hardware modification.
4. **Clean shutdown** — closed dialog while modem was enabled; no hang
or crash; reopened and re-initialized correctly.
5. **Extended run** — modem left running 10+ minutes on a busy APRS
channel; no crashes, no memory growth, GUI remained responsive.
6. **ASAN (RPi 5)** — Debug+ASAN build clean after fix; no new leaks or
UAF reported.

**RX sensitivity note:** A single decode was achieved with an HT at
close range. Consistent decode is limited by libmodem's single-slicer
VHF sensitivity — a known pre-existing limitation addressed in the third
PR in this series (Direwolf demodulator). This PR does not regress
decode behavior.

**300-baud HF decode** — not tested on-air; no 300-baud packet
infrastructure available in the test environment.

---

Built and smoke tested on: MacBook Air M2 / RPi 5 Debian trixie / Win 11
i3

Extracted from aethersdr#3466 (closed).

**Note:** The follow-on PR introducing `HdlcCodec`
([feat/ax25-hdlc-codec](https://github.com/K5PTB/AetherSDR/tree/feat/ax25-hdlc-codec))
is ready but will be held until this PR merges, so it can be opened
cleanly against `main`.

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

… HDLC framer (aethersdr#3475)

## Summary

Second of three focused PRs extracted from aethersdr#3466 per @jensenpat's review guidance, implementing part of the design laid out in aethersdr#2701 (Native Soft-TNC + APRS Integration). Introduces `HdlcCodec`, a self-contained pure-C++ HDLC framer with CRC-16-CCITT, and wires it into the VHF 1200-baud decode lane in place of libmodem's `bitstream_state`. libmodem is retained for HF demodulation and AX.25 frame parsing.

Also addresses non-blocking review notes 1 and 2 from aethersdr#3473: lane count restored in `ax25DemodDescription()` via `ax25DemodLaneCount()`, and two explanatory comments restored in `ax25BuildTransmitAudioFromFrame`.

## Motivation

libmodem's `bitstream_state` is an opaque internal struct — not unit-testable in isolation and not replaceable without forking libmodem. Extracting HDLC framing into `HdlcCodec` makes the state machine independently testable, and establishes the boundary where the Direwolf demodulator (PR 3) will plug in.

## Changes

- **New:** `src/core/tnc/HdlcCodec.h` / `HdlcCodec.cpp` — pure C++ HDLC bit-unstuffing and CRC-16-CCITT framer
- **New:** `tests/hdlc_codec_test.cpp` — 27 tests covering state transitions, FCS validation cross-checked against libmodem, two-frame decode, long payload, and reset
- `CMakeLists.txt`: added `HdlcCodec.cpp` to main build and shim test; added `hdlc_codec_test` target
- `AetherAx25LibmodemShim`: `DecodeLane` replaces `bitstreamState` with `HdlcCodec hdlcCodec`
- `ax25DemodLaneCount()` free function added; `ax25DemodDescription()` now includes lane count suffix
- Two explanatory comments restored in `ax25BuildTransmitAudioFromFrame` (minimum-frame-size invariant; host-omits-FCS note)

Squashed-from: aethersdr#3475

Co-authored-by: Paul <274291579+K5PTB@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Signed-off-by: jensenpat <84298137+jensenpat@users.noreply.github.com>