- Bot Actions (!buzz/!gps/!advert/!gpio1-4) ran their side effect
immediately during botScanCommands(), before quiet-hours/cooldown/
per-contact throttle were checked -- those gates only suppressed the
reply text, not the actual buzz/GPS toggle/advert/pin write. botCommandReply()
now only records what was requested; applyPendingBotActions() runs the
deferred effects once a wrapper's throttle checks pass and the ack sent,
mirroring the existing _locfix_requested pattern. resetPendingBotActions()
clears everything on every throttled/aborted path.
- CardKB's Fn+<letter> accent-popup shortcut bypassed the locked-input gate
(it called into KeyboardWidget directly instead of through the
enqueueKey()/dequeue path every other key uses, so it wasn't discarded
while _locked). Now checks _locked itself.
- Since a locked device now correctly ignores CardKB entirely, Fn+Esc
(single press) is added as CardKB's own lock/unlock gesture -- otherwise
a CardKB-only setup had no way to unlock. Esc rather than the adjacent
Fn+Backspace, to avoid an accidental press.
- botScanCommands() now parses up to two arguments per command instead of
one. Used by "!gps fix [seconds]" to override the default 90s timeout
(clamped 15-300s) for a poor sky view where 90s isn't always enough to
reach isLocFixReady()'s HDOP/satellite bar.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Turns GPS on (if it wasn't already), waits for a stabilised fix
(isValid() + >=8 satellites, then averages 10s of readings), sends the
position, and restores GPS to whatever state it was in before -- up to
a 90s timeout, after which it reports a partial fix (if it got any
samples) or plain failure.
Replies in two parts since a fix takes seconds-to-minutes, unlike every
other bot command here: an immediate "acquiring fix..." ack (through
the existing synchronous command path), then the actual position as a
separate follow-up message once ready, delivered to whichever
destination (DM/room/channel) the request came from. Only one fix can
be in flight at a time -- a second request while one is pending gets an
immediate "already pending" instead of silently replacing it.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Messages:
- Add/edit/delete channels on-device (new ChannelsView, owned by the renamed
MessagesScreen — was QuickMsgScreen, whose name no longer matched its scope).
Channel secret entry supports a typed passphrase (SHA-256'd, same primitive
the library already uses for the routing hash) or a raw 32-hex-char key.
- MyMesh::setChannelLocal() factors out the setChannel/saveChannels/
onChannelRemoved sequence previously duplicated across the two
CMD_SET_CHANNEL branches, shared now by the BLE and on-device paths.
Tools > Admin (new):
- Log into a repeater/room server's admin account and send CLI commands,
the on-device equivalent of the app's repeater-admin feature.
- Commands are organised into category tabs (System/Radio/Routing/Actions)
with common get/set fields (name, radio profile, tx power, repeat, advert
intervals, ...) plus a free-text "Custom command..." fallback for anything
else. A field row fetches the current value, opens it pre-filled for
editing, and sends the change — falling back to a blank editor if the
fetch fails or times out.
- The admin password persists and self-heals exactly like room logins in
Messages: saved on a confirmed admin-level login, forgotten on a failed
one, left alone if merely under-privileged.
- New MyMesh::sendAdminCommand()/AbstractUITask::onAdminReply() plumbing so
a reply reaches the UI without touching the existing BLE/app CLI-terminal
path (queueMessage's should_display gate is untouched).
Shared TabBar.h extracted from NearbyScreen/BotScreen's independently
duplicated tab-carousel rendering (now a third consumer via Admin) — also
fixes neighbouring tabs vanishing outright when they didn't fully fit;
they now truncate with an ellipsis instead.
Keyboard: the combined "Ext.Latin" alphabet split into 8 separate,
linguistically complete per-language keyboards (Polish, Czech, Slovak,
German, French, Spanish, Portuguese, Nordic), and fixed an OLED-only bug
where tall accented glyphs overlapped the keyboard's separator line
(SH1106's Lemon-font ascent constant was 2-3px short for them).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Auto-Reply Bot gains a third target (room servers, alongside DM and
channel), new {name}/{hops} reply placeholders, and a DM all/favourites
allow-list. BotScreen is redesigned as a circular tab carousel (Direct /
Channel / Room / Other, same interaction as Nearby Nodes' filter tabs)
instead of one long scrolling list, which also exposed and fixed a
leftover coupling where Channel/Room trigger-replies and their !command
handling secretly depended on the DM tab's Enable/Commands toggles —
each target's Enable and Commands are now fully independent.
- NodePrefs: bot_room_enabled/prefix/trigger/reply, bot_dm_scope,
bot_commands_ch/bot_commands_room (bot_commands_enabled repurposed as
DM-only). SCHEMA_SENTINEL 0xC0DE001D -> 0xC0DE001F; sizeof unchanged
at 2712 (new bytes absorbed existing padding, verified via a
standalone host compile + offsetof check).
- DataStore: persists all new fields; seeds bot_commands_ch/room from
the old shared bot_commands_enabled on upgrade so existing
channel/room command behaviour isn't silently lost.
- MyMeshBot: tryBotReplyRoom/tryBotRoomCommand mirror the channel bot's
shape but post via sendMessage (room relays to members itself);
requires an existing login session with that room, same as a manual
post would. botDmSenderAllowed() gates DM trigger-reply/commands on
the favourites bit when bot_dm_scope=Fav.
- MsgExpand: {name}/{hops} as optional trailing params (default
nullptr/-1, no existing caller affected) — deliberately not exposed
on the general compose keyboard, only on bot Reply fields.
- QuickMsgScreen/UITask: room-target picker (mirrors the channel
picker), routing through the existing room-login prompt when there's
no saved password yet.
- BotScreen: tab carousel (LEFT/RIGHT switches tabs, UP/DOWN moves
rows, Enter is now the only way to change a value); Enable split out
of Channel/Room's combo row; Quiet Hours gained a stepper sub-mode;
Commands moved from a shared toggle into each tab.
- docs/tools_screen.md updated for the new tab layout and behaviour.
Not build-verified — no PlatformIO toolchain in this environment.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Turn Nearby Nodes into the single node hub and make its navigation legible.
Node management (no phone app needed):
- MyMesh gains addDiscoveredContact() and deleteContactByKey(); CMD_REMOVE_CONTACT
now reuses deleteContactByKey() (one delete path: contact + blob + room password
+ UI cleanup + lazy write).
- Action menu (Hold Enter) gains Add contact (a new scanned node), Favourite /
Unfavourite (pin to the first free dial slot), and Delete contact (confirm first,
defaults to Cancel). Actions are offered contextually per row.
- A pinned contact shows a star in the list row, next to any live-share diamond.
Absorb passively-heard adverts: refreshStored() folds getRecentlyHeard() nodes that
aren't already listed as name+age rows (All filter only — AdvertPath has no type),
so "recently heard" no longer needs its own page.
Legible filter navigation: the filter (LEFT/RIGHT) is now a visible tab strip — the
active filter is a centred inverted pill, neighbours fan out and wrap around
(circular, first<->last). Only whole tabs are drawn so labels never wrap a line.
Rename Tools > "Nearby Nodes" to "Nodes".
Both solo envs build green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
UI-polish trio from CODE_REVIEW (biggest "feels finished" gain per line):
- Pill unread badges: new DisplayDriver::drawUnreadBadge()/unreadBadgeWidth()
draw a filled capsule with the count knocked out (corners knocked back for a
rounded look; inverts on a selected row). Replaces the bare right-aligned
digits in MODE_SELECT, the contact/channel pickers and favourites tiles.
No <stdio.h> in the header — fmtBadgeCount formats manually, clamps to 99+.
- Unified DM/CH history headers: DM_HIST and CHANNEL_HIST drew their titles by
hand (drawTextCentered + fillRect at lh+1), a different height/separator than
every other screen. Both now route through drawCenteredHeader().
- Trimmed default home carousel: new NodePrefs::HP_DEFAULT (Clock, Tools,
Shutdown, Favourites, Map; Messages + Settings always visible = 7 pages).
applyDefaults() seeds it instead of HP_ALL. Recent/Radio/BT/Advert/GPS/Sensors
are opt-in via Settings > Home Pages. Existing users keep their saved mask —
factory default only; no migration, no schema bump.
Both solo envs build green (OLED RAM 69.9%/Flash 62.8%; e-ink SUCCESS).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A discover response often arrives twice — the zero-hop direct copy and a
re-flooded copy relayed by another repeater carry different packet hashes,
so the mesh duplicate filter passes both. The standalone scan appended the
same node twice and app-driven discover forwarded both copies, so one
repeater showed up as two.
Track (tag, responder pubkey) for a short window in isDupDiscoverResp() and
drop the second copy in both the standalone and app-forward paths.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- locked device: a ringing alarm/timer was invisible — wakeForAlarm() now
holds the lock wake window for the whole ring, the lock screen draws the
alert overlay, and a key-dismissed ring falls back to a 5 s glance
- status bar: A / live-share / trail / repeater icons no longer vanish when
Bluetooth is off (moved outside the isSerialEnabled() gate)
- alert overlay: long text wraps to up to 3 lines inside the box instead of
overflowing the border (new UITask::renderAlertOverlay, shared with lock)
- MyMesh: force NUL on contact.name copied from an app frame (unterminated
name could overrun the AdvertPath strcpy)
- clock tools: ring/timer deadline compares now wrap-safe (signed diff),
matching the trail/loc-share timers
- messages: channel context menu freezes its target channel at open; fav
toggle no longer retargets the menu when the fav-only filter drops the row
All three solo envs build green (WioTrackerL1 OLED/Eink, GAT562-30S).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The prefs schema-migration block runs on every SCHEMA_SENTINEL mismatch, and
the "turn Favourites on" step was ungated. Since each firmware release bumps
the sentinel, every update re-applied `home_pages_mask |= HP_FAVOURITES`,
clobbering a user who had hidden it. Gate it to the transition that added
Favourites (sentinel < 0xC0DE0002), matching the trail_units_idx migration.
The Map/Trail carousel page shipped with no visibility toggle: pageBit(MAP)
returned -1 so it was always visible with no Settings row. Add a proper "Map"
toggle (new HPB_MAP/HP_MAP bit, pageBit/bitToPage cases, HOME_MAP settings
row) plus a gated one-time migration (sentinel < 0xC0DE0018) that keeps it
visible for upgraders. Bump SCHEMA_SENTINEL to 0xC0DE0018; page_order[] stays
a literal 11 so the persisted layout is unchanged.
Also fix the fresh-install seed, which used a stale 0x01FF literal that left
Favourites hidden on new installs despite its "all pages visible" comment;
seed NodePrefs::HP_ALL so fresh installs match upgraders (Favourites + Map).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
DM, room and channel history always stamped a message with receipt time
(rtc_clock.getCurrentTime()), even though the sender's real timestamp was
already available (and, for DMs/rooms, already threaded through to
addDMMsg -- just never used for display). Live-received messages hid this
because receipt lags origination by only seconds, but a room-sync replay or
an offline-queued message held by a repeater can arrive long after it was
sent, so a burst of backlog messages all showed as "just now".
storeDMMsg() now prefers msg_ts (falling back to receipt time only when
unknown). addChannelMsg() gained a timestamp parameter, threaded from
onChannelMessageRecv() down through AbstractUITask/UITask, with the same
fallback. The DM dedup check and outgoing-message timestamps are unaffected
(they use msg_ts directly, already correct).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
CMD_SET_CHANNEL clearing a slot (empty secret) left bot_channel_idx,
loc_share_channel_idx, and the per-channel melody bitmasks pointing at that
index. A new channel added later at the same slot would then silently
inherit the old one's bot target, Live Share target, or notification
melody. New onChannelRemoved() hook, mirroring onContactRemoved(), turns
the bot/Live Share channel target off (fail closed) and clears the melody
bits for that index.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
onContactOverwrite() (the contact-table-full LRU eviction path) deleted the
contact's blob and notified the companion app, but never called the new
onContactRemoved() cleanup -- so a Favourites Dial slot, Locator target, or
Live Share target could still go stale, just via the silent auto-evict path
instead of an explicit removal. This is likely the main real-world cause of
the "(gone)" tile the docs described, since auto-eviction happens far more
often than an explicit CMD_REMOVE_CONTACT.
Also sync the two docs that described the old (now wrong) behaviour:
Locator's "survives delete" claim, and Favourites Dial's "(gone) until
reassigned" claim.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Deleting a waypoint left the Locator pointed at coordinates that no longer
existed (it's a coordinate snapshot, so nothing noticed). Removing a contact
was worse: nothing cleared its favourite slot, its Locator/Live Share target,
or its per-contact mute/melody entry, so all four kept referencing a pubkey
that no longer resolved to anything.
- WaypointsView's Delete now calls UITask::clearTargetIfWaypoint() first.
- New AbstractUITask::onContactRemoved() hook, called from MyMesh.cpp's
CMD_REMOVE_CONTACT handler, clears the favourite slot, the Locator target,
and dm_notif/dm_melody entries for that pubkey. Live Share's DM target
turns auto-share off instead of guessing a new recipient.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The on-device login path already saved room passwords to /room_pw; do the
same for logins issued by the phone/USB app (CMD_SEND_LOGIN). The password
is stashed when the login is sent and persisted once the server confirms,
gated to ADV_TYPE_ROOM contacts; a failed login forgets any stale saved
password, mirroring the on-device path. This lets the device post to a
room standalone after a reboot without re-prompting.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
CMD_REMOVE_CONTACT already drops the contact's advert blob; also drop any
saved /room_pw login for that key so a deleted room doesn't leave an
orphaned (and now useless) password behind until FIFO eviction.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Log in to a room server from the device UI with no phone app: picking a
room prompts for its password (blank allowed for open rooms), and a
context-menu "Login..." allows re-login. Successful passwords are
persisted to a dedicated /room_pw file so a previously-used room logs
back in after reboot without retyping; a failed login forgets the
(now-stale) saved password so the next attempt prompts again.
The room-password file is written via a temp file + atomic rename
(new DataStore::commitFile helper), matching the crash-safety of
contacts/channels persistence.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Squash merge of feat/location-beacon-alerts-autopause (v1.21).
Features:
- Live Location Sharing — broadcast position as movement-gated [LOC]
messages to a channel or contact; live shares show as map pins with
distance/bearing in Nearby Nodes and a status-bar indicator. [LOC] is
parsed in DMs, channel messages and room messages; DM shares name the
sender.
- Locator (geofence) — arm a geofence around a saved waypoint or a person
(live [LOC] or last-known position), alert on arrive/leave or near/far,
with an optional homing beeper (gated to arrive/both modes). Arm from
Tools > Locator, Nearby Nodes, or Waypoints; target picker lists
favourites first, clearable via a "None" entry. Active target is drawn
as a flag on the map.
- One active target shared across Locator / Navigate / Map via a single
resolver (resolvePersonPos / activeTargetPos) that prefers a live
[LOC] share over the last-advertised GPS fix.
- Follow live contacts — Navigate to a live-sharing contact follows them
as they move and adds an ETA line; quick-share your own position from
the Map.
- Map & status-bar upgrades — home mini-map gets a north marker, scale
tick, and a connected trail line (was disconnected dots); status line
shows tracked-node count, an arrow + distance to the active Locator/Nav
target (falling back to the nearest live-tracked contact); GPS fix icon
in the top status bar, shown only on GPS boards with GPS enabled.
- Trail auto-pause — recording freezes on stops (banking elapsed time,
breaking the map line across the idle gap) and resumes on movement
without ending the session.
- Streaming trail simplification — GPS points are simplified in-stream
via a fixed-corridor (Reumann-Witkam) pass tuned for fidelity: straight
runs collapse to their endpoints, curves stay bounded to the Min-dist
tolerance, so the 512-point buffer covers a far longer route than a
flat point budget would suggest.
- Collapsible Tools (Location / Comms / System sections, fold-in-place
like Settings) and page-indicator icons on the home carousel.
- Waypoint coordinate editor — add a waypoint by scroll-editing lat/lon
digit by digit.
Fixes:
- Critical: low-heap hang and contact loss on RAM-tight builds. Halved
message-history scrollback rings (recovering ~14 KB free heap) and
made contacts/channels/prefs persistence atomic (temp file + rename),
so an interrupted save can no longer corrupt or wipe the store.
- Serial.write() bounded so a stalled USB host can't hang the device.
- Nearby Nodes: live [LOC] senders respect the type filter, sort by
shared position, and the list refreshes so live shares bubble up.
- Map: live contacts are labelled before waypoints.
- GPS status icon hidden when GPS is off in Settings.
- Splash screen no longer truncates a pre-release tag's own dash (e.g.
v1.21-rc1) when stripping the build's commit-hash suffix.
- Null-guarded the Locator target picker; clamped loc-share channel
index on load.
Under the hood:
- -Os size optimisation on the e-ink and GAT562 30S solo envs.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
constrain() is a min/max macro and NaN compares false against both bounds,
so a corrupted or layout-shifted prefs file that decodes a float field as
NaN/inf would pass it straight through to setParams() and could hang the
radio at boot (stuck after "Loading...", recoverable only by erasing flash).
Reset freq/bw/airtime_factor/rx_delay_base to safe defaults before the
existing constrain() so a non-finite value can never reach the driver.
Note: a defensive hardening of the boot path, not a confirmed root cause for
the reported brick.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Incoming room-server posts fired the notification and reached the app via
the offline queue, but never appeared when the room was opened directly on
the device: the on-device history (addDMMsg) was gated to ADV_TYPE_CHAT only,
while rooms (ADV_TYPE_ROOM) share that same history list (keyed by the
server's pubkey). Include ADV_TYPE_ROOM so room posts are stored too.
A room carries many guests, so prefix each stored post with its author —
resolved from the signed message's sender pubkey prefix (fallback: short
hex) — and have the DM history view split "Sender: text" for room contacts
so each line is attributed to the guest who wrote it, instead of showing the
room server's name for every message.
Builds verified: WioTrackerL1 + GAT562 30S solo.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Rename the repeater filter knobs from "politeness" to "forwarding
filters" across code comments, FEATURES.md and the docs, and correct
references that hadn't followed the controls when they moved from
Settings › Radio to the dedicated Tools › Repeater screen.
Comment/doc accuracy fixes:
- NodePrefs: user_radio_presets are now written by both Settings and
Repeater (shared picker), not Settings only.
- DiagnosticsScreen: class header listed only some rows; spell out the
full set (forwarded, signal, pool/queue, error flags, reset gesture).
- RepeaterScreen header no longer claims to show live forwarding stats
(they live on Diagnostics).
- tools_screen.md: drop the stale paragraph claiming the Repeater screen
shows live stats at the bottom — it is config-only.
Comments/docs only; no behaviour change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Settings > Radio gains a repeater toggle, 16 community-suggested radio
presets plus manual Freq/SF/BW/CR tuning (digit-by-digit Freq editor),
4 persisted user preset slots, and the packet pool bumped 16->32 so
queued retransmits stop starving incoming-packet allocation while
relaying.
Four opt-in politeness knobs (skip-advert, max-hops, yield, min-SNR),
all flood-only and off by default, plus overhear suppression that
cancels a queued retransmit if a peer relays the same packet first.
Adaptive Power Control is suppressed while relaying (pins TX power to
the ceiling) and duty-cycle RX is forced off, since a repeater needs
to hear and relay at consistent power.
A dedicated Tools > Repeater screen consolidates the toggle, the
politeness knobs, and live forwarding stats, plus an optional "Custom"
radio profile — a dedicated frequency/SF/BW/CR for relaying, separate
from the companion's own network, band-matched to the companion
frequency by default and used everywhere a radio change can happen
(boot, on-device toggle, app-driven CMD_SET_RADIO_PARAMS, Settings
radio edits) so the device never silently falls back to the wrong
params mid-relay.
Diagnostics gains the actually-forwarded packet count, real heap-free
via mallinfo(), a reset-counters popup, and hardened loop-detect
bounds. Also: a frequency-floor fix and a float-equality preset-match
fix in the repeater profile logic, deduped BW-table/valCol/profile-
seeding helpers shared between Settings and the Repeater screen, and
a build.sh fix tolerating control characters in `pio project config`'s
JSON dump.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Reply-bot overhaul on top of the trigger/reply auto-reply:
Robustness
- single botTriggerMatches() shared by DM/channel paths; named constants
(BOT_REPLY_COOLDOWN_MS / BOT_SCRATCH) replace magic numbers.
- per-contact DM throttle (8-entry ring) so a second sender isn't starved
while one contact is on cooldown.
- channel anti-loop: skip only when an incoming message equals our own reply
(was: any reply containing the trigger word — which silently killed channel
replies for the common "trigger word in reply" setup).
Features
- away / reply-to-all: a lone "*" trigger matches every message.
- independent DM vs channel triggers (bot_trigger / bot_trigger_ch); "*" works
on the channel too, bounded by cooldown + echo guard.
- query commands: a DM or monitored-channel message is scanned for "!word"
tokens — !ping/!batt/!loc/!time/!temp/!hops/!status/!help — merged into one
combined reply. DM = per-contact throttle, ignores quiet hours; channel =
broadcast, per-channel cooldown, respects quiet hours. !hops uses
getPathHashCount() (0 = direct).
- quiet hours (bot_quiet_start/end, local, wraps midnight) silence push replies.
- reply counter shown in the BotScreen header.
UI / storage
- BotScreen: 9 rows, scrolling list (no more cramming), field I/O via
fieldBuf()/fieldCap().
- prefs bot_commands_enabled / bot_quiet_start / bot_quiet_end / bot_trigger_ch
persisted; schema 0xC0DE000A → 0xC0DE000C with clamping + migration
(bot_trigger_ch seeded from bot_trigger on upgrade).
- docs: tools_screen bot section rewritten; FEATURES.md refreshed.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- FEATURES audit: refresh references the Nearby/Trail refactors made stale —
M4 (renderDiscoverDetail → renderScanDetail), L1 (SNR %.1f now in scan detail
+ ping; list cards show RSSI), Trail grid (round-step renderGrid + MIN_GRID_PX).
- MyMesh: drop the obsolete `// TODO: add expected ACK to table` — the code
right below already populates expected_ack_table.
- PopupMenu: render() is now the single source of truth for _scroll, clamping
the selection into view from the live-height cap, so handleInput() only moves
_sel and never depends on a stale _cap (M3). Behaviour-preserving.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
End-to-end delivery indicators on the device UI, drawn next to outgoing
messages and auto-scaled to the font (legible on landscape e-ink).
DM (and room servers, which share the DM path):
- Pending / delivered / failed marker driven by the real end-to-end ACK.
Pending shows a row of dots — one per send — so auto-resend progress is
visible before it resolves to ✓ / ✗.
- Auto-resend: new pref dm_resend_count (0-5, default 2), Settings ›
Messages › Resend. A pending DM whose ACK times out is re-sent (reusing
the original timestamp) until resends run out, then ✗. Driven from
UITask::loop so it completes in the background, independent of screen.
- Incoming dedup: a retry reuses the sender timestamp + text but carries a
fresh packet hash, so addDMMsg drops copies matching prefix+ts+text.
Channels (flood, no recipient ACK):
- ✓ only once a repeater echo confirms the send was relayed into the mesh;
no echo is normal, not a failure (no pending/fail shown). A small ring
tracks a burst of sends so each matches its echo. Receive-path hashing is
gated so the hot flood path is untouched when idle.
Shared:
- Markers shown in both the history list and the fullscreen message view.
- Reusable scalable mini-icon facility in icons.h (bitmap + auto-scale);
adding a new status icon is a bitmap plus one draw call.
No changes to the upstream mesh library (src/).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two independent, default-off toggles under Settings › Radio.
Pwr save: hardware RX duty-cycle (SX126x SetRxDutyCycle via
startReceiveDutyCycleAuto). The chip cycles RX↔sleep and wakes on a preamble —
no MCU state machine; recvRaw reads the packet exactly as in continuous RX.
Falls back to continuous RX on non-SX126x. (Replaces an earlier software-CAD
state machine that fought the hardware: polling a warm-sleeping chip gave a
phantom-busy channel that stalled TX ~4 s and dropped ACKs in the scan gaps.)
Auto pwr: Adaptive Power Control. tx_power_dbm becomes a ceiling; actual TX
power tracks the reverse-link SNR margin (measured above the per-SF demod floor,
EWMA-smoothed, proportional step with a deadband). Feedback comes from direct /
room-server ACKs and, for channels (no ACK), from hearing a repeater rebroadcast
our own flood; a lost confirmation ramps power back up so channel sends can't get
stranded below what the repeaters can hear.
Prefs schema 0xC0DE0009 (rx_powersave, tx_apc). Radio page / name bar show the
live TX power; noise floor reads n/a while duty-cycling.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Reverts onDiscoveredContact to the upstream 4-param signature, adding a
separate virtual onDiscoveredAdvert(bool was_flood) with a default no-op
implementation. Only MyMesh overrides the new virtual; simple_secure_chat
and any future implementors need no changes on upstream merges.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Self-review caught a regression I'd just introduced: feeding hasConnection()
from isClientConnected() (BLE *or* USB) broke the GPX-export collision warning
in TrailScreen, which relies on hasConnection() meaning "BLE app connected"
(BLE → USB dump is safe; otherwise the app may be on USB → warn).
Correct split:
- hasConnection() ← isBLEConnected() — BLE-specific consumers: BT status
indicator, pairing PIN, GPX-export warning.
- isClientConnected() (BLE bonded OR open USB-CDC port) used directly only by
the Auto buzzer mute and message-wake, which should react to either transport.
Builds clean on dual (OLED + e-ink) and pure-BLE.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
USB client presence IS detectable on nRF52 after all: (bool)Serial ==
tud_cdc_n_connected() (DTR — a host has the CDC port open). Add
isClientConnected() = BLE-bonded OR USB-CDC-open; DualSerialInterface
overrides it, base defaults to isConnected() (single-transport unchanged).
Wire it so each consumer gets the right signal:
- hasConnection() ← isClientConnected(): Auto buzzer mute + message-wake now
trigger on BLE or an open USB port (PR #14's intent, done correctly), but
not on charging-only (no host → DTR low).
- BT status indicator + pairing PIN stay on isBLEConnected() (BLE-specific).
Caveat: a plain serial monitor also asserts DTR, so it counts as connected.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
PR #14 (advert sound) changed setHasConnection() from isBLEConnected() to
isConnected() to also mute on USB — but DualSerialInterface::isConnected() is
hardcoded true (USB is always a send fallback) and ArduinoSerialInterface's is
too ("no way of knowing"), so USB-client presence is undetectable. The result
on every dual build: hasConnection() always true, which broke everything that
means "a companion app is connected":
- Auto buzzer mute → permanently muted (Auto behaved like Off)
- BT status indicator → always shown connected
- message-wake → display never woke for new messages
Revert to isBLEConnected() (the only detectable client signal). Mute-on-USB
isn't achievable; Buzzer = Off covers that case manually.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
M2: CMD_SET_DEFAULT_FLOOD_SCOPE used strlen() on the frame's 31-byte name
slot, which is not required to be NUL-terminated. Switched to strnlen()
so the search can't run past the field into the 16-byte key.
M4: NearbyScreen::renderDiscoverDetail computed strncpy(b64, ..., max_chars - 3)
where max_chars came from display width. On very narrow displays (width < ~28
at 6 px font) this became negative. Skip the pub-key line entirely when
max_chars < 4 so we don't risk a negative count and a bogus terminator.
L1: SNR was shown as truncated integer dB. Switched the detail view and
the 2-line discover cards to %.1f so they keep the 0.25 dB resolution
(consistent with the ping popup, which already used %.1f).
L4: Two fallback "?" sender placeholders used strncpy(buf, "?", sizeof(buf))
— functional but it memsets 21 unused bytes for a one-character string.
Replaced with strcpy.
M1 marked as not-a-bug after re-check: default_scope_name is char[31],
so the n < 31 guard correctly admits the max 30-char string + NUL.
FEATURES.md audit section updated with current status.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
C1/C2: onChannelMessageRecv and onChannelDataRecv both passed
findChannelIdx() through (uint8_t), so an unknown-secret packet
turned -1 into 255 and the message flowed into the offline queue,
UI history, notification and bot reply with a bogus channel index.
Drop such packets at the recv path; also harden addChannelMsg with
an MAX_GROUP_CHANNELS bounds check so a future caller can't poison
the channel-history ring buffer either.
H4: loadPrefsInt always reset trail_units_idx on sentinel mismatch,
even for jumps where the field was saved correctly (e.g. 0xC0DE0004
→ 0xC0DE0005). Scope the reset to sentinel == 0xC0DE0003, which is
the only case where the byte at that offset is actually the stale
sentinel low byte.
FEATURES.md: mark these three items as ✅ in the audit backlog.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Adds GxEPD2 framebuffer capture to the screenshot tool, extending the
existing OLED support to the Wio Tracker L1 e-ink variant.
Key changes:
- GxEPD2-patch: patched copy of GxEPD2_BW.h exposing getBuffer/getBufferSize
- GxEPDDisplay: override getBuffer/getBufferSize/getDisplayType/
screenshotWidth/screenshotHeight/screenshotRotation using GxEPD2's
own reported visible dimensions (WIDTH_VISIBLE, not physical WIDTH)
and live rotation value
- DisplayDriver: virtual screenshot interface so no C-style casts needed
- Protocol: 11-byte header with display_type, rotation, uint16 LE
width/height/chunk_idx/total_chunks (eliminates truncation for panels >255px)
- screenshot.py: full rot 0-3 decoder for GxEPD2 buffer layout; ERR frame
checked before length so device errors surface correctly; buffer-size
sanity check using panel-aware expected size
Conflict resolution: kept feat/eink-screenshot protocol and decoder
(11-byte header, e-ink decode) over the OLED-only 5-byte version on
wio-unified; battery and joystick fixes from wio-unified kept intact.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
handleScreenshotRequest() reinterpret-cast every DisplayDriver* first to
SH1106Display* then to SSD1306Display*; both casts always produce a
non-null pointer, so the SSD1306 fallback was unreachable and on any
SSD1306 build we called SH1106::getBuffer() on a SSD1306 object — UB
that only happened to work because both backing classes share an
Adafruit_GFX layout at offset zero.
Replace the cast hack with virtual getBuffer()/getBufferSize() on
DisplayDriver, overridden in SH1106Display and SSD1306Display. MyMesh
no longer needs to know about either concrete type. const-correct the
buffer pointer and widen bufferSize to uint16_t while passing through.
Also fix the matching Python decoder: it sanity-checked length before
dispatching on resp_code, so a 2-byte RESP_CODE_ERR frame was reported
as "Frame too short" instead of the actual device error. Move the
length check after the resp_code dispatch and log the error code byte.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Firmware (7-byte header):
Byte 7 = display->screenshotRotation() — GxEPD2/GFX rotation value (0-3).
DisplayDriver::screenshotRotation() defaults to 0 (OLED).
GxEPDDisplay::screenshotRotation() returns display.getRotation(), the
live GxEPD2 value (reflects runtime setDisplayRotation() calls, not just
the compile-time DISPLAY_ROTATION macro).
Python decoder:
- Parse 7-byte header; pass rotation to eink_buffer_to_image().
- Implement all four GxEPD2 drawPixel coordinate transforms:
rot 0: phys_x=lx, phys_y=ly
rot 1: phys_x=vis_w-1-ly, phys_y=lx
rot 2: phys_x=vis_w-1-lx, phys_y=vis_h-1-ly (180° flip)
rot 3: phys_x=ly, phys_y=vis_h-1-lx
- vis_w/vis_h derived from log dimensions + rotation parity (no constants).
- Print rot= in the status line so the user sees which rotation was used.
This fixes the 180° rotated image seen with rotation=2 (portrait inverted)
without hardcoding a flip — correct for any rotation value.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Firmware: add screenshotWidth()/screenshotHeight() virtual pair to
DisplayDriver (default: width()/height()). GxEPDDisplay overrides to
return display.width()/display.height() — the GxEPD2-reported values
that use WIDTH_VISIBLE (e.g. 122 for GxEPD2_213_B74), not the full
physical WIDTH stored in DisplayDriver (128). MyMesh uses these
instead of display->width()/height() for the screenshot header bytes.
Result for GxEPD2_213_B74 + DISPLAY_ROTATION=1:
header was 250×128 → now 250×122 (correct visible canvas)
Python decoder (tools/screenshot.py):
- Remove hardcoded EINK_PHYS_WIDTH=128 / EINK_VISIBLE_W=122 constants.
- Derive phys_stride from buffer_size / log_width (works for any panel).
- Fix phys_x formula: was (EINK_PHYS_WIDTH-1-ly = 127-ly),
now (vis_w-1-ly = log_height-1-ly = 121-ly for 213_B74).
Old formula addressed invisible columns 122-127; new formula correctly
maps logical rows 0..121 to visible physical columns 121..0.
- vis_w = log_height (no hardcoded trim; header now carries correct value).
This also works for square panels (e.g. 128×128 where WIDTH=WIDTH_VISIBLE):
header will carry 128×128, decoder produces a correct 128×128 image.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add lib/GxEPD2-patch/src/GxEPD2_BW.h: local patched copy of GxEPD2_BW.h
that exposes getBuffer()/getBufferSize() under ENABLE_SCREENSHOT guard.
Include guard (_GxEPD2_BW_H_) prevents double-inclusion of the installed
library version.
- GxEPDDisplay.h: use patched header via relative include when
ENABLE_SCREENSHOT so the patch takes precedence over the installed lib
(PlatformIO adds library paths before -I build_flags).
- DisplayDriver.h: add virtual getBuffer()/getBufferSize()/getDisplayType()
defaults (nullptr/0/0) under ENABLE_SCREENSHOT.
- SH1106Display.h / SSD1306Display.h / GxEPDDisplay.h: add concrete overrides;
getDisplayType() returns 0 (OLED) or 1 (e-ink).
- MyMesh.cpp/h: replace fragile C-cast with virtual dispatch in
handleScreenshotRequest(); extend 5-byte header to 6 bytes by appending
display_type so the host tool can decode the correct pixel layout.
- tools/screenshot.py: parse 6-byte header; add eink_buffer_to_image()
that decodes the row-major MSB-first GxEPD2 buffer with DISPLAY_ROTATION=1
(phys_x = 127-ly, phys_y = lx); dispatch on display_type.
- variants/wio-tracker-l1-eink/platformio.ini: add
[env:WioTrackerL1Eink_companion_dual_dev] with ENABLE_SCREENSHOT.
- variants/wio-tracker-l1/platformio.ini: unchanged (OLED env already exists).
Builds verified: WioTrackerL1Eink_companion_dual_dev SUCCESS,
WioTrackerL1Eink_companion_radio_ble SUCCESS (unaffected).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
