26e678e57034230e16616ce22c0159f6c791aef6
597
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 arcattackandClaude Opus 5
|
26e678e570 |
#67 part 2, ROOT-CAUSED AND FIXED: 1995 latent uninitialized fields, exposed by the port's allocator
THE HUNT (the deterministic rig repro made it a two-hour arc):
1. cdb write-watch armed from the Torso ctor (bp plants `ba w4 this+0x21c`
per torso -- ASLR-proof). The poison reproduced (atUpd=-1250 this run;
-3750 and int-15-as-float before)... and the watch stayed SILENT. Nobody
writes the garbage. The field is never INITIALIZED.
2. Confirmed in our ctor reconstruction: it inits every neighbour but skips
targetTwist @0x218 and twistAtUpdate @0x21C (the function that zeroes
them is a death-reset handler, not the ctor).
3. Confirmed in the BINARY: the real ctor @004b6b0c contains no store to
either offset -- a genuine 1995 latent bug (uninitialized read on the
copy path).
4. Why the pod never showed it: MemoryBlock arenas carve fresh OS-zeroed
pages, one pool per type, low churn -- first allocations read as zero.
The engine's own DEBUG_NEW_ON NaN-fill proves the developers knew the
hazard class. Why WE show it: mechrecon.hpp's Memory::Allocate shim
("a plain heap allocation is behaviour-equivalent" -- false) recycles
dirty heap. Replicant torsos spawned with garbage twist targets; the
limit clamp turned any garbage into FULL TWIST -- rendered "twisted full
right while not using TT" to every peer (playtest night 4, 3 reporters).
THE FIX -- environmental, class-wide, byte-faithful to the binary's code:
Memory::Allocate / AllocateArray / Alloc now ZERO-FILL, reproducing the pod's
EFFECTIVE allocation semantics for every never-stored field in every
reconstructed factory at once (Torso, Reservoir, all of them). No ctor gains
stores the binary lacks.
PROVEN on the rig: the replicant thor now spawns cur=0 target=0 atUpd=0
(was cur=-3.31613 = hard against the limit). Teardown clean.
Ships with part 1 (the dead-reckoning clock fix) in the next zip. Remaining
on #67 for the next games night: live confirmation that twist TRACKING looks
right in play, and whether the fire-from-centre facet (very plausibly the same
never-stored-field class, now zeroed) is cured with it.
Tools kept: scratchpad/torso_watch.cdb + rig_watch.ps1 (the ctor-armed
write-watch pattern -- reusable for any "who wrote this field" hunt).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
a3329a64da |
#67 torso replication, part 1: the dead-reckoning clock fix + the bug REPRODUCED on the rig
THE CLOCK FIX (landed, verified live). The replicant torso's extrapolation
window was hardwired to ZERO: both shim accessors returned the same field
(torso.hpp, the "dormant in single-player" note). The engine maintains both
clocks exactly as the binary expects -- WriteUpdateRecord stamps
lastUpdate=lastPerformance on the master, ReadUpdateRecord stamps
lastUpdate=Now() on the replicant (SIMULATE.cpp:276/296, the 1995 "HACK"
comment intact) -- so the fix is one mapping: GetCreationTime (misnamed; now
GetLastUpdateTime) reads lastUpdate. ComputeTargetTwist's
`current - lastUpdate` window is real again. Why legs always replicated
while torsos did not: Mover is ENGINE code reading these fields natively; the
Torso is our reconstruction with the collapsed shim. Rig-verified: the
copy-side log now shows distinct lastUpd/now values.
THE FIELD BUG REPRODUCED (2-pod rig, thor vs thor -- thor because the first
attempt used FOGDAY's Black Hawk, whose torso is authentically FIXED like the
Owens: hEn=0, limits +/-0.01 deg):
[torso-copy] cur=-3.31613 target=-3.31613 atUpd=-3750
The replicant renders at EXACTLY its twist limit (thor: +/-3.31613 rad)
because twistAtUpdate holds garbage (-3750) and the limit clamp slams it to
the stop -- "twisted full right but was not using TT", on demand. An earlier
run poisoned it with 2.1e-44 (= int 15 as float, suspiciously the torso's
subsystem index).
NARROWED: the tx/rx probes added here (BT_TORSO_LOG prints every torso record
both directions, header + raw payload dwords) logged ZERO records in the
poisoned runs -- Torso::Read/WriteUpdateRecord never executed. The garbage
therefore arrives OUTSIDE the torso record path: a raw state write during
mech spawn/consolidation spraying the torso's fields (stream-walker framing
or a consolidation blit). That writer is part 2's hunt; the probes and the
deterministic repro make it a short one.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
9bb95d8580 |
steam-networking KB: the first real multi-player field verification (night 4)
Three back-to-back Steam lobbies, 3+ players, host-and-join, no operator console: launch worked, DAMAGE APPLIED (previously suspected dead over Steam), 2 kills, clean round end, next mission launched. Records what worked, the one genuinely Steam-side defect found (#68 silent exit on a failed join), and the triage lesson -- a bug found in a Steam lobby is only a Steam bug if it lives in the lobby/token/transport layer; the three replication symptoms that night produced (#67/#70/#55) are above the seam and would reproduce on the relay. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
184121c597 |
kill-storm repro rig for #35 (owens laser crash): parameterized damage hook + honest NEGATIVE
BT_MP_FORCE_DMG=<n> now sets the per-tick probe damage (plain =1 keeps the original amount) -- kill-storm rigs need lethal ticks. rig_killstorm.ps1: offset-port relay + owens shooter under BT_AUTOFIRE vs a respawning victim. Result: NEGATIVE, twice. The respawn cycle (death anim + warp + handshake) caps the harvest at 1-2 kill-teardown windows per 4-minute round, and none crashed. With July's 254-volley negative the conclusion firms up: the window needs the reporter's slow-machine timing, not more attempts here. The field net (crash self-report + join.old.log rotation + the sweep guards) means the next real occurrence names its own site. Refined theory recorded in the ledger: six-beam volley vs a target dying mid-volley, 515-class teardown race. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
35bdb505df |
join bats keep the previous session's log as join.old.log
Conn Man's owens-laser crash stack (#35) was in his join.log -- and re-running join.bat deleted it before it could be sent. The bat now rotates instead: join.log -> join.old.log at launch. One generation of history, the exact insurance that would have closed #35 tonight. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
ecb9b338bd |
the X button means QUIT: a user-closed window exits for real instead of relaunching
Operator report (00:30): "my local instance auto-relaunches whenever I stop and
start a session, even if I close my window." Root cause is the pod's AUTHENTIC
immortality: an arcade cabinet's join loop never exits, so every RunMissions
return relaunches into the join wait -- and a human closing the window was
indistinguishable from a round ending. On a desktop that made the client
unkillable outside Task Manager, and explains every orphaned-plasma zombie
hunted this week (my own test teardowns included), plus tonight's "eventually
all I could find was my plasma display".
FIX, one flag + one gate:
* btl4main WndProc WM_CLOSE stamps gBTUserRequestedExit before DestroyWindow
(covers X, Alt-F4, taskbar close, Task Manager's polite phase).
* BTFE_RelaunchSelfAndExit -- the single choke point every relaunch path
funnels through (round end, console-pad lost, round abort, FE loop) --
exits for real when the flag is set, logging 'window closed by the user
-- exiting for real (no relaunch)' to the marshal log.
Round-end auto-rejoin is UNTOUCHED: the flag only sets on WM_CLOSE.
VERIFIED both directions on the built exe (4.11.591):
* WM_CLOSE posted to a live solo instance -> the whole process tree is GONE
8s later: no relaunched generation, no plasma orphan. [was: immortal]
* Offset-port rig, full launch -> StopMission -> both pods AUTO-REJOINED and
re-ACKed for the next round (2/2 ready, WAITING FOR OPERATOR); zero
exiting-for-real lines fired (nobody closed a window).
Requires the next zip for players; the operator's running instance keeps the
old behavior until restarted onto this build.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
33487ab846 |
night ledger: post-night fix session status (guards verified, dup-pilot repro NEGATIVE, five fixes landed)
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
ef0ec48f9a |
map dropdown = the AUTHENTIC console catalog; the relay names a phantom-map stall instead of holding silently
THE 40-MINUTE OUTAGE'S REAL FIX. The GUI's map dropdown offered every RES
name passing the type-14+26 existence check -- which includes INTERNAL
FRAGMENTS: artrucks is an include-node of arena1 (PROGRESS_LOG map anatomy:
arena1 -> {arenall -> cavern, artrucks}), not a mission. Picking it stalled
every pod's mission load forever with no error anywhere, and End/Re-arm kept
restoring the poisoned egg -- 22:02..22:49, three "different" failures, one
cause.
* eggmodel.CONSOLE_MAPS: the Mac 4.10 operator console's own adventure-tree
catalog (Console.ini; the same 8 the solo menu ships in btl4fe.cpp kMaps):
cavern grass rav polar3 polar4 arena1 arena2 dbase. ValueSets.maps now
offers exactly that, catalog order, filtered to what the RES carries
(permissive fallback only if the intersection is empty -- a foreign RES
must not present zero maps). artrucks is the one fragment the old filter
let through; now hidden.
* The relay WARNS at egg load/reload when the mission names a non-catalog
map (the camo-color-warning precedent: hand-edited eggs still work, the
operator just cannot miss it).
* The launch hold gains the PHANTOM-MISSION SIGNATURE diagnostic: 0/N ready
for 60s means every pod is stalled in shared mission content -- one line
naming the egg's map and the recovery (End Mission -> fix map -> Re-arm),
instead of the silent 10s HELD drumbeat the operator stared at for 40
minutes. Hint re-arms per launch.
VERIFIED: new suite scratchpad/test_map_guard.py (5 checks: catalog exact +
ordered, artrucks hidden, warning fires on a doctored egg, silent on a real
one, the 60s hint names the map); all five console suites pass. Python-only.
(While placing the hint, a blind text-insert broke the launch-received block's
indentation -- caught by py_compile before anything ran, repaired, and the
whole area re-verified by the rearm suite.)
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
becaddb662 |
console logs ROTATE instead of truncate -- tonight's evidence loss cannot repeat
Two destroyers, both in btoperator.py:
* Start Session zeroed operator_relay.log -- took the whole 23:08 session
(the reservoir-crash round's relay trace) minutes after it happened;
* Launch local instances os.remove()'d operator_N.log -- took the pod's
[crash] self-report block before it could be copied.
Both now rotate the previous file to <name>.1 (btrelay_park's pattern: drop
the old .1, os.replace, never block on a locked file). One generation of
history is exactly what post-mortems needed twice tonight. All four console
suites pass; picks up on the console's next restart.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
b7e4837738 |
seat identity swap (SAURON played as 'Draco'): the LEFT handler ate the pref of a re-seating player
THE CHAIN (reconstructed from the night's logs, then REPRODUCED offline in
scratchpad/test_seat_identity.py before fixing):
1. At round end every pod relaunches and seat-requests again. SAURON's
request was walk-up-assigned the departed Draco's old seat, and his pref
(callsign/mech) was correctly written -- the assign line even printed
callsign='SAURON'.
2. His request conn then closed BY DESIGN (the pod re-dials to HELLO) --
and the beacon-death branch of _drop_game, added 2026-07-26 for the
departed-player roster fix, treated any beacon death on an unclaimed
seat as a LEAVE: it printed a false 'PLAYER n LEFT' and POPPED the pref
written milliseconds earlier.
3. The next egg release _reload_egg_file()'d the DISK egg -- where the GUI's
Start Session had saved the ADOPTED roster names, including 'Draco' on
that row -- and with no pref left to override it, Draco's callsign
shipped on SAURON's seat. His plasma (and that round's score
attribution) wore the wrong name. Per-seat HELLO-vs-close ordering
roulette explains why only one seat swapped.
THE FIX: a LEFT-grace window (SEAT_LEFT_GRACE_SECONDS=15). A beacon that
dies younger than the grace is the pod's designed post-assign re-dial: keep
the pref and the reservation, print nothing. A real join-menu leaver has
held the seat far longer, so the departed-player roster fix keeps working
(pop + LEFT exactly as before); claimed seats were already exempt.
REGRESSION SUITE (new, offline, drives the real Relay class -- no ports):
A. designed instant close: pref survives, seat stays protected [was FAIL]
B. real leave (aged beacon): pref popped + seat freed [unchanged]
C. claimed seat: pref survives an unrelated conn death [unchanged]
D. a different identity assigned onto a held seat overwrites
the held pref (the operator's original suspicion, locked in) [unchanged]
All four console suites pass (rearm 25, net 17, roster 22, identity 7).
Python-only: no client update needed; the running console picks it up on its
next Start Session. Awaiting live verification next games night.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
e28dcbc161 |
the #35-class sweep + the games-night load crash: four lifecycle-window null-derefs guarded
THE CLASS (established by Gitea #35's 515 fix, confirmed growing tonight): code the binary could run bare because a pod's mission world always existed, crashing in the port's MP join/teardown/respawn windows where mission, player, or a roster link is briefly NULL. Swept every GetMissionPlayer()/ GetPlayerVehicle()/GetCurrentMission() chain in the tree (23+7 sites). GUARDED (the 515 pattern -- authentic behavior whenever the object exists, skip/degrade + loud log when mid-teardown): 1. DPLRenderer::SortAndReloadNameBitmaps -- runs in the round-STOP path (the end-of-round circle), raw mission+player+entity-manager derefs. 2. DPLRenderer::LoadOrdinalBitmaps -- called from (1); raw GetMissionPlayer()->GetInstance() evaluated on EVERY machine (the camera- director test derefs before it branches). 3. DPLRenderer::LoadNameBitmaps -- same window, raw GetPlayerCount() chain. 4. Reservoir::Reservoir master-gate block -- THE CAPTURED GAMES-NIGHT CRASH ([crash] btl4+0x4990d = this ctor inlined into CreateReservoirSubsystem, AV reading NULL+0x1d0): linkedSinks.Resolve() derefed raw; +0x1d0 is exactly the masterScale FILD of the resolved master heat-sink bank (heatSinkCount @0x1D0 -- the offset the crash named). Observed trigger: a DUPLICATE-PILOT egg (the seat-ghost bug put one identity in two seats) poisoning the ID registry, so Resolve() nulled during viewpoint-entity construction and at least two pods died loading the same egg. Now: loud '[spawn] FATAL-AVOIDED' + degrade to the inactive-copy shape. The duplicate-pilot ROOT CAUSE is the seat-ghost fix (tracked separately); this guard makes the failure survivable and self-identifying either way. CLASSIFIED SAFE, not touched: btl4mppr/btplayer/mechmppr/btl4gau3 chains (guarded or null-tolerant by design); APP.cpp's five Check(player) sites (mission state machine -- player exists by construction, original engine ordering); CAMMGR/CULTURAL GetCurrentMission chains (construction-time). VERIFIED: build clean (0 errors, no new unresolved externs); solo smoke on the guarded exe -- zero FATAL-AVOIDED lines (guards silent on the healthy path), subsystems simulating, no faults. STILL OWED: a full launch->stop rig cycle (exercises the guarded round-stop path live) and the duplicate-pilot-egg repro -- both blocked on port 1500 while the operator's console session is up. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
c7dcdf26eb |
games night 2026-07-26: full incident ledger (artrucks phantom map, reservoir load crash, seat ghost, owens laser suspect)
The night's evidence file, preserved against the log truncation that ate the primary sources twice. Highlights: map=artrucks (an art sub-node the dropdown offers as a mission) stalled every load for 40 minutes; the post-freeze reload crash resolved to CreateReservoirSubsystem+0x13d (null->0x1d0) with a duplicate-pilot egg as prime suspect; a held seat walk-up-assigned to a different player kept the old callsign (SAURON played as 'Draco'); and the 23:14 first domino was Conn Man in OWENS firing lasers -- matching an older verbal report never filed. Fix list + evidence-collection asks recorded. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
d3e724c254 |
cut 4.11.584 for tonight, and PROVE it still plays with the 554 everyone has
The merge brought Cyd's cockpit refit into the build testers will run tonight,
so the question that decides the evening is whether a 554 client can still join.
Rather than reason about it, tested it: extracted the real BT411_4.11.554.zip
and ran ONE 554 pod and ONE 584 pod against the same relay through a real
launch.
WIRE-COMPATIBLE. Both staged, both REGISTERED, the relay launched the mixed
round, and 30s in both were still registered AND still on the UDP fast path
(registered [2, 3] udp-known [2, 3]), zero drops, both processes alive.
=> NOBODY HAS TO UPDATE TONIGHT. The new zip is an upgrade, not a
flag day (unlike 554, which changed the scoreboard wire format).
Supporting evidence for why that held: the merge did NOT change the attribute
table's shape -- 36 ATTRIBUTE_ENTRY(Mech,...) rows before and after. Cyd's
crouch work POPULATED an existing pad slot (0x37 DuckState, previously
attrPad), so no index shifted.
The zip itself (dist/BT411_4.11.584.zip, 47.0 MB):
* built clean at HEAD
|
||
|
arcattackandClaude Opus 5
|
2b0506ddfd |
OPERATOR_GUIDE: fix the blockquote swallowing the secret/port paragraph
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
6b0402badc |
park_relay.cmd: resolve a real interpreter -- a second admin account has NO working python
Setting up remote access exposed this: 'python' works for the operator's own account only. It resolves through pyenv SHIMS driven by PYENV/PYENV_HOME/ PYENV_ROOT, and all three are USER-scoped to that account; the single machine-wide PATH entry points at a Python39-32 folder that no longer exists. So a dedicated remote-admin login (the account just created for SSH/RDP) gets NO python at all -- parking the relay from the road would have failed with 'python is not recognized', at night, with players waiting. park_relay.cmd now resolves an interpreter explicitly and prints which one: BT_PYTHON override -> the project's pyenv 3.11.4 (calling account's profile, then the operator's -- Administrators can read it) -> the machine-wide py launcher -> whatever is on PATH -> a clear error naming the fix. Verified both as the operator and with a simulated second-account environment (machine PATH only, PYENV cleared): both resolve 3.11.4. Also dropped a delayed-expansion bug in the first cut (!CAND! without EnableDelayedExpansion would have silently evaluated to nothing). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
05fdd319d6 |
parked relay: a supervisor that keeps it alive + a REMOTE restart the operator can actually reach
Completes the travelling-operator deployment. The relay stays parked on the
home machine (so no player ever edits join.bat) and the console dials in from
anywhere -- but until now two things needed hands on that machine: bringing the
relay back when it died, and restarting it (the only way to clear a wedge or
pick up a roster resize). Both are covered now.
tools/btrelay_park.py -- the supervisor:
* runs the relay with cwd=content\ , which is what pins WHICH
operator_secret.txt is live (the trap documented last commit);
* relaunches on ANY exit, with 2/5/15/30/60s backoff when a relay dies inside
20s, so a permanent fault (port taken, missing egg) cannot become a spin;
* rotates content\parked_relay.log to .1 first, so the dead generation's
evidence survives the relaunch that replaces it;
* Ctrl-C stops it for good. NOT a Windows service on purpose: session 0
would hide the window an operator wants to tail.
tools/park_relay.cmd -- double-click to park; a shortcut to it in shell:startup
gives start-after-reboot with no admin rights.
`restart` on the control port (btconsole.py): sets restart_requested, the run
loop returns, relay_main exits RESTART_EXIT_CODE=42 and the supervisor relaunches
-- re-reading the egg. REFUSED while a mission is running (launches_sent >= 2
and not stop_sent): bouncing then would drop every pod out of a live round.
Local stdin gets the same command for parity.
btoperator.py: Restart Session in REMOTE mode now sends `restart` and reconnects
6s later instead of just tearing down our own link -- the old behaviour looked
like "Restart does nothing" against a parked relay. (QTimer had to be imported;
it was missing, which py_compile does not catch -- it would have been a runtime
NameError on the first click.)
VERIFIED by scratchpad/test_parked_supervisor.py, 12/12, and the full
test_remote_console_e2e.py re-run green afterwards (15/15):
* kill the relay -> a NEW pid is serving again, and the old log is kept as .1;
* remote `restart` -> ACKed, new pid, egg re-read, serving again;
* the supervisor distinguishes the two -- "operator requested a restart" vs
the crash/backoff path -- proven from its own log, not assumed;
* the mid-mission guard is present and wired to launches_sent/stop_sent.
Two harness defects fixed while proving it, both mine: a check written with
`or True` that could never fail (replaced with two real assertions against the
supervisor's log), and a recv that treated the relay's correct
close-after-restart as a ConnectionResetError failure.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
05f1ffb194 |
verify the parked-relay/remote-console path END TO END (it works; 15/15)
The operator asked whether the console is actually set for the parked-relay
deployment or whether I was just vouching for earlier work. Honest answer was
that prior verification was unit-level (relay log lines fed to an offscreen
widget) plus LOCAL-relay rig cycles -- the remote path had never been driven for
real. Now it has:
parked relay (standalone btconsole.py, cwd=content, all interfaces)
+ 2 real pods dialling in
+ the REAL operator GUI in REMOTE mode over a real TCP control socket
15/15, twice, with clean teardown. Proven: AUTH; the relay REGISTERS both pods;
the GUI adopts the roster and lights the seats; LAUNCH / END MISSION / Re-arm all
ENABLE over the remote link (the exact things that were broken before today);
LAUNCH really reaches the pods (RunMission pair) and END MISSION really stops
them; mission settings cross the wire and rewrite the relay's OWN egg
(map=cavern time=night weather=soup verified in the relay log); and the parked
relay SURVIVES the operator disconnecting -- the property the whole deployment
depends on.
A LAN address is used deliberately, not loopback: btoperator treats
localhost/127.0.0.1 in the relay-host field as LOCAL mode, so a loopback test
would have silently exercised the wrong code path and proved nothing.
Three defects found in my own harness on the way, all fixed here (none in the
product):
1. Pods announced BT_SELF on the LAN address while FOGDAY.EGG's roster lists
127.0.0.1:1502/1602 -- seat identity mismatch, so they took the egg and
closed. The relay was RIGHT to refuse: "LAUNCH pressed but NO players are
seated yet". Pods now announce the address the egg's roster lists.
2. The roster assertion counted EGG-CONFIGURED callsigns (always present) and
would have passed with zero live players. It now counts the relay's own
REGISTERED lines.
3. The relay's `set` really rewrites its egg -- pointed at a TRACKED file it
dirtied content\FOGDAY.EGG (caught by git status, restored with git
checkout). The test now runs on a throwaway copy and deletes it.
Also: cleanup now walks the PROCESS TREE, because btl4's front-end relaunches
itself for the mission, so the surviving pod is a CHILD of the spawned PID --
killing only our own handles left a live pod holding its log open. Still
PID-only, never by image name.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
9df7dff787 |
KB: the parked-relay/remote-console mode is COMPLETE -- record it so it stops being re-derived
Operator caught me re-investigating a feature we already shipped today. Root
cause was a self-contradicting knowledge base, not just my compaction: the
§Modes entry still said "Remote relay -- KNOWN BROKEN: the LAUNCH button can
never enable in this mode", while §Mode-specific traps 60 lines later documented
the same thing as FIXED. Reading the summary line first is exactly what a
future session does. Swept: that was the only stale copy.
Now recorded once, authoritatively (context/operator-console.md §Parked relay +
remote console, and a sysop recipe in docs/OPERATOR_GUIDE.md):
* THE PRODUCTION TOPOLOGY the operator actually wants: the relay parked
permanently on the home machine so no player ever edits join.bat, with the
console dialling in from anywhere (cell internet is fine -- outbound only).
* The exact standalone park command, and why the control port is what makes a
GUI-less relay usable at all (backgrounded, it has no stdin).
* The COMPLETE remote command set as a table, read out of _ctl_command /
_ctl_get_mission / _ctl_set_mission: launch, stop, rearm|newround, get,
set (the six CONTROL_SET_KEYS, written to the relay's OWN egg, effective
next round), ping, plus the log stream and the roster re-issued at AUTH.
Written down so the next session reads instead of re-deriving.
* The hard limits: remote mode ATTACHES -- it cannot start or restart the
relay, cannot resize the roster, cannot edit callsign/mech/colour.
* Multiple operators ARE accepted concurrently (ctl_conns is a list, each
AUTHing independently), so hand-off is seamless -- but nothing arbitrates
two people pressing LAUNCH.
Two live hazards found while verifying, both new:
1. THE SECRET'S CWD TRAP. control_secret() opens a bare relative
"operator_secret.txt" and SILENTLY GENERATES A NEW ONE if absent, so the
live secret depends on the relay's working directory. This repo really has
TWO secret files with DIFFERENT values (repo root and content\) -- a remote
operator holding the wrong one fails AUTH as a CONTROL_AUTH_TIMEOUT drop,
with no "wrong password" anywhere. Documented in both files.
2. operator_secret.txt was untracked but NOT ignored -- one `git add -A` from
publishing a live credential. Now in .gitignore.
Also corrected a nearby overreach: "the console must STAY CONNECTED" is about
the console channel to the PODS; an operator GUI dropping off a parked relay's
control port is harmless and the pods never notice.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
12d6862c1f |
keylight: unship it -- opt-in (BT_KEYLIGHT=1), gone from all player docs
Operator decision: the player base is not on Windows 11, so the RGB keyboard
mirror should not be shipped or promised. The feature is NOT deleted -- Cyd's
implementation stays intact behind the gate for whoever wants it -- it is
simply off by default and invisible:
* PadRIO gate flips from opt-out (BT_KEYLIGHT=0 disables) to OPT-IN
(BT_KEYLIGHT=1 enables). Default boots produce zero keylight lines, and a
stub build and a real build now behave identically for players: nothing.
* Removed from every user-facing artifact: the README paragraph, the
CONTROLS.md section, the CONTROLS.html footer line.
* environ.ini template: the entry stays (it is the discoverable home for the
opt-in) but now says OFF-unless-1, Windows 11 22H2+, and why it is
unshipped.
* mkdist's cut-time stub warning removed -- it existed to protect a README
promise that no longer exists.
* context/glass-cockpit.md records the decision, the opt-in, and the
build-time SDK stub gate in one place.
VERIFIED: default boot logs zero keylight lines; BT_KEYLIGHT=1 engages the
feature (the stub on this machine: its one honest log line + the key map).
All three console suites pass; checkctx CLEAN.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
c4202d92ca |
zip-doc audit: last stale key claim + a keylight-stub warning at cut time
Full audit of every user-facing artifact in a freshly cut zip (33 checks, all
passing -- README substitutions/controls/migration story, CONTROLS.txt ASCII
flatten, CONTROLS.html loader + wrap, launchers byte-identical, exe carries the
migration):
* README's intro still said 'Press V any time to toggle the external camera'
-- the one key claim outside the rewritten controls block, stale for every
fresh/migrated install since the board took V. Backtick now.
* mkdist warns at cut time when the exe carries the KEYLIGHT STUB (build
machine's SDK < 10.0.22000), because the README promises the RGB feature
-- nobody should ship a stub build believing that promise ships with it.
The audit zip itself was deleted after verification: its version stamp predates
today's doc commits, and a release cut should rebuild the exe at final HEAD
first (the 554 procedure).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
e46fc706a6 |
CONTROLS.html: load your bindings.txt and the board becomes YOUR board
The interactive controls page was a static picture of the stock layout -- a
player's carried migration rows, rebinds and HOTAS setup were invisible to it,
and the mismatch grew with every customization. Now a "Load your bindings.txt"
box (file picker + drag-drop) parses the real bindings grammar client-side and
rewrites the interactive keyboard in place:
* every key shows the loaded file's truth; keys that differ from the stock
board get a hazard outline (the legend says so); hover readouts follow
because they were already attribute-driven;
* the game's bindings-row-wins rule is modelled: the PgUp/PgDn volume and
backtick view BUILT-INS keep their face only while their key is unbound;
* rows that have no keyboard geometry (pad / padaxis / wizard joyaxis-
joybutton-joyhat / MOUSE keys) land in an "also in your file" list;
* one click restores the stock view. Nothing is uploaded -- FileReader on a
user-chosen local file, works from the extracted zip on file://.
Key-name registry walks the three key blocks in DOM order (Shift/Ctrl/Alt
resolve left-then-right; the numpad's U+2212 minus and its ASCII twin both map
to NUMPADMINUS), address meanings are harvested from the stock board itself
plus a small table for the slots the default board leaves off (unwired columns,
the cabinet-only throttle bank, PANIC).
Two self-inflicted bugs found by the harness and fixed before landing: the
address harvest invented a "BTN" label for the many stock keys that have none,
which made EVERY default row read as a customization; and generated axis labels
("AIM +") differed from the page's hand-authored vocabulary ("AIM UP"), custom-
marking the whole stock numpad. Both now speak the board's own labels.
VERIFIED headlessly (no browser extension needed): scratchpad/
test_controls_page.py injects a self-test harness into a copy of the page,
feeds it scratchpad/fixture_migrated_bindings.txt -- a REAL board-2 migrated
file (full defaults + two authored rows + a wizard section) -- and runs it
under chrome --headless=new --dump-dom. 11/11: the two authored rows carried
and marked, zero stray custom marks across the rest of the board, both
built-ins survive, wizard rows listed, status line honest, and reset restores
the stock board exactly.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
6eaed84fad |
CONTROLS.html: catch it up with the volume move + the bindings migration
The interactive controls page had drifted from the markdown fix-ups: PgUp/PgDn rendered as unbound keys with no meaning (the volume keys live there now), and the Rebinding section still promised 'an existing file is never overwritten' -- made false by the board-2 migration. Both corrected; the page and CONTROLS.txt now tell the same story. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
ab91b5e7c1 |
clickbank: never target a degenerate window (the 0x0 Plasma window ate all 144 clicks)
The picker matched any visible window whose title contains the substring and took windows[0] -- which for a live game is 'BattleTech - Plasma', a 0x0-client window. Every posted click landed in it: 144 clicks, zero dispatches, and what looked exactly like the click/render alignment regression being checked for. Windows with a client smaller than 50x50 are skipped now; verified by re-running the full 72-button pass with the broad 'BattleTech' title straight through to 72/72 dispatches. (Found during the post-merge alignment regression check: boot geometry, a mid-session resize to an odd 1120x640 letterbox, and a minimize/restore cycle -- 72/72 dispatched in every round, 288 clicks total, zero Gitea #56 tripwire lines. The merge did NOT regress click-vs-render alignment.) Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
8edce41f88 |
bindings: one-time migration to the board-2 layout -- the two-population fork is gone
Operator decision: "make this version devour their custom settings and re-output
the correct binding file with the updates -- I'd like to eliminate competing
code paths." Before this, the preserved-bindings.txt convention meant upgraders
stayed frozen on whatever layout their file was written under, forever: two
player populations, two sets of true documentation, and every future default
improvement reaching only fresh installs.
THE MIGRATION (PadBindingProfile::MigrateBindingsFile, run once at load):
* A file carrying the new "# bindings-board 2" marker is left alone -- the
check is the first thing that runs, so this is a no-op forever after.
* Otherwise: every row matching the UNION OF EVERY DEFAULT SET EVER SHIPPED
(85 canonical rows, harvested from c2ee729..1cda880 and embedded as a
table; comments stripped, whitespace collapsed, uppercased) is dropped --
those rows were never a player's choice, they were just the old board.
* Rows the player actually authored are CARRIED with their original text and
comments, and WIN over the new defaults: the losing default row is emitted
commented out as "# (yours wins) ...", keyed by the control ("KEY T",
"PAD A", "PADAXIS LX") so key rows only displace key rows.
* The joystick wizard's marker-delimited section is preserved VERBATIM (the
same markers L4JOY rewrites between), so HOTAS players migrate without
re-running the wizard.
* The previous file is saved as bindings.old.txt first. Restoring it over
bindings.txt just re-migrates next boot -- deliberate: the old WORLD is not
restorable, only the player's own rows persist. That is the point.
VERIFIED end-to-end with the real exe (three planted scenarios):
A. old-world file (8 old-default rows + 2 authored rows incl. a rate tweak +
a wizard section): both authored rows carried with comments, both
conflicting new defaults stood down with "(yours wins)", wizard section
byte-preserved, backup written, log line states all counts, and the
migrated file parses with ZERO malformed lines.
B. already-migrated file: byte-identical md5 after a full boot, no migration
line, no backup touched. (First attempt at this check clobbered its own
evidence by running case C first -- re-run properly.)
C. no file: fresh defaults now carry the marker as line 1, no migration.
The operator's real bindings.txt was stashed before testing and restored
untouched; it will migrate on their own next launch, as intended.
Docs flipped to the new reality: the README's "your keys DO NOT change"
upgrade promise is replaced by the migration story (customs + stick kept, old
stock keys replaced, bindings.old.txt escape hatch); CONTROLS.md gains the same
note; context/glass-cockpit.md records the mechanism, the 85-row table's
provenance, and why restoring the backup re-migrates.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
150961176c |
merge fix-ups: the four review findings + a build portability gate the merge exposed
Applied on top of the glass-cockpit-refit merge, all from the pre-merge review: 1. README CONTROLS TABLE REWRITTEN for the new default board. It still taught "W/S throttle, A/D turn" while the defaults moved flight to the numpad and handed the letter rows to the MFD banks -- a fresh install would read controls that do not work. The new table teaches the numpad flight block, the panel-on-your-keys groups, F7 for control mode (was M), backtick for the camera (was V), the coolant valves on 1-3/QWE with the detent warning kept, and PgUp/PgDn volume. UPGRADERS ARE TOLD EXPLICITLY that their preserved bindings.txt keeps their old keys, and that deleting it opts into the board. 2. BACKTICK VIEW-TOGGLE GUARDED. The bindings-row-wins rule (KeyHasBinding) covered V and J/K/L but not backtick itself, which polled unconditionally -- and BACKTICK is a nameable, unbound key, so binding it to a button double-dispatched (button + camera). Both halves of the view poll are now guarded. 3. ENVIRON.INI REFUSES ONE-SHOTS. The loader applied ANY key=value line, and BT_JOYCONFIG in the file would defeat the #66 one-shot in a sneaky way: the wizard DELETES the process var after running, so in every relaunched child there is no real env var left to win over the file -- capture wizard at every mission start. BT_JOYCONFIG is now skipped with the reason in the template header, which also warns that test/debug hooks (BT_MP_FORCE_DMG and friends) do not belong in a file that silently persists forever. 4. VOLUME DOCUMENTED EVERYWHERE IT WAS MISSING. CONTROLS.md gains the PgUp/PgDn row (the old "- / = (see below)" cell pointed at a note that never mentioned volume) and the note that the keys moved so they no longer share a key with anything in any layout; the bindings template header reserves PgUp/PgDn informally. 5. L4KEYLIGHT BUILD GATE (found by building the merge on this machine). The C++/WinRT Dynamic Lighting TU does not compile against SDK 10.0.19041 -- its bundled cppwinrt fails inside winrt/base itself (C2039 'wait_for'). CMake now detects an SDK older than 10.0.22000 and builds a dormant stub with the same five-function scalar interface (logs once at Start, identical behaviour otherwise) -- extending the feature's own runtime philosophy ("no Dynamic Lighting -> log once, stay dormant") to build time. Dynamic Lighting is a Windows 11 22H2 feature, so nothing real is lost on an older build machine, and machines with a newer SDK build the real mirror unchanged. VERIFIED on the merged tree (4.11.572): * build clean -- 0 compile errors; the 40 /FORCE-tolerated unresolved externals are byte-identical to every pre-merge build (20 CreateStreamedSubsystem + 20 DefaultData; the earlier claim that all 40 were CSS was shorthand -- corrected here for the record). * solo boot: environ.ini template written + loaded, mode resolver announces the surround, historical bands L276/R276/T223/B336 confirmed, keylight stub logs its dormant line, BT_SHOT capture shows the under-glass button treatment and the corrected hat labels, 0 faults. * PgUp/PgDn volume PROVEN LIVE via injected keys: 5%->10%->15%->10%, saved to volume.cfg. (Side catch: content\volume.cfg had been sitting at 0.00 from an old test -- anyone using this tree had a silent game; reset.) * full relay cycle on the merged exe: 2 pods staged, launch pair, mission, StopMission, round RESET -- all clean. * all three console suites still pass; checkctx CLEAN. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
62e89018f5 |
Merge glass-cockpit-refit: cockpit scaling + one button geometry, the keyboard button board, RGB keylight, crouch, the cwd guard
Cyd's 9-commit branch, reviewed before merge (clean merge-tree, zero overlap with the console/relay work that landed after his fork point; his L4VB16 refit preserves the #48 plane-audit tripwires, and his lamp-decode fix corrects the #47 flash rendering). Four review findings are fixed in the follow-up commit: the stale README controls table, the unguarded backtick view-toggle, the environ.ini one-shot loophole, and volume-key documentation (the -/= -> PgUp/ PgDn move itself landed pre-merge so this branch's Comm-bank -/= bindings are collision-free). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
9384c38c08 |
volume keys: -/= -> PgUp/PgDn, ending the zoom double-fire (prep for the cockpit-refit merge)
The issue #26 volume poll watched VK_OEM_MINUS/VK_OEM_PLUS -- the same physical keys the authentic 1995 typed-character target-zoom hotkeys ('+'/'-') live on, so a single press zoomed AND stepped the volume (operator-reported annoyance). The incoming glass-cockpit-refit branch also binds -/= as Comm-bank buttons, which would have made it a TRIPLE dispatch on fresh installs. PgUp/PgDn produce no WM_CHAR at all, so they cannot collide with any typed hotkey in any install; they are unbound in every bindings layout including the refit's 74-key board; and they exist on tenkeyless keyboards. Numpad +/- were considered and rejected: their typed characters feed the same zoom channel that made -/= wrong. README/CONTROLS documentation follows in the merge-fixes commit (the whole controls table is being redone there for the new default board). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
a60ada782f |
operator-console KB: record what survives of the 1995 console (BTCNSL.CPP message defs; the Macintosh clue)
The changelog line in game/original/BT/BTCNSL.CPP -- '06/03/95 GAH Added
corresponding Macintosh message definitions' -- implies the 1995 sysop station
was a Mac application in a separate codebase, which explains its total absence
from the pod-side archives. Recorded so the provenance question ('is our
console a port or a recreation?') has a durable answer: wire-faithful
recreation, from the surviving parser side.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
c9e561d9ba |
console review round 2: six must-fixes from the adversarial pass (one of its fixes corrected)
The 4-dimension review of 4736cba..820caf8 returned 14 surviving findings. All six must-fixes plus the four deferables are in; one of the review's own prescriptions was wrong and is fixed differently (below). THE REAPER, FINAL DOCTRINE (blocker + major). Rev 2 -- "reap anything silent in the staging window" -- was still wrong twice over: a pad sends exactly ONE message in its life (the egg ACK, L4NET.CPP:1259) and is then silent FOREVER, so any manual-launch hold >180s reaped every healthy ACKed pad and force-relaunched their clients; and a REGISTERED game conn is quiet on TCP while loading, so with BT_RELAY_TCP_ONLY=1 (or blocked UDP) the reaper _abort_round()ed the whole night every ~3 minutes blaming a healthy player. The rule is now: an app-level deadline is valid only while a RESPONSE IS OWED. Only egg-sent-never-ACKed pads are reaped, with the debt clock starting at EGG SEND (a pad that sat through a long held-egg wait gets its full window -- an edge neither review round caught); game conns are never reaped at all. Half-open ghosts are TCP keepalive's job (~130s, faster than the deadline anyway). RE-ARM GUARDED ON EVERY CHANNEL (major). The launch_at guard lived only on the LAUNCH-press path; the ctl `rearm` command, stdin, and the always-lit GUI button reached _rearm_for_new_round unconditionally -- one press mid-mission zeroed the counters, permanently killing the mission clock AND making End Mission print "no mission is running": an unstoppable round, the exact class this feature exists to eliminate. Now refused (loudly) while a mission is running or a pair is in flight, and the button greys while launched. THE REVIEW'S OWN FIX WAS WRONG here: it prescribed refusing on `launches_sent >= 2`, but that stays 2 after a FINISHED round -- applying it verbatim resurrected the original dead-LAUNCH wedge, caught immediately by the regression suite. "Running" is `launches_sent >= 2 AND not stop_sent`. RE-ARM RE-KEYS SEATS (major). It restored the template roster but left seat_beacons/seat_prefs keyed by the trimmed round's positional ids, so round 2's trim minted a departed player's tag into the egg and trimmed a present player's out, shifting every callsign/mech a slot. The re-key block is factored out of _maybe_reset_round (_rekey_seats_to_roster) and shared. RESTART SESSION NO LONGER BAKES A WALK-UP'S NAME INTO THE EGG (major). _stop_session and _start_session now restore the stashed configured callsign/mech into the cells BEFORE _collect_egg can snapshot them (_restore_seat_defaults); previously the departed name went into the egg (name= plus rasterized bitmaps) and was then re-captured as the seat's permanent "configured default". LATE REMOTE OPERATOR GETS A ROSTER (major). The only line the GUI can adopt tags from was printed once at relay startup and aged out of the 400-line control history in ~33 minutes of stats chatter -- AUTH now re-issues the live roster line ahead of the replay, so a remote operator connecting at any point gets pilot lights and a working LAUNCH button. DEFERABLES, all four: _drop_game blames the tag stashed AT REGISTRATION (the live-roster resolve named the wrong tag for a trimmed-round conn dying after a re-arm restore -- and the tag-trusting GUI would clear the wrong seat); an operator-BLANKED callsign cell now restores (empty string is a real configured value; the falsy skip left the departed name up and re-captured it); a returning player displaces their own half-open registered ghost (same IP, no mission running) instead of eating ROSTER FULL until keepalive fires; udp_spoofed now rides the [relay-stats] line when non-zero and the warn set is capped at 256. VERIFIED: rearm suite grown to 25 checks (ACKed pad never reaped however silent; never-ACKed pad reaped; game conns never reaped; the egg-send debt clock; re-arm refused mid-mission, allowed after round end) -- plus net 17/17, roster 22/22, checkctx CLEAN. Live rig: mid-mission `rearm` refused with the message and the mission survived; End Mission -> re-arm -> full re-seat -> second mission launched. One bounded artifact observed and documented: each waiting pod bounces once (identity resync) after an explicit re-arm. Docs rewritten to the final doctrine (context/operator-console.md reaper + re-arm + roster-replay + udp_spoofed sections; OPERATOR_GUIDE + tooltip now say re-arm is between-rounds-only and warn about the one-bounce resync). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
 CydandClaude Opus 5
|
9d536294e4 |
Mouse buttons are bindable; mouse LOOK scoped (not built)
BUTTONS (shipped). Win32 hands mouse buttons out as virtual keys, and the
binding path already stores VKs and polls GetAsyncKeyState -- so this needed no
verb, no parser change and no new machinery, just names:
key MOUSE4 button 0x40 # side buttons
key MOUSE5 button 0x46
key MOUSEMIDDLE button 0x41
MOUSELEFT/MOUSERIGHT are named too, with a warning in the file header and the
docs: they are how the player presses cockpit buttons (left = press, right =
latch), and the poll is focus-guarded but not click-aware, so binding either
ALSO fires on every panel click. The side and middle buttons are the free
ones. Axes work as well (`key MOUSE4 axis Throttle slew + 0.7`).
One guard: the RGB lamp mirror skips mouse VKs when building its map -- a
keyboard lamp array has no mouse buttons to paint (BTPadBindingIsMouseKey).
Verified live: a bindings.txt carrying two mouse rows loads 76 keys (74 + 2),
and a real injected XBUTTON1 press produced `[emitter] FIRED #1`. Clean A/B --
the earlier run whose SendInput failed the struct-size check logged no fire at
all. Shipped commented-out examples in the default profile so the capability
is discoverable.
LOOK (scoped only). docs/MOUSELOOK_PLAN.md. Mouse MOVEMENT is a real feature,
not a table entry, and the hard part is not reading the device:
- The cursor is already spoken for. The glass cockpit's premise is that all
72 buttons are mouse targets; mouse-look wants a captured, hidden, recentred
cursor. Both cannot be true, so this is a MODE question first -- four
models compared, recommending hold-to-look (costs nothing when unbound,
cannot strand a player, behaves the same in surround and exploded layouts).
- The channels are POSITIONAL (JoystickX/Y are -1..1 deflections the mapper
reads per control mode); a mouse gives deltas. Accumulate-into-a-virtual-
stick reuses every existing rule; a rate model would need new mapper
semantics.
- Control mode changes what it MEANS: BASIC steers with the stick, MID/ADV
twist the torso -- so mouse-look steers in one and aims in the other.
Proposed grammar (new row type, so old builds skip it with a warning),
where the code goes, ~250 lines, and a verification plan whose load-bearing
item is "panel clicks still work when a mouselook binding exists but is not
engaged". Also flagged: the pod had NO mouse, so nothing here is
reconstruction -- every choice is design, judged on feel, and it must stay out
of the pod build's path.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
820caf8765 |
console review follow-up: two edge cases in my own roster fix, found and fixed
Self-review of |
||
|
arcattackandClaude Opus 5
|
3a694485bf |
operator UI: a departed player no longer keeps their seat in the roster
Operator report: "if a player disconnects from the console, it leaves their name
up in the roster ... it should turn their seat back to empty." I had not noticed
or fixed this -- the earlier review raised adjacent roster issues (lights mapped
through the untrimmed tag list after a trim) but not this one.
CAUSE. The GUI only ever WROTE walk-up names into the table. On a disconnect
the relay pops seat_info, the write-back loop hit `if not info: continue`, and
the cell kept the departed callsign for the rest of the session -- while the
pilot light beside it correctly said "waiting". So the table and the light
disagreed and a free seat looked occupied.
Two changes, because the relay's two seat-clearing signals are NOT symmetric:
* SessionMonitor now pops seat_info when a pilot goes idle from EITHER signal.
"PLAYER n LEFT" is only printed inside the beacon-close branch gated on
`if seat not in self.by_host` (the relay's own comment: "never claimed:
player left"), so a player who was fully REGISTERED and then dropped
produced only `game[...] dropped` -- the common case, and the one that left
the name up. Keying "seat is empty" off LEFT alone could never have worked.
* _refresh_pod_status stashes the operator's configured callsign/mech PER
OCCUPANCY when a seat fills, and restores it when the seat empties.
Per-occupancy rather than once at egg load, so an edit the operator makes
while a seat is empty is respected instead of being overwritten by a stale
snapshot.
It restores the CONFIGURED PILOT, not a literal blank, deliberately: that cell is
editable and is what gets written to the egg on Save, so a placeholder like
"-- empty --" would end up in the mission file. The "unoccupied" signal is the
grey `waiting` light beside it.
NOT CHANGED, and explained instead: a vacated seat is held for its player for 90s
(seat_reclaim) so a crash or a reconnect returns them to the same seat and mech.
Assignment skips claimed/reserved/reclaim-held seats, so a brand-new joiner in
that window gets the next free seat, or ROSTER FULL if there wasn't one; after 90s
the seat is fair game. The operator said they were happy with players keeping
their seat/address, so this stays -- it is now documented in both docs rather than
silently removed.
VERIFIED: scratchpad/test_operator_roster.py (new, 14 checks) drives the REAL Qt
widget offscreen and feeds it the REAL relay log lines: a walk-up name appears;
after REGISTERED then `dropped` the light returns to waiting AND the seat shows
the configured pilot again; the seat is reusable by a different player and clears
for them too; the `PLAYER n LEFT` path clears it as well; an operator edit made
while the seat was empty survives an occupancy; and a neighbouring row is never
touched by another seat's traffic. Other suites still pass (rearm 19/19, net
17/17); checkctx CLEAN.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
ab5828a866 |
relay/console: fix the three remaining hardening items from the review
1. UDP ENDPOINT HIJACK. `from_host` in a UDP envelope is the sender's OWN claim,
and the relay used it directly to refresh that host's downstream endpoint --
so ANY datagram claiming host N silently stole host N's traffic (a zombie pod
from a previous round, a stale NAT mapping, or anyone who guessed a host id).
The victim simply stopped receiving on a channel that still looked healthy.
The claim is now bound to the identity we actually authenticated: the IP of
that host's live TCP game connection. The PORT is deliberately not checked --
it moves on a NAT rebind, which is the whole reason the endpoint map refreshes
per datagram -- and a genuine IP change cannot happen without the TCP
connection breaking and re-registering, so a legitimate pod is never rejected.
Rejections are counted (udp_spoofed) and logged once per offending (host, IP).
Known limit: two pods on one machine share an IP, so this cannot separate
them; same-machine trust is assumed.
2. THE EGG ACK COULD BE MISSED ENTIRELY -- a silent, unrecoverable wedge. The
pod's ACK is the launch gate's ONLY signal, and it was detected by parsing a
single recv() at fixed offset 0 behind a `len(data) >= 24` test. On a real
network (loopback hid both cases) the 28-byte ACK arriving SPLIT -- 16 bytes
then 12 -- was dropped by that test and never looked at again, and two
COALESCED messages meant only the first was read. A missed ACK means that
seat never counts toward the gate, so the round can never be released.
_console_read now buffers per connection and walks every complete frame
(16 + messageLength); an implausible length drops the connection WITH A REASON
rather than mis-reading that pod all night, since a stream protocol cannot be
resynced by guessing. Bounds: CONSOLE_MSG_MAX 64K, CONSOLE_INBUF_MAX 1 MiB.
3. REMOTE-OPERATOR MODE WAS HALF A CONSOLE. LAUNCH and END MISSION could never
enable (both conditions required `console_proc is not None`, which the remote
path never sets), the pilot lights were permanently blank (SessionMonitor was
built with an empty tag list, so `seated` was always 0), and Launch-local was
refused by the same guard even though the code below it already built
BT_RELAY from the remote host. All three enable conditions now accept EITHER
channel, and the monitor ADOPTS THE ROSTER from the relay's
"roster: N pilot(s) -> hostIDs [...]: [...]" line -- which the relay replays to
every newly AUTHed operator -- so a remote operator gets real lights and a
real seat count.
VERIFIED
* scratchpad/test_relay_net.py (new, 17 checks): spoofed datagram cannot move
an endpoint and is counted; a NAT rebind on the same IP still is honoured; an
unregistered host id still dropped; the ACK is found whole, split in two,
split three ways mid-header, and when hiding behind another message;
a garbage length drops the conn; the roster line is adopted, maps a
subsequent SEATED onto a real tag, lifts `seated` off 0, and re-feeding it
preserves known state.
* Live 2-pod rig: real pods ACKed through the new reassembly path (2/2 ready,
zero desync drops), the mission launched and ran, UDP flowed throughout
(93 rx / 85 tx, udp-known [2,3]) with ZERO false spoof rejections, then a
clean StopMission + round RESET.
* scratchpad/test_relay_rearm.py still 19/19; checkctx CLEAN.
Docs updated to match (context/operator-console.md gains reassembly and
anti-hijack sections; the guide no longer claims remote mode is crippled).
Remaining open items are now recorded in the topic's frontmatter: mesh mode's
operator buttons are inert by construction (no stdin reader in that path -- left
alone, mesh self-launches), _log_launch_readiness can cry STALL on a healthy
launch-with-whoever, and a straggler's late FIN is still misread as a pod dying
mid-load.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
CydandClaude Opus 5
|
5f88a4eeb2 |
The keyboard becomes the button board (RP412 bindings port) + a controls page
Ported RP412's bindings design and made it the DEFAULT: the letter and number rows are the MFD button banks laid out WHERE THEY SIT ON THE PANEL, flight moves to the numpad so the board stays free, F1-F12 are the map's two columns, and G/B stay unbound as the physical gap between the lower clusters. Expressed in BT's OWN grammar (slew/deflect/set, the Turn channel) -- bindings.txt is a documented compatibility surface and was not touched. Coverage: 61 of 72 addresses on the keyboard, and the 11 absent ones are exactly those the pod never wired (0x16/0x17/0x1E/0x1F column gaps, 0x38-0x3E intercom/door). All 72 stay clickable on the panel. It lands on BT's addresses unreasonably well: 1-4 + QWER are the ENTIRE coolant system (Condensers 1-6, flush, balance), F6/F7 the display and control-mode cycles, F9-F12 Generators A-D, F4 the crouch button reconstructed earlier today. THE DELIBERATE COST. A bound key is removed from the authentic 1995 typed-hotkey channel so it cannot double-dispatch, and this board binds nearly everything -- so 5 (Quad page), z (Eng1), t/y/u/i/o (pilot select) and +/- (target zoom) are given up. Unbind a key to get its 1995 meaning back. Documented in the file header, the markdown and the page. NEW RULE: A BINDINGS ROW WINS OVER A BUILT-IN CONVENIENCE KEY (PadRIO::KeyHasBinding). V and J/K/L are board buttons now, and those polls read GetAsyncKeyState directly -- without this they would have fired BOTH the button and the built-in. View toggle lives on BACKTICK alone. CONTROLS.MAP rewritten to mirror the board so glass/pod/dev still feel identical (the 2026-07-21 settlement): 90 bindings, 0 parse complaints. HAT LABELS CORRECTED [T1]. INPUT_PATH_AUDIT flagged 0x41-0x44 and was right. Settled from the streamed mapping (BT_CTRLMAP_LOG): elem 66 -> subsys 17 attrID 14, i.e. 0x42 is the TORSO subsystem, not a look; 0x44/0x43/0x41 are the mapper's LookLeft/LookRight/LookBehind (attrID 10/11/12). The .RES has no "TORSO CENTER" string -- its names are LookBehind/Down/Forward/Left/Right -- so the audit's wording was loose but its substance correct. Swept L4GLASSWIN, L4PADPANEL and L4VB16. docs/CONTROLS.html: interactive keyboard (hover a key for its address and meaning, MFD clusters outlined), pad diagram, panel map, the under-glass press-target figure, radar placement thumbnails and the environ.ini table -- modelled on RP412's page, rewritten for BT's addresses and hazards. mkdist wraps the fragment in a doctype/charset shell and ships it beside CONTROLS.txt. Verified: bindings.txt regenerates and loads (74 keys, 0 errors); CONTROLS.MAP 90 bindings, 0 errors; 72/72 panel addresses still dispatch; surround / exploded / dock / pod / dev boot and simulate with zero faults. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
CydandClaude Opus 5
|
56d9971b0e |
-fit: pin the backbuffer to the CANVAS, not the client (the display was wrong)
Field report with a screenshot: `btl4.exe -fit` renders tiny MFDs hugging the corners, an over-wide world view, and a horizontally squashed scene. CAUSE. L4VIDEO pins the windowed backbuffer to the CLIENT RECT (Gitea #56, so a device Reset cannot silently re-derive it). That is correct precisely while client == canvas -- true on a normal boot, and FALSE under -fit, which sizes the window to the whole monitor BEFORE the device exists. The backbuffer came out 3440x1440 against a 1452x1059 canvas, and everything downstream disagreed: - BTApplyWorldViewport and BTDrawCockpitPanels lay out from the BACKBUFFER, so the canvas they drew was 3440 wide: bands unchanged, MFDs still 320x240 (hence tiny), view rect 2888x881 instead of 900x500; - the world projection used the aspect of the INTENDED view (1.8) inside a 3.28 viewport, stretching the scene; - Present then scaled that oversized backbuffer into the 1974x1440 letterbox, squashing it horizontally by 0.574; - and every button was hit-tested through gBTCockpitCanvasW/H while being drawn in backbuffer space -- the exact #50 divergence the pinning comment warns about, unnoticed because nothing had clicked under -fit. FIX: when the cockpit surround is up, the backbuffer IS the canvas, whatever the window is doing. Pin to gBTCockpitCanvasW/H and let the letterbox do the scaling -- which is what the rest of the design already assumed. Drag-resize was always right for this reason (the device is created while client == canvas and a later resize does not recreate it); only -fit, which changes the client first, exposed it. MY EARLIER VERIFICATION WAS WRONG. I called -fit good from a screenshot that was centred and letterboxed, without checking that the backbuffer matched the canvas or that a single button could be clicked. Both are checked now. Verified: -fit on 3440x1440 renders correct proportions (captured), and 72/72 buttons dispatch under -fit; surround/exploded/dock/pod/dev all boot and simulate with zero faults. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
CydandClaude Opus 5
|
a7bcf1cbab |
Proposal: ship the authentic single-folder dist layout (docs only, NOT implemented)
The bare-launch crash was a symptom; the cause is that BT411 hands players a developer tree. mkdist zips tracked repo paths verbatim, so the dist inherited build\Release\btl4.exe + content\ -- while the 1995 pod (BTL4OPT.EXE sits beside BTL4.RES to this day, in content\) and RP411/RP412 all ship ONE folder, where cwd is right by construction and a double-click just works. docs/DIST_LAYOUT_PLAN.md sets out the proposed tree, the file-by-file change (~50 lines total, no game code), what it buys, and the part that actually needs a decision: the upgrade story. Four migration options with a recommendation, plus the wrinkle that decides between them -- the zip root is VERSIONED, so every version may already land in its own folder, in which case the documented extract-over-top has never worked and the migration cost is near zero. Four open questions listed for the other author, including whether the .bat launchers still earn their keep and whether the pod cabinet's own install shape should be confirmed against Nick's notes before we call one folder authentic for the cabinet too. Not implemented pending that discussion. The 4.11.560 cwd guard stands either way and makes both layouts work. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
CydandClaude Opus 5
|
78a98dafc7 |
KB: record the cwd landmine + why the folder layout differs from RP/4.10
The 1995 pod and RP411/RP412 ship ONE folder (BTL4OPT.EXE sits beside BTL4.RES to this day, in content\); BT411's build\Release + content\ split is an artifact of mkdist zipping tracked repo paths, not a decision -- and it is the reason this class of bug exists here and nowhere else. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
CydandClaude Opus 5
|
8408165d92 |
Find content\ ourselves: a bare exe launch no longer scatters config and dies
Field report: `.\btl4.exe -fit` from build\Release opens the mission console,
then the mission crashes on Launch.
ROOT CAUSE (pre-existing landmine, newly reachable). The engine resolves
BTL4.RES, VIDEO\, BTDPL.INI, the eggs AND both player config files
(bindings.txt, environ.ini) RELATIVE TO CWD. Every launcher .bat does
`cd content` first, so this never showed. A bare launch of the exe therefore
found no resources ("Resource file btl4.res v1.0.0.0 is obsolete!"), wrote a
stray bindings.txt / environ.ini / btl4.log NEXT TO THE EXE -- so the player's
real ones in content\ looked like they had never been created -- and killed the
mission generation the menu launched (the child inherits the parent's cwd).
Recent work made that path inviting rather than obscure: glass is the desktop
default, a zero-arg launch opens the menu, and -fit is documented as something
you pass on the command line.
FIX: BTEnsureContentDirectory() runs before anything resolves a relative path
(the log file included, so a bare launch logs where the launchers log). If cwd
has no BTL4.RES, probe from the exe's own directory -- ..\..\content for the
shipped layout, plus the flattened and in-content cases -- and
SetCurrentDirectory there. Says so in the boot line when it had to go looking.
Belt and braces: the environ.ini writer refuses to write unless BTL4.RES is
beside it, so a failed probe cannot scatter a settings file either.
ALSO: -fit was silently dropped when the menu relaunched into a mission. The
front end rebuilds the child's command line from scratch, and the flag was
parsed at window creation -- code the menu process exits long before reaching.
Parsed early into gBTFitDisplay now, and appended to the relaunch.
Verified: bare `btl4.exe -fit` from build\Release leaves ZERO files next to the
exe, writes environ.ini + btl4.log into content\, and the mission child's exact
command line (-net 1501 -platform glass) from that same wrong cwd now logs
"cwd: found the content directory from the exe path", loads resources and
enters RunMissions instead of dying.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
CydandClaude Opus 5
|
02b1b50e45 |
RGB keyboard lamp mirror + the shipped controls reference
KEYBOARD LAMP MIRROR (L4KEYLIGHT). Ported from RP412, itself a port of vRIO's KeyboardLampMirror. Keys bound to a lamp address in bindings.txt glow with the panel palette through Windows Dynamic Lighting -- yellow for the map's side columns (0x10-0x1F), red for the rest -- flashing in step with the on-screen buttons (its LampLevel copy matches the FIXED BTLampBrightnessOf). Per-key boards light each bound key; zone-lit boards mirror the strongest lamp board-wide. All WinRT runs on a private worker thread: watcher, claim, and a 100 ms paint loop that repaints only on change. RP412's packing hazard does NOT transfer: it forces default struct packing there because its engine is /Zp1, which would break the WinRT ABI. BT411's BT_OPTS carries no /Zp, so only the dialect flags are needed -- /std:c++17 /permissive- per-file, since the project otherwise builds C++14 /permissive. The scalars-only interface is kept anyway so the isolation survives if packing is ever added. Wired in PadRIO: map built from bindings.keyBindings (ActionButton binds only, first-binding-wins per key), fed from PadRIO::SetLamp, stopped in the dtor (which hands the LEDs back to Windows). BT_KEYLIGHT=0 opts out; a machine without Dynamic Lighting logs once and stays dormant. Verified live: claimed this machine's 24-zone keyboard and mirrors 25 bound keys; BT_KEYLIGHT=0 and the pod profile produce zero keylight lines and run clean. SHIPPED CONTROLS REFERENCE. docs/CONTROLS.md carries the keyboard table plus the full 72-BUTTON POD MAP, grouped by the display each bank surrounds, with the coolant-valve detent warning (1-5-50-CLOSED, one press past max shuts the loop) and the jam/eject procedure. mkdist flattens it to ASCII as CONTROLS.txt at the zip root, the README's own idiom. players/README.txt gained pointers to it, to environ.ini, to -fit, and to the RGB mirror. KB: context/glass-cockpit.md and docs/GLASS_COCKPIT.md updated for the whole branch -- layout modes, L4RIOBANK and the under-glass rule, the letterbox fit and its ordering trap, player-tunable displays, the measured legend grid, the closed dead-button backlog, and this mirror. Verified: five-mode regression (surround/exploded/dock/pod/dev) boots and simulates with zero faults; mkdist writes a 5149-byte pure-ASCII CONTROLS.txt. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
CydandClaude Opus 5
|
1debbeb240 |
Mech::DuckRequest -- the CROUCH button, and a dead-button census that lied
The 2026-07-20 audit listed 8 panel buttons dispatching a streamed message with
no reconstructed handler. Only ONE was still missing: generator on/off,
ToggleSeekVoltage, EjectAmmo, ToggleCooling and BalanceCoolant had all landed
between 07-20 and 07-25 while pod-hardware.md and open-questions.md still called
them dead. docs/INPUT_PATH_AUDIT.md had already flagged the census as "stale in
both directions" and was right -- swept both files, and recorded the rule: check
the code, not the census.
DuckRequest @0049fa00 (id 0x1a, RIO 0x13 -- the manual gives crouch a whole
section). The binary's entire body:
if (0 < *(int *)(param_2 + 0xc)) { *(undefined4 *)(param_1 + 0x398) = 1; }
Press-only; sets duckState (mech+0x398, attribute 0x37). A one-shot REQUEST
flag, not a posture toggle -- the handler never clears it and the only other
writer in the whole binary is the mech reset (part_012.c:9439, the same reset
that zeroes incomingLock, which is how that region was already mapped).
WHY THE FLAG HAS NO READER, AND WHY THAT IS CORRECT. duckState has ZERO readers
anywhere in the decomp, because it is published as an ATTRIBUTE and consumed
through DATABINDING: content/GAUGE/L4GAUGE.CFG runs a 3-frame bduck.pcc
oneOfSeveralPixInt bound to DuckState -- "crouch mode: button 4" on the map's
legend column. Verified live by capturing the Secondary/Radar window before and
after a 0x13 press: the crouch icon goes grey -> orange. So the handler is
COMPLETE. A crouch pose invented here would be a stand-in for data we have not
found (no SQUAT clip name survives in the decomp or in content/, only
DuckServo01.wav in AUDIO1.RES).
Bonus: those gauge widgets sit at offsets 537/430/322/215/108 -- a 107 pitch,
independently corroborating the map legend grid measured from pixels.
Still open from that census: MechRIOMapper's own Keypress @004d2514 (id 0x19).
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
CydandClaude Opus 5
|
61563c9efe |
Glass cockpit refit: one mode resolver, one button geometry, and a cockpit that scales
Replicates the RP412 cockpit line into BT411's own architecture -- porting the geometry and keeping our renderers, so BT_SHOT single-frame verification stays intact. MODE RESOLVER. Where the secondary displays go was decided in TWO places with duplicated precedence, and the boot banner read NEITHER -- it announced "per-display cockpit windows" for every glass boot, surround included. The split had also broken BT_COCKPIT=0: documented as the dock-bottom opt-out, the profile block converted it to BT_GLASS_PANELS=1, so the docked strip was UNREACHABLE under the glass profile. One resolver now, consumed by the banner, the pad-panel decision and the window sizing; BT_GLASS_PANELS is explicit-only; dock/window modes auto-raise BT_PAD_PANEL so the button field always has a home. L4RIOBANK -- one button geometry. Both renderers carried their own copy and had drifted: an MFD button was 156x138 reaching under the glass in the exploded window and a 76x24 sliver entirely OUTSIDE the glass in the surround. One module owns it now, both are consumers, placement stays per-renderer. The pod's under-glass rule (RP412 L4MFDVIEW): reach half the glass in behind the display, leave a lamp strip clearing the edge, paint buttons first and imagery over -- so the lamp reads as a bar and practically the whole display is the press target. The strip scales off the display's SHORT axis (the map is portrait) with a readable floor. Retired L4GLASSWIN's three local placers and seven layout constants. LAMP FLASH DECODE was wrong [T1]. BTLampBrightnessOf returned max(state1, state2) and blanked on the alternate phase; RIO::LampState (L4RIO.h [T0]) says solid shows state 1 and flashing ALTERNATES the two. Agrees only when one state is Off -- true for the Panic lamp, which is why it survived -- but L4LAMP.cpp:252 commands flashFast + state1Dim + state2Bright, a dim->bright pulse that rendered as a hard bright->off blink. Three copies existed (l4vb16.h, L4GLASSWIN, L4PADPANEL), all three wrong; the two locals now forward to the one fixed inline. THE COCKPIT SCALES. The fixed canvas was stretched into whatever the client area was, so a window dragged to a different shape squashed the instruments (the projection was aspect-corrected in task #20; the panels never were). Now one uniform scale, centred, leftover black. D3D9 applies it as a Present destination rect, which DISCARD forbids -- so the WINDOWED swap effect becomes COPY when the surround is up and multisampling is off. The click mapping had to follow (mapping against the full client drifts the hit test off every button by the bar width), as did the world aspect (under a uniform scale it is the view rect's own). -fit / -windowed-fullscreen: borderless over the monitor. - ordering trap: the first WM_SIZE beats the device, so a -fit boot logged aspect=3.14 and applied it on frame 1. The letterbox INTENT is decided in btl4main; L4VIDEO only confirms or withdraws it. PLAYER-TUNABLE DISPLAYS. BT_MFD_SCALE (+ _UL/_UC/_UR/_LL/_LR), BT_RADAR_SCALE, BT_RADAR_POS (CENTER/LEFT/RIGHT/MIDLEFT/MIDRIGHT). The surround BANDS derive from the resolved sizes -- that is why the sizes could not stay constants: the band a display hangs in has to grow with it or the canvas clips it. 100% reproduces the historical L276 R276 T223 B336 exactly. A corner map goes flush to the CANVAS edge and the lower MFD slides beside it (measuring off the view edge overlapped them by 232px). MAP LEGEND GRID -- measured, not inherited. scratchpad/measurelegend.py over a native capture: top 3, cell 102, pitch 107 of 640. RP412's map is 13 + 6x102 @ 105 -- same cell height, different top and pitch, so its numbers do NOT transfer. Our old even division had the pitch right by luck and sat 3px high of the labels. environ.ini. It was read ~300 lines into WinMain, AFTER the platform-profile block had run its getenv()s -- so every setting the profile reads was silently ignored FROM THE FILE and only worked as a real env var. It also putenv()'d comments verbatim. Now loaded immediately after the first-breath line, comments skipped, the real environment WINS over the file, and a fully documented default is written on first run (the bindings.txt convention: untracked, so extract-over-top never clobbers a player's settings). VERIFICATION HARNESS (new, reusable): BT_RIOBANK_LOG=1 dumps every bank; checkbank.py proves no address is SHADOWED (an address whose rect is covered by earlier buttons is dead however big it looks -- the overlapping under-glass banks make that a live hazard); clickbank.py posts a real click at every button centre. Verified: 72/72 placed, 0 shadowed, 72/72 dispatched in BOTH modes, after a resize, at 150%/135%, and at 75%+BOTTOMRIGHT; wide/tall drags and -fit undistorted on a 3440x1440; exploded/dock/pod/dev un-regressed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
arcattackandClaude Opus 5
|
1cda880c6d |
console/relay: document it properly + fix two regressions the review caught
DOCS (the ask: after a compaction this session lost track of how the console
works and launched the wrong program, twice).
* NEW context/operator-console.md -- the dedicated topic that was missing.
Leads with the thing I got wrong: btoperator.py is the PySide6 GUI the
operator uses; btconsole.py is the headless relay it spawns. Then ports,
the route table, the roster/seat/identity model, the full round lifecycle
with every launch gate, liveness, mode-specific traps, and log locations.
* NEW docs/OPERATOR_GUIDE.md -- sysop-facing: start the console, set up a
mission, watch pods arrive, launch, run back-to-back rounds, what to press
when LAUNCH looks dead, a troubleshooting table keyed on the exact log
lines, and what to save BEFORE restarting a session (Start Session
truncates operator_relay.log, so restarting to clear a problem destroys the
evidence of it).
* CLAUDE.md: two Quick Lookup rows + a DO-NOT entry naming the two programs,
so the distinction survives the next compaction.
* context/multiplayer.md: a pointer out of the scattered console notes to the
new topic (they were buried across ~8 places in a large file, which is
exactly why they evaporated).
FIXES -- both are regressions in my own previous commit, found by the review
pass, and one would have made a games night WORSE:
* THE REAPER WOULD HAVE KILLED HEALTHY PLAYERS. It was gated only on "no
mission running", which a round RESET satisfies -- so it was armed for the
whole BETWEEN-ROUNDS wait, and that is a period when a pod is legitimately
byte-silent: its seat beacon is write-only for the process lifetime
(L4NET.CPP: "the relay ignores its silence") and its console pad has no egg
yet so it cannot ACK. A real night showed 12-minute and 5-minute gaps; the
180s deadline would have dropped healthy pods and forced their clients to
relaunch. It now runs ONLY in the active staging window, skips pads with no
egg and conns that have not HELLO'd, and last_seen is also stamped from
inbound UDP (a pod streaming updates while its TCP idles was being counted
as silent). Half-open detection is keepalive's job; this is just a backstop.
* UnboundLocalError in the operator UI. My end_sent reset was an `elif` in
the chain that assigns `head`, so that branch left `head` unbound -- a crash
on the first status refresh after End Mission, which is exactly the path the
relay's "StopMission sent" line produces. Moved out of the chain.
Plus one pre-existing wedge with the same symptom as the reported bug, live-
proven in operator_relay.log (~5 minutes of a night lost): _abort_round clears
eggs_released BEFORE the survivors' sockets close, so _maybe_reset_round's own
`if not self.eggs_released: return` skips the template restore forever -- the
roster stays trimmed, the release gates can never be met, and walk-ups get
ROSTER FULL. _rearm_for_new_round's restore is therefore now UNCONDITIONAL (it
was gated on eggs_released, which made Re-arm useless in the one state that most
needs it) and it clears round_hold_until so an abort's settle window is not
inherited. Aborts are ordinary: nothing on the wire distinguishes a straggler's
late FIN from a pod dying mid-load.
scratchpad/test_relay_rearm.py now 19 checks, all passing, including the two new
regression guards (a 15-minute-silent waiting pod is NOT reaped; a pad that was
never sent an egg is NOT reaped) and the mid-pair no-re-arm guard.
Still open, recorded in the new topic's frontmatter: the UDP endpoint map trusts
the sender's self-declared fromHost; the egg-ACK is a fixed-offset parse of one
recv with no reassembly; remote-operator mode can never enable LAUNCH.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
4736cba1ca |
relay: LAUNCH is never silently inert again -- fix the post-round wedge (operator report)
THE SYMPTOM: "after a game ends I push LAUNCH and nothing happens until I reset
the session entirely or reboot the console." The operator also observed the
mirror image, which is the tell: with a STABLE group, relaunching worked fine
several times in a row.
ROOT CAUSE. Two gates, both all-or-nothing, and a UI latch that agreed with
them:
* btconsole.py: the manual LAUNCH branch is guarded on `launches_sent == 0`,
but a finished mission leaves it at 2. The only paths back to 0 were
_check_launch_gate (needs EVERY seat of the last round to re-ACK) and
_maybe_reset_round (needs EVERY pod gone). A games night lives between
those -- most pods rejoin, one player closes their window. Worse, the
"not all seats filled" diagnostic sits INSIDE the `launches_sent == 0`
guard, so in exactly that state NOTHING was printed.
* btoperator.py: the LAUNCH button is gated on `not monitor.launched`, and
`launched` was cleared ONLY by the two relay lines those same gates emit
("WAITING FOR OPERATOR", "round RESET"). So the button was greyed out in
precisely the wedged state -- the click was a no-op by construction.
That is why a stable group worked (everyone re-ACKs -> gate fires) and why only
a session restart recovered (fresh relay process = fresh state).
FIXES
* An explicit operator LAUNCH is now sufficient authority to start a new
round: _rearm_for_new_round() clears the finished round's state and restores
the template roster/egg, KEEPING seats/beacons/connections, so whoever is
here stays here. Triggered on a press while a round is latched.
* New `rearm`/`newround` operator command + a Re-arm button, so recovery never
needs a session restart. Proven live over the control port.
* Every operator command is logged AT RECEIPT with the state that decides its
fate, so "did it arrive or was it ignored?" is answerable from the log.
* A stale End Mission can no longer kill the next mission: _tick_stop consumed
a stop_requested set between rounds by latching it until launches_sent hit 2,
then StopMissioning the new mission in the tick it launched -- also
indistinguishable from "launch did nothing". It is now consumed + announced
while idle. The UI's end_sent likewise reset per round, not per session
(End Mission used to work exactly once a night).
* The UI clears `launched` on "StopMission sent" and on the re-arm line, so the
button returns when the round actually ends, and only greys once a command
has really been written (a dead relay now says so instead of logging
">> sent" into the void -- _console_finished leaves console_proc non-None).
HARDENING (the "reboot the console" half)
* SO_KEEPALIVE on every accepted socket + a last_seen deadline and a reaper:
nothing detected a pod that vanished WITHOUT a FIN (sleeping laptop, dropped
Wi-Fi, killed process, NAT timeout). Such a conn kept acked=True forever,
which permanently blocked the round reset and held a phantom seat. The
reaper stands down while a mission is running.
* Every pod-facing send went blocking with NO timeout on the single-threaded
selector loop, so one wedged peer could freeze the entire relay. All sends
now go through _send_all_guarded (10s timeout, drops the peer on failure).
* _send_egg no longer calls getpeername() on a possibly-dead socket.
REGRESSION I INTRODUCED AND CAUGHT ON THE RIG: the first cut re-armed whenever
launches_sent >= 1, which also matched the NORMAL state between RunMission #1 and
#2 (launch_requested deliberately stays set across the pair). That reset the
pair mid-flight and re-released eggs every tick -- a re-arm storm that kicked
every pod into an identity-resync loop. The re-arm now additionally requires
`launch_at is None`, i.e. no launch sequence in flight. Verified: 0 REJOIN lines
and exactly 1 RE-ARM line across a full 3-mission rig session.
VERIFIED
* scratchpad/test_relay_rearm.py -- 6 groups, all passing: the exact wedge
state recovers; a stale stop is consumed while idle; staging is NOT restarted
under an impatient operator; mid-pair never re-arms; the happy path is
untouched and RunMission still reaches the pods; the reaper drops a silent
conn and keeps a live one, and stands down mid-mission.
* Live 2-pod rig (scratchpad/rig_relay.ps1 + relay_ctl.py over the control
port): two clean back-to-back missions, StopMission, round RESET, a live
`rearm`, and LAUNCH-with-nobody-present now printing why instead of nothing.
NOT reproduced end-to-end: the field wedge needs 3+ pods with staggered rejoins
(this rig's pods re-exec together, so the all-gone reset always fires). The
state itself is covered by the unit test. Awaiting live confirmation on a real
games night.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
48d47ef806 |
#45 review pass: correct three claims the fix's own comments got wrong
An adversarial review of the change returned NO BLOCKERS but caught three
statements that were wrong or unverifiable. All three are corrections to my own
comments/docs, not behaviour changes -- the verified binary is unchanged.
1. The phantom-kill note had the ORDERING backwards. It claimed the owner's
record "overwrites it on the next record -- the phantom stops being visible".
The engine's event priorities run the other way: an inbound update record is
posted at UpdateEventPriority == MaxEventPriority and drained by
ExecuteBackgroundTask, while the rerouted score message sits at
EntityManagerEventPriority == DefaultEventPriority and is popped later -- so
the correction usually lands FIRST and the phantom AFTER it. The artifact is
BOUNDED, not masked: on the killer's pod only, that victim's KILLS reads +1
until the next record, i.e. <= one 2s heartbeat instead of "until the victim
next scores or dies". Swept in btplayer.cpp, KD_SCOREBOARD_PLAN.md and
RECONCILE.md.
2. The corpus figures were not reproducible, and were mis-tiered. I cited
"125 of 125 SCORE type=2 over 255 node-logs" and "0 of 8800 DMG rows" -- true
when measured, but the corpus is append-only and this fix's OWN verification
runs grew it from ~255 to 425 files, so nobody can re-derive them. Replaced
everywhere with invariant RATIOS that hold at any corpus size, re-measured
here:
* 0 of 18818 DMG rows are inst=R (damage is applied master-side only)
* 439 of 439 DEATH inst=R rows read killer=0:0 killdmg=0.000, against
0 of 225 inst=M rows (sole lastInflictingID writer: mech.cpp:746)
Also re-tagged T1 -> T2: log-corpus field evidence is T2 per the CLAUDE.md
tier table. Lesson recorded in the banner: cite ratios, not counts, for a
corpus your own rigs append to.
3. The recordLength guard's justification was wrong (and the plan's risk 5 with
it). A short legacy record does NOT desync the stream -- both sides advance
by the WRITER's stamped recordLength and no reader asserts a length. The real
hazard is a read PAST THE ALLOCATION: an inbound update message lives in a
per-event heap block sized from messageLength, so on a TRAILING legacy record
the two new fields would be read off the end of it.
Plus: the record struct's own field comment still said "deathCount -- the DEATHS
column", the exact wrong-field belief #45 removed, sitting on the wire struct.
Filed the review's other finding on #60: a SECOND decomp export gap, index
jumping FUN_0049fe80 (ends 0x49ffc8) -> FUN_004a1232, 4714 unexported bytes
containing Mech::TakeDamageMessageHandler @0x4a0230 -- the producer of every
score and death message in the game. Verified independently against
functions_index.tsv before posting.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
a52207d779 |
#45 scoreboard: reclaim the binary's DEATHS field, add a heartbeat, sweep the false KB claims
Follow-up to
|
||
|
arcattackandClaude Opus 5
|
4fa7eee54f |
K/D scoreboard ROOT-CAUSED for real (#45): the kill credit was rerouted, not lost
Kills/deaths only ever appeared on ONE machine. The cause is not a missing
tally -- it is Entity::Dispatch:
if (GetInstance() == ReplicantInstance) // ENTITY.cpp:244-251
application->SendMessage(ownerID, EntityManagerClientID, message);
BTPostKillScore runs on the VICTIM's node (the only node whose mech carries a
populated lastInflictingID), resolves the killer's Player -- a REPLICANT there --
and Dispatch()es to it. The engine reroutes that to the owning host, so
++killCount lands on the killer's own PC and nowhere else, and nothing carried it
back out: Player__UpdateRecord is currentScore + dropZoneLocation only. Every
other pod's copy therefore read 0 all mission. playerLink was never NULL for
these kills -- the dispatch proves it resolved.
Field proof over the whole corpus (255 node-logs): 125 of 125 SCORE type=2 rows
credit the LOGGING node's own player, not one credits a remote pilot; and 0 of
8800 DMG rows target a replicant, so no other node can even know the killer.
This means the owner's counter is already the single authoritative copy,
incremented exactly once per kill. So the "design decision" the plan was blocked
on collapses: there is no second writer to reconcile, only a one-way
owner->replicant mirror to add.
* BTPlayer__UpdateRecord = Player::UpdateRecord + killTally + deathTally, with
Read/WriteUpdateRecord overrides. No new data member, no new virtual ->
sizeof(BTPlayer)==0x28c and every existing offset lock still holds.
* Both counter writes already ForceUpdate() (:508 death, :829/:840 score), so
no dirty-bit edit was needed (plan risk 6 avoided).
* recordLength guard in ReadUpdateRecord: a pod on a build without the
extension degrades to "remote counters don't move" instead of reading a
neighbouring record as a kill count.
* Size locks added per plan risk 2 (no false offsetof assert).
Rig-verified, 2-node loopback (scratchpad/rig45_up.ps1, BT_MP_FORCE_DMG):
owner 3:1 finished kills 3/deaths 2 and the peer read kills=3 deaths=2; owner 2:1
finished kills 2/deaths 3 and the peer read kills=2 deaths=3. Exact convergence
where a remote column previously never left 0. 3 respawns per side with intact
death sequences (deathCount is only overwritten on replicant copies; the
handshake runs on masters), no crash, no GLITCH rows.
Kept byte-faithful: the kill handler's partner increment (the binary's
inc [ebx+0x27c] / inc [edx+0x27c] wrong-column slip) is still reproduced. It
always lands on a replicant copy, so the owner's record now overwrites it -- the
rig caught the correction repeatedly (wasKills=3 -> kills=2). The phantom kill
stops being visible without silently deciding the fidelity question.
Still open and deliberately untouched: which field the LOCAL pod's DEATHS column
should read (+0x280 vs deathCount), last-hitter-takes-all/ram kills, and the
slip's fidelity. Awaiting live multi-pod verification by a human.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
da617e6f8f |
README + release notes: match reports UPLOAD THEMSELVES -- stop asking testers to send them
The operator caught this: I had carried the old "please send the operator your
matchlog" instruction into the release notes, and the shipped README has said
it for a while. It is out of date.
Since 2026-07-22 the client auto-uploads its match forensic log to the relay
after every round: btl4main.cpp:914 calls BTRelayUploadMatchLog() once
RunMissions() returns, and the console receives it on route -9 and saves it as
matchlogs/<peer>_<name> (btconsole.py:1026-1042). content/matchlogs/ is
already full of them.
So the instruction was making every tester hunt down and attach a file the
operator already had. Corrected to state the real contract, including the
gaps the auto-upload genuinely cannot cover:
* RELAY modes only -- join.bat and join_lan.bat both set BT_RELAY, so both
upload. play_solo.bat and play_steam.bat set none, so those still need a
manual send.
* The upload fires when a round ends CLEANLY (after RunMissions returns), so
a mid-round CRASH loses it -- exactly the case we most want the file for.
* join.log is NEVER uploaded -- no mechanism exists -- so that one is still
a manual ask, and is now called out separately rather than buried.
Also added the same correction to dist/RELEASE_NOTES_4.11.550.md.
The zip cut before this (BT411_4.11.550.zip) carries the stale README and must
be re-cut.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
4360908d38 |
README (the shipped tester doc): repair two corrupted paths + bring it current with today's wave
Asked to double-check the launchers and make sure the user-facing docs in the
zip are up to date. players/README.txt is the ONLY user-facing doc mkdist
ships (content/VIDEO/RESULT2.TXT is a game asset), and it was last touched
2026-07-24 -- before all of today's work.
1. CORRUPTED PATHS (real defect, worst possible placement). The file
contained literal CONTROL BYTES where two paths should be:
offset 440 0x08 ...personal files (content<BS>indings.txt with...
offset 490 0x0b ...joystick setup, content<VT>olume.cfg with...
i.e. `content\bindings.txt` and `content\volume.cfg` had been written
through a string that interpreted \b as backspace and \v as vertical tab.
Testers read "contentindings.txt" / "contentolume.cfg" -- files that do not
exist -- in the UPGRADE paragraph, the one place that tells them which
personal files to preserve when extracting over an old install. Repaired
by byte, and verified: 0 control bytes remain in all four packaging
variants. (The other two occurrences, content\join.log and a later
content\bindings.txt, were undamaged -- \j and the second \b happened to
survive.)
2. CONTENT now current with today's 13-issue wave, tester-relevant parts only:
* AMMO BAY FIRE is now LETHAL (#46) -- flagged prominently as a gameplay
change, with the ~10 s fuse, what it costs, and the EJECT-hold escape.
Previously these fires were harmless; a tester who ignores one now dies.
* ENG-page buttons FLASH on a jam or bay fire (#47) -- the restored pod
annunciator, so testers know to look for it.
* KNOWN ISSUE + WORKAROUND for #62: the BUS MODE button's second press
detaches a weapon from its generator and the auto re-attach is dead, so a
weapon can go permanently dark. Documented with the recovery (press a
GENERATOR SELECT A/B/C/D button) to pre-empt a wave of "my weapon broke"
reports.
Verified by simulating mkdist's own marker processing across all four
steam-on/off x expire-on/off variants: new content survives every path, no
leftover {VERSION}/{STEAM}/{EXPIRE} markers reach the tester, no control bytes.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
373889760c |
#66: joyconfig.bat no longer hangs the game -- clear BT_JOYCONFIG once the wizard is done
Found while sweeping the shipped launchers for the PATH-shadowing bug class.
Previously recorded ONLY as row 10 of docs/INPUT_PATH_AUDIT.md and never
filed, so it was invisible on the tracker.
joyconfig.bat sets BOTH `BT_JOYCONFIG=1` and `BT_FE_SOLO=1`, so after the
wizard writes bindings.txt the boot continues into the FRONT END, which hands
off by CreateProcessW()ing the next generation with the CURRENT environment.
That clear-list (btl4console.cpp) covers only BT_FE_EGG/PODS/SECS/LOOP --
BT_JOYCONFIG is never cleared -- so the handed-off generation re-entered
BTJoyConfigWizard() and blocked on _getch() BEHIND the 3D window: an
unrecoverable hang, hitting exactly the players most likely to run that bat
(flight stick / HOTAS / pedals).
Fix: SetEnvironmentVariableA("BT_JOYCONFIG", NULL) immediately after the
wizard returns. Cleared in the WIN32 environment (what CreateProcess
inherits, matching the console's own pattern) rather than the CRT copy, so
exactly one generation ever runs it. Chosen over extending the console's
clear-list because it also covers any future relaunch path.
NOT YET REBUILT: the exe was locked by a live play session when this landed.
Must be rebuilt + verified (run joyconfig.bat, finish the wizard, confirm no
second "=== BattleTech joystick setup ===" banner and that the game proceeds
into the menu) before the next zip is cut.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
arcattackandClaude Opus 5
|
bf40c53e1e |
Launchers: make the handoff wait tool-independent -- my first version broke on any PATH carrying Git-Bash/MSYS
Regression in my own fix from an hour ago, reported immediately by the
operator on relaunch:
find: '/I': No such file or directory
find: 'btl4.exe': No such file or directory
The game has exited. ...
The wait loop called bare `find`, which resolves to the MSYS/Git-Bash **Unix**
find on any box with git on PATH (confirmed here: `which find` ->
/usr/bin/find, `which timeout` -> /usr/bin/timeout). So the loop errored out
and fell straight through -- restoring exactly the misleading "The game has
exited" message it was meant to remove. `timeout` was the same exposure one
line later (GNU coreutils ships one, with different arguments), so the block
had TWO PATH-dependent failures.
Harmless to the game (the loop only reports; it never touched the session),
but it made the fix a no-op for anyone with git installed -- which is every
developer and a fair number of players.
FIX: no external tools at all.
* detection is now a pure-cmd `for /f "tokens=1"` over tasklist output,
comparing the first token to btl4.exe -- no find/findstr;
* tasklist and timeout are both invoked by ABSOLUTE %SystemRoot%\System32
path, so a shadowing PATH cannot reach them.
Failure modes stay safe: if tasklist is unavailable the variable never gets
set and the loop degrades to reporting immediately (the old behaviour, never a
hang), and a genuinely dead exe still falls straight through so the #41 launch
forensics are untouched.
VERIFIED both branches by running the shipped detection lines from THIS bash
shell -- i.e. the hostile shadowed PATH that broke v1, a harder case than a
playtester's clean environment:
game running -> RESULT: WOULD-WAIT [process detected]
absent-name control -> RESULT: WOULD-EXIT [no process]
No find/timeout errors in either.
(players/ is the shipped source per tools/mkdist.py:78; the gitignored root
copies re-synced byte-identical.)
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|