ApplyNetworkSettings disposed the station link without clearing it, so
ApplyStationLinkSettings disposed it a second time and threw. Only the
second function touches the link now, and StationLink.Dispose can be
called twice.
Contacts from the network are queued and applied together on the window
thread. Each one used to rescore the whole log and redraw the log window,
which kept the window thread busy for minutes during a resync.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The bridge launched its helper only through Wine. It now starts any
process, so on Windows it runs the helper directly. WineBridgeChannel is
renamed ProcessBridgeChannel.
WindowsControlSetup sets up the control side on Windows. The digital
dialog, AppSession, README and docs/digital-bridge.md describe the
Windows setup. publish-windows.sh builds a Windows release.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
N1MM records a phone message from the entry window: Ctrl+Shift+F1 to
Ctrl+Shift+F12 records into the file that key plays, and the same pair again
stops and saves it. Escape and sending anything stop one too — there is one
microphone, and it cannot be recorded and transmitted at the same time. The
status line says what is being recorded and what was saved, which is where
N1MM says it.
The recording being made is the session's rather than the window's, for the
same reason: the voice keyer window and these keys record into the same
microphone, and now each can see what the other started.
Alt+K opens the function key editor for the mode the radio is on, which is
N1MM's shortcut for what right-clicking a key button already did.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0157i7PiZyqBkquxS9A6FTdQ
A phone function key names a wav file instead of words, which is what N1MM's
own SSB file holds: `F1 CQ,{WAVDIR}\Cq.wav`. `{WAVDIR}` is the folder named
after whoever is at the microphone, a name is taken relative to `Wav`, and
commas play several files in a row. An SSB `.mc` file written for N1MM plays
here.
Window ▸ Voice Keyer lists every recording the messages name, says which of
them are there for the operator picked at the top, plays one, and records over
one. The recording is written beside the wanted file and moved over it at the
end, so a recorder that will not start takes nothing away.
`CommandSoundCard` runs `aplay` and `arecord`, and both commands are settings:
the audio path belongs to whatever the station runs, and a library would pick
one of them for the operator. A recorder is stopped with an interrupt so it
closes its wav file, and killed only if it ignores that. PTT is cwdaemon's
`<ESC>a` for as long as the recording plays, off for a station on VOX.
Two things follow. The QTC window sends on SSB: N1MM's four recordings on the
Ready, Agn, Cfm and Close buttons, with the lines still read out by the
operator. And alternating CQ duels on phone, because the voice keyer says when
a recording has finished — which it had to, since the CQ message is picked by
the radio's mode and would otherwise have been keyed as Morse.
Nothing here has been played on a radio: this machine has no sound card, so
what is tested is the program being run, the file filled in, the exit code and
the interrupt, with shell commands standing in for the two.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0157i7PiZyqBkquxS9A6FTdQ
The score window and the stored Contest.Multiplier1Name row both said
"Prefixes" for an Asian entrant, which counts DXCC entities.
Contest.MultiplierNames is now MultiplierNamesFor(StationInfo me), the
same shape as ExchangeFieldsFor. Every other contest returns the same
list it did before; All Asian returns "Countries" from Asia and
"Prefixes" from everywhere else. ContestDefinitions.For takes the
station as well, because that name is written to the log for N1MM.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0157i7PiZyqBkquxS9A6FTdQ
The QTC window put nothing on the air on RTTY except the lines themselves,
so the operator typed the rest into the digital window's transmit pane and
copied the lines by hand. N1MM sends a message for each step and fills the
lines in from a click. Both are here now, with N1MM's messages, N1MM's
defaults and N1MM's settings behind them.
The messages, all of them settings in the QTC window's setup area:
- RX Ready, `WAERXReadyText`, `{ENTERLF}I AM QRV BK`
- All Agn, `WAEAllAgnText`, `{ENTERLF}ALL AGN ALL AGN BK`
- Agn n, `WAEAGNText`, `{ENTERLF}AGN QTC: $ $ $ `, where `$` is the line
- Save, `WAESaveQTCText`, `{ENTERLF}R R R ALL OK BK`
- R U QRV, `WAERUQRVText`, `{ENTERLF}R U QRV BK`
- Save, reading traffic out, `WAESSaveQTCText`, `{ENTERLF}GL DE {MYCALL}`
Each keys the transmitter, sends the text and drops it again, and an empty
one sends nothing. The line spacing default is N1MM's `{ENTERLF}` rather
than `{ENTER}`, which was wrong.
The buttons: R U QRV and Snd Hdr were hidden while reading traffic out on
RTTY, though that is where N1MM starts — R U QRV is the first button
pressed. All Agn is new, and so is the S button on each line, which strikes
a line through a QTC that arrived too broken to read so the rows after it
keep their places; Agn n clears the strike and asks for that line again.
Clicking a line in the digital window fills the next row, which is how the
series is copied on RTTY: the other station sends the header and all ten
lines in one transmission. A click anywhere in a line hands the whole line
to the QTC window, which reads the time, the callsign and the serial number
out of it by shape and flashes the line to show it was taken. A line that is
not a QTC is left to mean what a click always means there.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0157i7PiZyqBkquxS9A6FTdQ
The README and the comments said the pool "goes out" and named it in
places where the reader cannot tell what is sent. They now name the
message and what it holds: the highest number taken on each band.
One of them was also wrong. The pool has not been sent on a reset since
the reset started counting from the log, and the README still said it was.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015NZkFxHde1iifAQemFTHxf
The button started the numbers again at one. N1MM's manual says it "sets
serial numbers on all computers to the highest value in any of the logs,
plus 1", and its own handler re-reads the highest number per band out of
the log, so a mixed network came apart at exactly the point the button is
meant to repair: the N1MM stations counted on from the highest number and
this one started at one, sending numbers that had already gone out.
ContestSession.RestartNumbers fills the pool from the log, which is where
the reset now lands. SerialNumbersFrom is gone with it, and so is the pool
this station sent afterwards: N1MM sends none, because each station works
the same number out of its own copy of the log.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015NZkFxHde1iifAQemFTHxf
Each station worked out its next number as the highest in its log plus one,
so two operators typing at the same moment both sent the same number and
one of the two contacts was wrong in the sponsor's log.
SerialNumberPool holds the highest number taken per band, which is N1MM's
QSONumbers array. A number is taken at the first character of the callsign,
as N1MM takes it, and ReserveNr tells the other stations before the contact
is worked. A number at or below one this station has used is answered with
the whole pool, N1MM's QSONRS, and the station that is behind counts on from
there. The pool also goes out when a contest is opened and when the numbers
are started again.
Contest.NumbersByBand says whether the numbers run per band or as one
sequence for the station. CQ WPX runs one per band for a multi-two and a
multi-multi; every other contest here runs one sequence. The band index in
the message is N1MM's, so the array lines up with theirs.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015NZkFxHde1iifAQemFTHxf
Seven tabs now, in N1MM's order, and the messages behind them.
**Options** holds what there is something behind: Log QSOs at all Stations,
which decides whether the contacts the others log are kept here; the
message trace, moved off the Messages tab where N1MM does not keep it; and
Run1/Run2, which used to be guessed from the radio number and is now the
operator's choice, sent in every `STATUS` and every `XMIT`. The transmit
interlocks stay out: this program does not block a transmitter, and a
station that claimed it did would be waited on for nothing.
**Call Stack Targets.** A callsign heard at one station and handed to the
position that will work it — `STACKANOTHERCALL` with the call alone,
`STACKCALL` with the contact behind it. The target is picked from the tab
or from the station list, and what arrives goes on the call stack of the
radio the operator is on, sorted by whether it is a multiplier here.
**The master** is the station the entry follows for cluster spots and the
time. `MASTER` names it to everybody, so the column agrees everywhere.
**The clock.** `TIME` carries a station's clock and Force Time Sync sends
it. N1MM sets the clock from it when it is running as an administrator and
says how far out it is when it is not; this program always does the second,
because setting the clock needs root and a contest logger has no business
asking for it. Anything over two seconds is called out.
**Serial numbers.** Reset Serial Numbers at all Stations sends
`RESETQSONRS` and every station starts again at one. The contacts already
logged keep the numbers they were sent with.
**What the entry agrees on** is now checked rather than assumed:
| What | How |
|---|---|
| the contest | `CONTESTNAME` as soon as a connection opens, and again in every `CHECKSUM` |
| the country file | its issue date — the publisher's `=VER20260826` — rides in `STATUS` |
| the computer name | longer than N1MM's fifteen characters is named at startup; a second computer answering to this one's name is named when its beacon arrives |
| the station number | two stations on the same number are named when the second says so |
`CountryFile` reads its issue date, which nothing had asked it for before.
Left for later: the master relaying its cluster spots, and the shared
serial-number pool where stations claim numbers in advance.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
Three of the messages this program sends were its own invention or in the
wrong units, so an N1MM beside it read them wrongly or not at all.
| Message | N1MM | What was sent |
|---|---|---|
| `XMIT` | transmitting, the function key's caption, Run2, F1 pressed, stop sending | transmitting, the radio number |
| `PASSFREQ` | the frequency in hundredths of a kilohertz, and nothing else | the frequency in kilohertz, and a callsign |
| `FREQ` | the frequency in hundredths of a kilohertz | `BANDMAP`, which N1MM has no such message for |
So N1MM showed a radio number where the other operators read what a
station is sending, read a pass on 14205.5 as 140.25, and never saw this
station move at all. `BANDMAP` is gone: the frequency goes out in `FREQ`
when the radio moves and the mode, run and the rest ride in the `STATUS`
that follows it, which is where N1MM carries them.
The `Msg` column now shows what a station is sending, which is what the
caption is for. `PassCall` and `RadioNumber` are gone from the station:
neither is on the wire, so neither was ever anything but this program
talking to itself.
**A station that reconnects resynchronises by itself.** N1MM sends
`LASTQAT` — when I last worked anybody — as soon as a connection opens,
and a station holding nothing that late asks for the hours in between. An
empty log asks for nothing: a program that has just been started has no
business pulling the whole contest across, and N1MM suppresses it the same
way. The answer goes over the connection this station opens, which is not
always up when the message arrives, so a resync that has nothing to ask
over waits for it rather than being lost — and one that still cannot send
stops and says so instead of leaving the row resynchronising for the rest
of the contest.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
A station that was away — a laptop that lost the network, one that was
switched off — comes back missing whatever the others logged while it was
gone. N1MM fills that in by walking the log an hour at a time, and this is
that walk, on the same messages:
| Message | What it says |
|---|---|
| `CHECKSUM` | here is what I hold for your contacts in this hour, and the contest and callsign I am logging |
| `DELETEQS` | drop your copies of my contacts in that hour, I am sending them again |
| `RESYNCQSO` | one of those contacts |
| `CONFIRMED` | here is what the hour hashes to at my end — which also asks for the next hour |
Each station is the authority on the contacts it made, so an hour that
disagrees is replaced rather than merged. The answer to one block asks for
the next, and the walk ends when the block starts after the end of the
period. The Resynchronize tab counts the hours that agreed, the hours that
were sent again and the hours that still disagreed, which are N1MM's
columns.
**`LogChecksum`** is N1MM's: a CRC32 over the `DXLOG` columns of the block,
ordered by time and then callsign, leaving out what each station works out
for itself — points, the three multiplier flags, the country prefix, the
continent, the WPX prefix and `IsOriginal`. Two stations may disagree about
those and still hold the same log. A checksum that disagrees for any other
reason costs one block of traffic, which is what a resync is for.
**The link still holds no log.** It asks one through `ContactLog` for the
contacts of a computer in a block of time, and `SessionContactLog` answers
on the window thread, because that is where the log is touched.
**Every contact logged here now carries the computer that made it.** N1MM
stamps `NetBiosName` and `NetworkedCompNr` on every row and this program
did not, so its own contacts belonged to no station and a resync could not
ask for them.
**Rescore** writes the points and multiplier flags of a period back to the
database. The log in memory scores itself whenever it changes; the stored
rows do not, and N1MM reads them rather than working them out. A contact
that arrives from another station after the one that took a multiplier
leaves those columns behind. A station can also ask the others to rescore
when its resync finishes, which is the flag on the end of `CHECKSUM`.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
The window was one flat table of columns this program made up. N1MM's has
tabs and a fixed set of columns, and an operator who knows one has to be
able to read the other.
**The window** now has Stations, Actions and Messages. Stations is a grid
with N1MM's headers in N1MM's order — Computer, IP Address, Pass, Run,
10, 100, Freq, Op, Msg, Send, Receive, Band, R/M, Mode — and a row per
station, this computer included. Right-click a row to pass the contact
being worked to that station, to talk, to show the connection log, or to
close and open its connection. Actions holds the talk line, the pass
frequency, the echo and the networking switch; Messages holds the chat,
the connection failures, the echo round trips and a trace of every
message that arrives.
**`STATUS`** is what fills those columns. N1MM sends it every few seconds
whether anything has changed or not — the pass frequency, the frequency,
run or search and pounce, the operator, the rates over the last ten and
the last hundred contacts, the mode, the categories and the version — so
a station sitting on one frequency for an hour still has a current row.
`BANDMAP` only says a station moved. The five transmit-interlock flags
are sent as false: this program does not block a transmitter.
**`Rates.PerHour`** reads the two rate columns out of the log. Count back
over the last N contacts, stop at a gap longer than the contest's minimum
off time, and treat a span under a minute as a minute, so three contacts
in five seconds read as 180 an hour rather than 2160.
A connection closed by hand stays closed. Without that flag the next
beacon opens it again within ten seconds, which makes the menu item look
broken.
Left out, because there is nothing behind them: N1MM's Resynchronize and
Rescore tabs, the transmit interlocks of its Options tab, the Master
column and the call-stacking target tab.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
Three pieces on top of the link:
**The network status window**, on the entry window's Window menu. A row
per station, this computer included the way N1MM shows it: number,
address, operator, band, mode, run, transmit, pass frequency, the last
message type, how long ago it arrived, the counts each way and the echo
round trip. A station that broadcast the wrong version is named under the
table in red — that is the one fault where everything looks connected
and nothing arrives. The box at the bottom sends a line of chat, and
Echo asks every station whether it is there.
**Where this station is** goes out once a second when it has changed.
The frequency, the mode and run have a dozen places they can change from
— the radio moving, a band button, a QSY typed into the callsign box —
so it is read and compared rather than announced from each of them.
**The ten-minute rule.** `BandChangeRules` already counted changes and
the stay on a band, and the entry window already showed the countdown,
but only a user-defined contest carried a rule, so every built-in
contest allowed anything. `ForCategory` gives a multi-operator entry
with one or two transmitters a ten-minute stay, which is N1MM's fallback
in `ContestInstance.BandChangeTimerDuration` for every contest that does
not name its own. Its per-contest table is a few hundred cases in a
decompiled hash switch and is not repeated.
Config ▸ Edit Networked-Computer Names now covers both networks: the
12060 broadcast to other programs and the 12070 link, with the station
number, the port, the version to claim and stations named by address for
a network where a broadcast does not reach.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
The macro buttons went wrong at the end of a transmission, both faults in
DigitalEngineSender:
- The engine reports the transmitter drop on its own thread.
WhenTransmitChanged took the state lock, decided the message had ended,
released the lock, and only then called Buffer.Ended(), which clears
what has gone to the engine. A macro pressed on the last character got
through StartAsync in that gap and had already flushed its own text into
Sent, so Ended() wiped the new text off the pane while the engine
transmitted it. Ended() is now called inside the same lock.
- The pane only started again when the engine reported a drop. Two macros
in a row keep the transmitter up, so that report never came and Sent
grew with every press. Everything in Sent is locked, because it is in
the engine and cannot be taken back, so the whole pane became
read-only. TypeAhead.Started() drops the last message's sent text and
keeps what was typed ahead, and StartAsync calls it whenever it keys a
new transmission.
The rest of this commit is the digital transmit work these fixes sit on:
the pane as one coloured box, the sender's three keying states, the
type-ahead feeder paced by the clock with the engine's count as a brake,
{RX} flushing what is left in one piece, and the entry window's function
keys reading the digital macros on a digital mode.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PdAYHcdRktqKry7nk414TU
The engine probe answered two questions on the air, and they point opposite
ways.
TxBufLen is the number of characters left to transmit. Seventeen characters
pushed, five decoded back, and it read 12, counting down to 0 as the message
went out. So the pump no longer counts character times off the clock: it asks
the engine, feeds while the answer is under Lead, and asks again. The baud rate
is now only the fallback for an engine that will not answer.
A backspace is not an edit. Pushed in as a character it made the count go up by
four and the text went out unchanged, so text the engine has been given cannot
be taken back, only aborted. EngineTypeAhead was built on the opposite belief
and is gone, along with the setting that chose it and the interface that existed
to switch between the two.
A count that stops going down means the engine is holding what it has: MMTTY set
to Word out keeps a word until the space after it. The pump feeds one character
per look while that lasts, so the space arrives and the word goes out rather
than the message sitting there.
The probe kept two bugs of its own that this run showed: it read the count a
millisecond after pushing, saw 0 and called the message finished, and it started
a step while the previous unkey was still in flight, so that step ran with the
engine down. It now needs two empty answers in a row and waits for the engine to
say it has stopped. It also asks what {RX} does to a word the engine is holding,
which decides whether a macro without a trailing space loses its last word.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RtspmWmS7f8kUvcyaHpRWZ
MMTTY has a type-ahead buffer of its own: characters go into it, a backspace
takes back one it has not transmitted, and TxBufLen says how many are left. Used
that way it paces itself, so there is no gap between characters to tune and no
baud rate to keep in step with the engine.
EngineTypeAhead does that, and Config > Digital picks between it and the pump
that is there now. Off is still the default: three things it rests on have never
been seen with a real engine.
tools/Nonemm.EngineProbe asks the engine those three questions and writes the
answers to a file. It starts MMTTY through the bridge, pushes a message, polls
TxBufLen while it goes out, backspaces over text that has and has not been
transmitted, and logs what came back on the receive side and when.
The bridge learns one verb for it: `buffer` reads TxBufLen and answers with the
count, or -1 when the control will not say. A property the control does not know
is a log line rather than an error, since it stops nothing.
TypeAhead and EngineTypeAhead share the TransmitBuffer interface, which is what
the digital window now works through, so the window does not know which one it
has.
docs/unfinished.md states what each buffer assumes and how to run the probe.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RtspmWmS7f8kUvcyaHpRWZ
The transmit pane is now the type-ahead buffer. What has gone to the engine is
in red beside the box and cannot be reached; what is still to go is in the box
and can be typed over, added to or deleted while the engine works through it.
A macro fills the pane rather than the engine, so the operator can correct a
call the message is still sending.
The cursor holds the pump: nothing behind it goes out, so an engine that
catches up with the operator idles instead of transmitting half a word. With
the box out of focus there is no cursor and everything goes.
Escape and the RX button drop what has not gone out and stop the engine where
it is. Enter puts in the carriage return the engine takes as a new line. The
pane starts empty for the next message.
The baud rate the pump counts by is in the digital setup window, beside the
alignment frequency. It has to match the engine: 45.45 unless the contest says
otherwise.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RoGtneMQaz4M9w7Kk49AVD
Ctrl+O asks who is at the radio, as it does in N1MM, and the callsign goes on
every contact logged from then on. A multi-operator station changes hands every
few hours, so each operator also keeps what he wants the program to look like:
where the windows sit, the theme, and the folder his recordings are in.
The windows on the screen are stored against the operator leaving, before the
dialog opens, and the operator taking over gets his own back. A window he opens
later is placed as it appears. An operator who is new to the list starts with
whatever is on the screen.
Nothing plays the recordings yet, so the folder is stored and waits on voice
keying. {OPERATOR} in a message sends the callsign.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RoGtneMQaz4M9w7Kk49AVD
MMTTY and 2Tone have no socket or pipe interface: N1MM hosts XMMT.ocx and
exchanges window messages with the engine. A Linux process cannot load that
control, so bridge/nonemm-mmtty-bridge.exe hosts it under Wine and passes
lines over its standard input and output. docs/digital-bridge.md states the
protocol and what the control needs.
The window is N1MM's: receive pane with coloured callsigns, grab list, call
stacking, twenty-four macro buttons and the engine controls. One left click
copies what is under it — a callsign to the callsign box, anything else to the
exchange box the contest keeps for that kind of value.
Config > Digital registers XMMT.ocx in the Wine prefix on its own, making the
prefix first if it is not there. The engine, the bridge and the control start
at the copies shipped beside the program.
The bridge reads the control's own events. OnTranslateMessage carries only the
messages the control has no event for, so nothing was ever decoded through it.
hamlib's data mode names read as digital modes now, and a mode typed into the
callsign box changes mode the way a frequency changes band, so a station with
no radio can reach RTTY at all.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RoGtneMQaz4M9w7Kk49AVD
MultSqlString = CallHist counts every station the call history file
lists, once per callsign, which is how N1MM dedups it. The file reaches
the scoring code through ContestLog.History, and a QsoContext now
carries it; without a file loaded nothing counts.
BonusPoints2 = +50, calls.txt reads the callsigns from the support-files
folder and adds 50 to the contact, *2 doubles it and a plain 50 scores
50 instead. The folder comes from the registry, which the app passes
when it reads the .udc files. N1MM's other form, where the file is
grids.txt and the bonus is looked up by grid square, is not read.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RoGtneMQaz4M9w7Kk49AVD
AppSession.TakeFromNetwork wrote the contact to the store and then called
OpenContest to see it, which built a new ContestSession, read the whole log
back and built new entry positions. Anything the operator was typing at either
radio went with them, and the entry boxes were rebuilt under the other
operator's hands. It also ran on the socket thread, so it was changing the log
while the windows were reading it.
ContestSession now applies the three messages itself: AddFromNetwork,
ReplaceFromNetwork and DeleteFromNetwork write the store, change the log in
place and raise Changed. None of them raises Logged, Edited or Deleted, so
nothing goes back out to the other stations. Points and multipliers are still
worked out here from the rules rather than trusted.
A contact that arrives goes in where its timestamp says it belongs, through
the new ContestLog.Insert, and the log is scored again: of two stations that
worked the same multiplier, the one that worked it first keeps it, however
late the message turns up. Add would have appended it at the end and given the
multiplier to the wrong contact.
The socket thread now hands the update to the thread the windows run on, which
is where every other change to the log is made.
The matching rule moved with the code, so ContestSession answers what a
message from N1MM refers to: the contact id, then the call and time it had
before the edit.
Still not tested against a second station. Two of these cannot run on one host
to try it, because they share one UDP port.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RoGtneMQaz4M9w7Kk49AVD
The menu tree comes from N1MM's EntryWindow.Designer.cs: File, Edit, View,
Tools, Config, Window, with the items under the names N1MM gives them. The
window list moved out of View into Window; the call history file moved from
Config into File > Import. Ctrl+W, Ctrl+L and Ctrl+M now work, so the keys
printed in the menu are real. Two commands that had no menu item before:
the CW and SSB function key definitions, and the QTC setup area.
The rest of the Edit menu:
- Ctrl+N asks for a note and puts it in the comment of the contact in the
boxes, or of the last logged contact when nothing is typed.
- Edit Current Contact opens the Edit Contact form over what is typed but
not logged. Update hands the values back to the entry boxes.
- Ctrl+Q and Ctrl+A load a logged contact into the boxes and paint them
pale yellow. Enter writes the change back, Esc puts back what was being
typed, and stepping forward past the newest contact leaves quick edit.
- Ctrl+U raises the received serial, or a numeric exchange box when the
contest has no serial.
- Ctrl+F shows the call being typed in the log window, then the next
contact with that call.
RadioPosition is now OperatingPosition: it is the operator at one radio,
not a place on a band.
Looked at under Xvfb on a 3775-contact log: quick edit back and forward,
Esc, the note dialog, find, and Edit Current Contact putting a zone back
into the entry boxes.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two differences from N1MM in the spot path, both found by reading Packet.cs.
Randomising moved every CW spot. PacketSpot.Randomize moves a spot only when it
is not a dupe, not a self spot, not on a CQ frequency, is on CW and is not
working split, and Packet skips it altogether for a station passed to another
band. Moving a dupe or a split station makes it harder to find, not easier, so
this now leaves the same ones alone. Whether a station has been worked comes
from the log, so AppSession asks it before moving a spot.
Split needed knowing where a station is listening, which is in the comment. Qsx
reads it by N1MM's rules: QSX, UP, DOWN, DN, U or D, the word standing alone or
after a space, a bare number after QSX meaning kilohertz inside the band and
anything else an offset from the spot, an offset over a hundred kilohertz
refused unless it says QSX, DN70 read as a grid square rather than five down,
and a listening frequency outside the band thrown away.
Spots reached the bandmap one at a time, each one redrawing every window that
watches it. N1MM queues them and flushes the queue once a second, which is what
this does now: the filters and the randomiser run over the batch, and the
bandmap takes it in one call and raises one change.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
N1MM has two columns of band buttons beside its entry window, CW in one and
phone in the other, with the band the radio is on marked. This has the same.
A button does not jump to the band edge every time: BandMemory keeps the
frequency the operator was last on for each band and mode, so leaving 20 metres
at 14032 and coming back lands on 14032. The first visit to a band starts where
that part of it starts, which the band plan says — the bottom edge for CW, the
CW-to-phone boundary for phone — and a band the plan says nothing about starts
at its bottom edge.
Phone picks the sideband the band calls for, so 40 metres is lower sideband and
20 is upper.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The messages were twelve boxes in a dialog, which meant no labels, no
search-and-pounce set, and no way to read a file somebody published. They are
now the file itself: right-click any function key button under the entry
window, or the two buttons in Config ▸ Keyer and messages, and the text of an
N1MM .mc file opens in a plain editor with Import, Export and Back to the
defaults.
MessageFile reads that format by N1MM's rules. A line starting with # is a
comment; every other line is a label, a comma and the message, with && standing
for one & in a label. Which key a line belongs to is decided by where it is and
nothing else: the first twelve lines are F1 to F12 while running, the next
twelve the same keys while searching, and a file that stops part way through the
second twelve leaves the rest sending what they send while running. A blank line
in the middle is a key with nothing in it, which is what the manual says; the
newline that ends the last line is not, which it does not say, but no .mc file
in the world means its final newline as a message.
Two things fall out of this. The buttons now say what the file says, and they
change when the operator moves between running and searching, so the labels are
the documentation N1MM's manual suggests writing them as. And the search set is
real: ESM's F2 while searching can be a different message from the one it sends
while running, which is how the file was always meant to be used.
Settings keep the file text and carry the twelve stored messages into it on
first read, with the labels this program used then, so nobody loses what they
had typed.
The editor is deliberately a text box rather than a grid of fields: importing,
exporting and editing are then the same thing, and the comments an operator
writes in the file survive being edited here. The telnet window's buttons can
use the same editor later.
Running it caught two: Save as the default button ate the Enter key that a
multi-line editor needs, and the button labels did not follow the move between
run and search until they were redrawn rather than only rebuilt.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The packet window showed traffic and took a command line, and everything else
about the cluster lived in a dialog under Config. It is now the window N1MM has,
with the same five tabs.
Telnet shows the traffic with spot lines in green, what went out in blue and
lines from a preferred spotter in bold. Double clicking a spot line, or "Jump to
this spot", puts the radio there with the call in the entry window. Scrolling
stops while the pointer is over the traffic. The client keeps the last two
hundred lines, so a window opened mid-contest is not blank.
Clusters keeps the operator's nodes with their ports, passwords and after-login
commands, connects and disconnects, and holds the logon settings. Download
fetches the published list of telnet nodes from NG3K — around fifty, with the
sysop's call and a note about each — and clicking one fills the boxes in. N1MM
downloads its list from its own web service, which asks the operator to opt in
to data collection and is N1MM's to run, so this reads a public page instead.
ClusterList takes any page with telnet:// links in a table; a page it cannot
read leaves the stored list alone.
Filters decide which spots reach the bandmap: bands, modes, beacons, busted
calls, stations outside the call history file, blacklisted spotters and calls,
spots from outside your country, continent or a list of prefixes, and how long a
spot stays on the map. A busted spot is a call the callsign database has never
heard that is one character away from one it knows; a call nothing resembles is
kept, because that is what a new station looks like. Nothing is filtered out of
the traffic itself — the operator sees everything the node sends.
Buttons edits the twelve command buttons. A button takes what N1MM's takes: the
message macros, several commands separated by semicolons, or {CONN} and the name
of a favourite, which connects to that node instead of sending anything. The
label takes the macros too. Right-clicking a button opens the editor.
Config ▸ Cluster now opens this window on the Clusters tab rather than a dialog
of its own, which is where N1MM keeps those settings.
Three things that could take the program down while a cluster was connected:
Settings.Load read a null where the property is not nullable. A file that names
a key with a null value — one written before the property existed and then
edited — put that null straight through, because the property's own default only
runs when the key is missing. Opening the telnet window then threw on the first
list it touched. Every null is now put back to the default the property
declares, walking into the stored records and the lists of them.
Bandmap was written from the cluster's thread and read from the window's, so a
dictionary could be modified while a window enumerated it. Every method locks
now.
ClusterClient disposed its token source while its own loop still used it, and
the retry delay sat outside the catch, so a disconnect faulted the loop task.
Along the way the message macros were checked against N1MM's function-key
documentation, and several were wrong. {LOGGEDCALL} is N1MM's {LASTCALL}, the
serial is #, and there is no {MYZONE}; {NAME} and {GRIDSQUARE} stand for the
other station's name and grid, not ours; {OTHERMHZ} is the radio the operator is
not on. The single-character macros * and ! were missing. Added from the same
table: {LASTCALL}, {PREVNR}, {NAMEANDSPACE}, {CHNAME}, {GRID}, the two grid
bearings and the grid distance, {FREQ}, {FREQROUND}, the other-radio
frequencies, {TIMESTAMP} and {TIME2}. Frequencies are formatted the way N1MM
formats them, with R for the decimal point on CW. The macros that pass a station
to the other band take the second radio as a new argument, and stand for nothing
at a one-radio station.
Left out, and written down: saving spots to a database, which N1MM keeps in its
admin database rather than in the log file the two programs share; the
special-calls list; the two-character busted check, which is a few hundred
thousand lookups per spot against a few hundred for one character; and N1MM's
action macros, which need a different shape than an expander that returns a
string.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Ctrl+Z opens the QTC window for the station being worked, which is the key N1MM
puts it on, and the window is laid out the way N1MM lays it out: the header,
ten lines of time, callsign and serial number, an Agn and a Cfm button beside
each, Clear, Close and Cancel underneath, and the same colours — green for a
line that will be saved, red for one still empty, yellow for one that cannot be
read. Enter and Tab move forward and space moves within a line, so a line can be
walked while listening to it.
Sending, the lines are filled from the log and cannot be edited: the contacts
not reported to anybody yet, oldest first, never the station being worked, and
never past the ten any station may have. The rule for what has been reported is
N1MM's — a contact counts as reported when a QTC row carries its call and the
serial we sent it — so the pair is kept in the row the way N1MM keeps it.
Which way traffic goes is not the operator's choice on CW and SSB: Europe
receives and the rest of the world sends. On RTTY it goes both ways, and there
the window carries a switch. That is the one piece of the workflow that is ours
rather than N1MM's, which toggles a QTC mode outside the window instead.
Nothing here transmits. N1MM sends the lines on CW and RTTY and plays recorded
voice messages on SSB; we have no QTC message templates, no voice keyer and no
digital window, so Agn, Cfm and RX Ready move the cursor and the operator sends
by hand.
Running it caught what the tests could not: focus set in the constructor does
not stick, so the first thing typed into a freshly opened window was dropped and
every field after it took the value of the one before.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Ctrl+B calls CQ on one radio, and when that message has gone out, moves to the
other and calls there. The keyboard, the entry window and the SO2R box follow
each turn. N1MM calls this dueling CQs and puts it on the same key.
The turn is taken when the keyer says the message has ended, not when a timer
guesses it has. So MessageSender grew a Finished event and a ReportsCompletion
flag, and both keyers fill them in: cwdaemon answers the <ESC>h reply request
that now goes out in front of every message, and a WinKeyer clears the busy bit
in the status bytes it sends of its own accord. The status-byte reading is in
WinkeyerStatus, away from the serial port, because that is the half that can be
tested without a keyer on the desk. A keyer that reports nothing refuses to
start alternating CQ rather than keying the second radio over the first.
AlternatingCq itself takes the keyer, a callback that calls CQ on a radio, the
gap and a wait function, so the alternation is tested without sleeping. The gap
is in Config ▸ Keyer and messages and will not go below 100 ms, which is N1MM's
floor too: an SO2R box works relays.
docs/keying.md writes down why cwdaemon does the timing and we do not. N1MM
keys DTR itself with a coarse sleep, a busy-wait and a margin that grows every
time the sleep overshoots, and it raises the thread to TIME_CRITICAL for the
length of the message. The busy-wait ports to Linux; the priority does not,
without CAP_SYS_NICE, and a garbage collection mid-element is audible. A direct
serial keyer stays a reasonable third option, to be taken knowingly.
Running it against a fake daemon that takes 1.5 seconds to play a message: six
CQs went out back to back and the keyboard moved between the two entry windows
each time. Escape stopped it, let the message in flight finish, and started
nothing further.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Behind the bandmap scale, the CW part of the band is one shade and the phone
part another, with a third for a digital block. That is what N1MM's graphical
bandmap does: ResizeBandPanels puts a CW, a Digi and an SSB panel behind the
scale, sized from its CWBands, DigiBands and SSBBands tables. Anything past a
band edge is shaded red, which is the only place the edges themselves show, and
it only appears now that the slice can reach past an edge.
The defaults are N1MM's: CW ends at 1840, 3600, 7125, 14150 and so on, and the
digital pair is empty until somebody fills it in, exactly as N1MM ships. They
are one worldwide set rather than one per ITU region, which is also N1MM's
answer — the reason recorded for leaving this out, that the segments differ by
region, was wrong about what N1MM holds. 60M and everything above 2M get no
entry, because N1MM's own numbers there contradict themselves: its 1.25M CW top
is below its phone start, and its 70CM CW top is above the band.
Config ▸ Sub bands edits them, one row per band in kilohertz, with a Defaults
button. Only the rows that differ are written to the settings file, so a plan
nobody has touched leaves nothing behind. BandPlan is an object rather than a
static table, so the stored changes layer over the defaults and the bandmap
picks up a save without a restart.
The numbers are a drawing aid, not a licence. Nothing stops a contact being
logged either side of a boundary.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The Available Mults and Qs window: a column per band, the multipliers at the top
of each column and then the stations that are only points. Clicking one puts the
radio there with the call in the entry window, the same as the bandmap does.
Every spot on the bandmap goes through ContestLog.Judge, and what comes back a
dupe is dropped, so the colours and the ordering agree with every other window.
A spot holds a callsign and a frequency and nothing else. For CQ WW that gives
the country multiplier, which is derived from the call, but not the zone, which
is what the other station sends. Where a call history file has an entry for the
call its exchange is filled into the candidate first, so a published CQZone
column brings the zone in and a Sect column makes Sweepstakes work at all.
Filling the zone from the country file instead was rejected: it is wrong for
every large country, and a multiplier that is not there is better than one that
is not real.
A spot has no mode either, so the candidate is judged in the mode the radio is
in. In a mixed-mode contest that means the answer follows the operator.
RadioPosition kept two private mappings — exchange value into a contact, and
call history field into an exchange box — and both are needed here, so they move
to ExchangeSlots and both callers share them.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
N1MM's format, so a file published for a contest is read as it is: # comments,
!!Order!! naming the columns of the lines after it, semicolons in preference to
commas so user text can hold a comma, and a default column order for a file
that names none. The directives that change what is stored are acted on;
!!Order!! can appear again part way through, because these files are usually
several files stuck together.
Leaving the callsign box fills what the file knows into the exchange boxes that
are still empty. A box with something in it is left alone: what the station
actually sends beats what somebody published months ago. The zone comes from
the CqZone or the ITUZone column depending on which the contest asks for, and a
file with no Sect falls back to State.
The check window's two missing columns are in. History lists matching calls
from the file. Exchange is not about callsigns at all: it offers the values the
exchange box with the cursor in it can hold, which is what N1MM's exchange pane
does — it searches the contest's list of valid exchanges, not the log.
CheckCandidate.Call is now Text, because an exchange value is not a call.
The section-validating directives are read and passed over rather than acted
on. Acting on them means holding N1MM's section lists and its rules about
retired sections, and getting that wrong throws away good data.
Running it caught what the unit tests could not: the exchange was filled and
scored but the boxes on screen still looked empty, because nothing resynced
them.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
LoggingSession mixed two things: the contest, which one station has one of,
and what the operator is typing, which belongs to a radio. It is now
ContestSession and RadioPosition. Two positions share one log, one score and
one run of serial numbers, so a station worked on radio 1 is a dupe on radio 2.
Frequency, mode, split, what is typed and whether you are running belong to
each radio on its own.
A second radio opens a second entry window. It has no menu of its own: it is
another view of the same contest, not another program. Only the window for
radio 1 opens the database and the connections.
Ctrl+Tab moves the operator to the other radio and the keyboard follows.
Ctrl+Shift+Tab puts both radios in the headphones.
OtrspBox speaks OTRSP to an SO2R box over a serial port: TX1/TX2 for the key,
RX1/RX2 and RX1S for the headphones, AUXnnn for an output line. It follows the
active radio, and every message points the box at this radio before keying, so
a message cannot go out of the radio the operator has just left. A command that
would change nothing is not sent, because the box works relays; N1MM does the
same. The box takes a Stream, so what it sends is tested without a serial port.
Still missing: alternating CQ, and voice keying on the second radio.
Looked at under Xvfb with two fake radios, one of them split: both windows up,
radio 1 showing 14008.00 into 14020.0, radio 2 on 14030.00, Ctrl+Tab moving the
keyboard between them.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Split. RadioState carries a transmit frequency, zero meaning the radio
transmits where it listens, so split is not a second flag that can disagree
with it. The contact stores it in N1MM's QSX column, the entry window shows
14008.00 followed by the transmit frequency, and the bandmap draws a red bar
there. SetSplitAsync sets the frequency before turning split on, or a radio
last split somewhere else transmits there.
Every command now goes out with a + in front, asking rigctld for its extended
answer: named fields ended by an RPRT line. The raw answer is bare values with
no terminator, so the client had to know how many lines each command returns —
there was a `command == "m"` special case for the one that returns two. Add a
third such command and get the count wrong once, and every later answer is read
against the wrong command for the rest of the session. RigctldReply parses the
extended form and is tested on its own.
A radio that cannot do split answers RPRT -11. That is an answer, not a broken
connection, so the frequency and mode it did report still count.
Two radios. Settings hold a list rather than one host and port, with the single
radio an older settings file holds carried into it. Both radios are read and
both show on the bandmap, the active one green and the other orange, but only
the active one drives the entry window: the second radio moving must not drag
the operator off the station being worked. Ctrl+Tab swaps, and a contact
records which radio made it.
This is not a full two-radio operating position — no second entry window, no
alternating CQ, no audio switching. It is two radios read and logged correctly.
Tested against a stand-in for rigctld over a real socket, and looked at under
Xvfb with two fake radios, one of them split.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The old client read the socket with a StreamReader, so the option negotiation a
node sends on connect landed in the text as control bytes and made a mess of
the first lines. TelnetStream now answers it: it agrees to ECHO and
SUPPRESS-GO-AHEAD, refuses everything else, swallows subnegotiations, and
tracks what it already answered so the two ends do not reply to each other for
ever. A command split across two reads is still understood, because a socket
hands over whatever has turned up rather than whole messages.
LineAssembler keeps the tail that has no line ending yet. Nodes write their
login prompt as "login: " with nothing after it, so the old ReadLineAsync sat
waiting for a line that never came and only got through because the greeting
happened to mention a matching word.
Login now follows N1MM: it looks for LOGON, ENTER CALL, LOG IN and the rest,
and sends the callsign anyway after ten seconds if no prompt turns up. A
password is sent when the node asks for one and the operator configured one.
The commands go out after that, not straight after the call. A connection that
has heard nothing for four minutes gets a blank line so the node does not drop
it as idle.
The old prompt check matched "call" anywhere in a line, which any spot comment
could trigger.
SendSpotAsync sends the node's dx command. Alt+P, or Edit / Spot It, spots the
call being typed at the current frequency, or the last contact logged when
nothing is typed, and puts it on our own bandmap without waiting for it to come
back round from the node.
The spot parser now takes a line with no colon after the spotter, and finds the
time when DXSpider has put the spotter's grid after it.
The cluster tests run over a real socket against a node fake that sends the
negotiation, the prompt and spot lines.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Double-click a cell in the log window to change it. The columns follow the
contest exchange instead of being fixed in the XAML, so CQ WW gets a Zone
column and Sweepstakes gets Nr, Prec, Ck and Sec.
QsoEditor holds the column list and applies one field edit. Validation comes
from the ExchangeFieldKind the contest declared: a CQ zone is 1 to 40, an ITU
zone 1 to 90, a section is looked up in the ARRL list, a grid must parse.
A refused edit returns the reason and leaves the log alone, so the cell
reverts. Changing the callsign runs the country lookup again.
Delete, or the right-click menu, removes the selected contact after a
confirmation. Either way the log is rescored, so a multiplier the removed
contact held passes to the next contact that claims it.
Both go out to the other stations in N1MM's own messages: contactreplace
carries oldcall and oldtimestamp, contactdelete names the contact. An incoming
edit or delete is matched by contact id first, falling back to call plus
timestamp because N1MM does not know our ids.
Country, continent and the two prefixes were filled in twice, once when
logging and once when editing. They now come from CountryFields.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The function keys send through cwdaemon or a WinKeyer, with N1MM's message
macros. Escape stops sending.
Settings are read with the reflection serializer rather than a generated one:
the generated one hands back null for every property the file leaves out
instead of the value the property is declared with, which crashed the program
the first time a new setting was added.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Checked by opening a log this program wrote in N1MM 1.0.11031. Three things it
rejected:
- a log naming a contest with no row in the Contest table throws "No current
row" in Contest.FromRow, so the definition row is written whenever a contest
is opened;
- an empty overlay category is answered with "Invalid Overlay Category:", so an
entry with no overlay now says "N/A", and the dialog offers N1MM's list;
- the sent exchange omits the report, which is what N1MM's own contest dialog
asks for.
With those three fixed N1MM opens the log and shows the contacts.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The entry window types contacts and colours them as they are typed; the log,
check, bandmap, score and packet windows read the same session. Contacts are
shared with the other stations of a multi-operator entry in N1MM's contact
message, so an N1MM station on the same network sees them too.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
