A client connected to several rigs decodes all of them at once, and the
decode stream carries every rig's traffic to every browser. The decoder
panels listed all of it: a station a background rig copied on another band
appeared in the APRS list next to the selected rig's, the vessel counts and
the "latest seen" lines counted both, the status lines said "Receiving"
because some other rig was, and the CW pane interleaved two rigs into one
stream of text that read as neither.
The page is about the rig the operator selected — the one whose spectrum is
on screen and whose audio is playing — so each panel now shows what that rig
heard: rows, counts, latest-seen, status, the live picture a WEFAX or SSTV
frame is painting, and the CW pane.
Nothing is dropped on the way in. The map is the whole station's view, has
its own rig filter, and would empty out if the plugins stopped feeding it, so
the histories still hold every rig and the map still plots them. That also
means a switch loses nothing: the runtime gained a rerender hook, which the
rig switch calls, and switching back brings the other rig's traffic up again.
The CW pane is the exception — a running stream of text cannot be unpicked
after the fact — so it starts empty on the rig switched to.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SyX26FCpMQxiBoC7r5K1A7
Signed-off-by: Stan Grams <sjg@haxx.space>
Spectrum dominates the server↔client connection, and all three things that
govern its cost were working against a poor link.
**It was polled, one round trip per frame.** The client asked for a frame every
50 ms on a dedicated connection and waited for the reply, so the frame rate was
capped at 1/RTT — on a 200 ms link, five frames a second no matter what was
configured. Add SubscribeSpectrum alongside the existing SubscribeMeter: the
server pushes frames from a per-rig broadcast that rig_task fills only while
somebody is subscribed. A server too old to know the command answers with an
error and leaves the connection usable, so the client falls back to polling on
the same connection without reconnecting.
**Bins were JSON floats.** 1024 bins spelled out as decimal text is around
10 KB a frame, ~200 KB/s at full rate — while the very next hop, client to
browser, already sends the same information as base64 i8 in about 1.4 KB. Bins
now travel base64-encoded whole dBFS, the resolution the display draws at
anyway. Decoding still accepts the old array form.
**Nothing was tunable.** [sdr].spectrum_fft_size and [sdr].spectrum_interval_ms
replace the compile-time FFT size and cadence; [[remotes]].spectrum_interval_ms
lets the client ask for less. 512 bins at 5 frames/s is roughly 3.5 KB/s
against roughly 200 KB/s before.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SyX26FCpMQxiBoC7r5K1A7
Signed-off-by: Stan Grams <sjg@haxx.space>
trx-rs.toml.example was maintained by hand and had fallen well behind: no
[[rigs]], no [[remotes]], no [timeouts], no bandplan or decode-history
settings, and a [frontends.http].default_rig_id that had been renamed.
Generate it from the config structs instead, so a new field shows up the moment
it exists, and add a test that fails when the checked-in copy drifts:
cargo run -p trx-config --example generate_example
Section comments come from a small table; a section without an entry is still
emitted, so forgetting a comment can never drop a setting from the example.
The manual was wrong about the basics. It listed five config search paths, none
of which the loader has ever looked at (the real order is ./trx-rs.toml → XDG →
/etc), called --print-config output "fully commented" when it carries no
comments at all, and documented a TRX_PLUGIN_DIRS variable no code reads. It
also still described [frontends.rigctl].port as the bind port years after
rig_ports replaced it. Fixed, and the new configuration features are written
up alongside.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SyX26FCpMQxiBoC7r5K1A7
Signed-off-by: Stan Grams <sjg@haxx.space>
The row of controls under the plot held four different control heights,
units as loose text beside the field they belonged to, and a quarter of
its width as a hole in the middle. Between about 1100 and 1400 px it
came apart: the bandwidth cluster wrapped to two lines while the level
cluster stayed on one, so the two sat at heights that matched neither
each other nor anything else on the page.
Every control stays, in its order, with its name and its behaviour.
This is the styling and the layout.
A field is now one box -- name, value and unit inside a single border --
so a number cannot be read apart from the unit it is in. Fields,
buttons, the peak-hold select and the contrast slider are one height,
border-box so a button's own border cannot add two pixels to it, and
2.4rem under a coarse pointer where a fingertip needs the room. The
contrast readout holds a fixed, tabular slot, so the row no longer
twitches between 1.0 and 0.9.
The container wraps and a cluster does not: a cluster that will not fit
drops whole to the next line and starts it left-aligned. The slack
goes to a spacer rather than to `space-between`, which is what opened
the hole.
Two things this turned up. The select carries `status-input` for other
layouts' sake, which drew a box inside the field's box. And the narrow
-screen rules lived in a media query earlier in the file than the rules
they override -- identical specificity, so the later one won and the
phone layout had been overflowing sideways rather than stacking. The
narrow rules now sit directly after what they override.
The layout test measures the strip at three widths: one height across
every control, no overflow, inside the plot, and clusters either
sharing a line or each having one -- never one floating against the
middle of the other.
docs/Spectrum-Controls-Rework.md records what was wrong and what was
deliberately left alone: the two different Autos, the settings that do
not persist, the one-shot buttons, and Sweet-spot's silence while it
retunes the SDR. Those are behaviour, and are for another day.
Signed-off-by: Stan Grams <sjg@haxx.space>
A receiver spreads by being linked to, and there was nothing to link to:
the routes carried the tab and nothing else, so "listen to this" could
only ever mean a screenshot and a frequency typed out in a message.
The query string now carries the dial -- rig, frequency, mode and
bandwidth -- in both directions. Opening a link selects the rig, sets
the mode, tunes, then applies the bandwidth: a mode change brings its
own default bandwidth with it, so an explicit bw has to land after it.
Frequencies are read the way someone writes them by hand (7074k,
14.074M) and written back as whole Hz, so what comes out of the address
bar is the same link in canonical form.
After that the address bar keeps up with the dial, which is what makes
it copyable at any moment rather than only at load. It is rewritten
with replaceState -- tuning is not navigation, and a swept dial would
otherwise bury the back button. A link button in the top bar copies
the current link; it folds into the overflow menu when the bar is tight.
Applying a link changes the radio, so an rx session says so instead of
failing control calls one at a time. A tab listening to a virtual
channel leaves the address alone rather than publishing a frequency the
rig is not on, and bw is skipped in both directions on rigs without
filter control, which would only refuse it.
The fixture pinned every state frame to 100 MHz plus jitter to keep
frames distinct, so no test could observe tuning at all. The jitter
moves to the S-meter and the fixture echoes set_freq/set_mode/
set_bandwidth, as it already did for squelch.
Signed-off-by: Stan Grams <sjg@haxx.space>
The bookmark fix addressed one instance of a defect the TypeScript
migration left across the feature entries. app.js stopped being a
classic script, so its top-level declarations are no longer shared
globals, but the converted entries kept reading them as window
properties that nothing publishes.
Restore the broken behavior:
- ais, aprs, hf-aprs read serverLat, serverLon and haversineKm as
undefined, so every positioned packet rendered an empty distance.
- ais, aprs, hf-aprs, cw, sat, vdes, wefax, wspr called an undefined
postPath, so clear-history and decoder toggles threw.
- scheduler read authRole as undefined, so the lazy-load path never
self-initialized and the Settings tab opened an inert scheduler.
- background-decode read authEnabled as undefined, so control gating
fell back to role-only.
- vchan read fifteen application values and services as undefined:
mode and bandwidth sync, the out-of-band hint, RX audio restart, and
the frequency field all silently no-opped on a virtual channel.
- vchan wrapped window.refreshFreqDisplay, capturing an undefined
original exactly as it did for setRigFrequency, so leaving a channel
never restored the application's own frequency display.
- _audioChannelOverride was a const that nothing could assign, so RX
audio always subscribed to the primary channel.
- ftx-family read fmtTime, a helper legacy ft8.js owned locally, so
decode bar timestamps rendered empty.
Declare the contract once in plugins/host.ts and import it from the
feature entries, rather than restoring globals that
docs/frontend-architecture.md excludes. trx.state gains jogUnit,
rxActive and audioChannelOverride, and makes lastModeName writable;
trx.core gains the tuning, RDS, WFM, jog and RX audio services the
entries need. vchan interception moves to an interceptFreqDisplay
service method that refreshFreqDisplay calls, matching the frequency,
mode and bandwidth interception it already registers.
Reading registry-built elements through a strict lookup is the same
defect as in bookmarks: renderTimelineNeedle guards its result, but
schedulerEl throws, so the now-initializing scheduler crashed on the
timeline needle group that its own SVG creates.
Feature tests move onto a shared host fixture, and entries that now
import a common module are bundled through bundleEntry like the other
shared-module entries. Covers scheduler self-initialization and the
distance path that the bare window reads broke.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GdyUjuXejCEfiub675z6cz
Signed-off-by: Stan Grams <sjg@haxx.space>
sjgandClaude Opus 5