The shared audio queue (maxsize reduced from 80 to 20) was not flushed
when the monitor frequency changed — only when the monitor was disabled.
This caused up to 4 seconds of stale old-frequency audio to play after
clicking to tune, making click-to-tune appear non-functional.
Now flushes the queue whenever the VFO frequency changes, so audio at
the new frequency begins within ~50ms (one FFT frame).
Two root causes for the waterfall/monitor lockup when scrolling past the
2.4 MHz RTL-SDR span:
1. safe_terminate() sent SIGKILL but never called wait(), leaving a
zombie process that kept the USB device handle open. The subsequent
capture restart failed the USB probe and the monitor could not use
the shared IQ path, falling back to a process-based monitor that
stole the SDR from the waterfall.
2. When the frontend created a new WebSocket after a failure, the old
handler's finally block called _set_shared_capture_state(running=False)
which could race with the new handler's running=True, making the
shared monitor path unavailable. Added a generation counter so only
the owning handler can clear the shared state.
When restarting capture for a new frequency, the USB handle from the
just-killed process may not be released by the kernel in time for the
rtl_test probe inside claim_sdr_device. Add retry logic (up to 4
attempts with 0.4s backoff) matching the pattern already used by the
audio start endpoint.
Also clean up stale shared-monitor state in the frontend error handler
so the monitor button is not left disabled when the capture restart
fails.
- _waitForPlayback now only succeeds on playing/timeupdate events, not
loadeddata/canplay which fire from just the WAV header before real
audio arrives
- stopMonitor() pauses audio and updates UI immediately instead of
blocking on the backend stop request (1+ second delay)
- Reduced backend audio stop sleep from 1.0s to 0.15s; the start
retry loop already handles USB contention
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When dashboards (satellite, ADS-B, AIS) are loaded via iframe with
?embedded=true, the full navigation bar was still rendered, creating
a "UI in UI" effect. Pass the embedded query param from route handlers
to templates and conditionally skip the nav include.
Fixes#144
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The previous fix (f29ae3d) introduced a regression: when VIS parity
check failed or the VIS code was unrecognized, the detector entered
DETECTED state permanently and never resumed scanning. Now it resets
to IDLE on validation failure and only enters DETECTED on success.
Also resets partial image progress counter between consecutive decodes
and adds SDR device claiming to general SSTV route to prevent conflicts.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New mode providing real-time space weather data from NOAA SWPC, NASA SDO,
and HamQSL APIs. Includes Kp index, solar wind, X-ray flux charts, HF band
conditions, D-RAP absorption maps, aurora forecast, solar imagery, flare
probability, and active solar regions. No SDR hardware required.
Bumps version to 2.20.0. Updates all documentation including README, FEATURES,
USAGE, UI_GUIDE, help modal, and GitHub Pages site.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update default ACARS frequencies for North America to 131.725/131.825 MHz and add ISS (145.825 MHz) as a selectable APRS frequency region.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The analytics summary, health, and export were only reading from legacy
DataStores (app_module.wifi_networks, bt_devices) which the v2 WiFi and
Bluetooth scanners don't populate. Now checks v2 scanner singletons
first and falls back to legacy stores.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a unified analytics mode under the Security nav group that aggregates
data across all signal modes. Includes emergency squawk alerting (7700/7600/7500),
vertical rate anomaly detection, ACARS/VDL2-to-ADS-B flight correlation,
geofence zones with enter/exit detection for aircraft/vessels/APRS stations,
temporal pattern detection, RSSI history tracking, Meshtastic topology mapping,
and JSON/CSV data export.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
dumpvdl2 2.6.0 requires the complete output specifier format
'decoded:json:file:path=-' instead of just 'decoded:json'.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add VDL2 (VHF Digital Link Mode 2) decoding via dumpvdl2 as a new mode,
and promote ACARS from ADS-B-dashboard-only to a first-class standalone
mode in the main SPA. Both aviation datalink modes now have full nav
entries, sidebar partials with region-based frequency selectors, and
SSE streaming. VDL2 also integrated into the ADS-B dashboard as a
collapsible sidebar alongside ACARS.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Bluetooth enhancements (service data inspector, appearance codes, MAC
cluster tracking, behavioral flags, IRK badges, distance estimation),
ACARS SoapySDR multi-backend support, dump1090 stale process cleanup,
GPS error state, and proximity radar/signal card UI improvements.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New modes:
- BT Locate: SAR Bluetooth device location with GPS-tagged signal trail,
RSSI-based proximity bands, audio alerts, and IRK auto-extraction from
paired devices (macOS plist / Linux BlueZ)
- GPS: Real-time position tracking with live map, speed, heading, altitude,
satellite info, and track recording via gpsd
Bug fixes:
- Fix ABBA deadlock between session lock and aggregator lock in BT Locate
- Fix bleak scan lifecycle tracking in BluetoothScanner (is_scanning property
now cross-checks backend state)
- Fix map tile persistence when switching modes
- Use 15s max_age window for fresh detections in BT Locate poll loop
Documentation:
- Update README, FEATURES.md, USAGE.md, and GitHub Pages with new modes
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Satellites added via CelesTrak import or TLE paste are now stored in
SQLite and survive page reloads and app restarts. Adds CRUD API
endpoints and wires frontend sidebar + dashboard to use them.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add .gitignore entry for data/subghz/captures/ to prevent large
IQ recording files from being committed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Throttle audio waterfall rendering (50ms→200ms), eliminate per-frame
Array.from() allocation, drain stale pipe buffer before streaming,
increase chunk size to 8192, and remove debug logging from animation
hot paths.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Starting ffmpeg at decoder launch caused a pipe-buffer deadlock: ffmpeg
stdout filled up (~64KB on Linux) before the browser connected to the
audio stream, back-pressuring the entire pipeline and freezing dsd-fme
stderr output (no text data, no syncs, no calls).
New architecture: a mux thread always drains dsd-fme stdout to keep the
pipeline flowing. ffmpeg starts lazily per-client when /dmr/audio/stream
is requested (matching the listening post pattern). The mux forwards
decoded audio to the active ffmpeg with silence fill during voice gaps,
and discards audio when no client is connected.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Digital voice is intermittent — dsd-fme only outputs PCM during active
voice transmissions. Without input, ffmpeg never wrote the WAV header
and the browser got an empty response. Add an audio bridge thread that
feeds 100ms silence chunks during voice gaps so ffmpeg always has input
and the browser receives a continuous WAV stream. Add auto-reconnect
on the frontend if the audio stream drops while the decoder is running.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Stream decoded digital voice audio to the browser via ffmpeg pipeline
(dsd-fme 8kHz PCM → ffmpeg → 44.1kHz WAV → chunked HTTP). Persist
frequency/protocol/gain/ppm settings in localStorage so they survive
page navigation. Add bookmark system for saving and recalling frequencies.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Wait for server-side WebSocket stop confirmation before closing the
connection, ensuring the IQ process is fully terminated and the USB
device released. Add retry logic with back-off in the audio start
endpoint as defense-in-depth for any remaining timing gaps.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The box-drawing character filter was dropping ANY line containing │ or ─,
including dsd-fme data lines that use these as column separators (e.g.
"DMR BS │ Slot 1 │ TG: 12345 │ SRC: 67890"). Now only filters lines
that are purely decorative (no alphanumeric content).
Also adds -J /dev/stderr so dsd-fme writes its event log to stderr
where we capture it, and debug logging of raw stderr lines.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
-fd means D-STAR in dsd-fme, not DMR — causing sync detection
(shared C4FM modulation) but no decoded data. DMR Simplex is -fs.
Also fix -o - (invalid in dsd-fme) to -o null for headless servers,
add D-STAR flag mapping, and handle TGT/SRC output format in parser.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
