The map module was tagging all decode markers (APRS, AIS, VDES,
FT8/FT4/FT2/WSPR locators) with the global rig picker's active rig
instead of the actual source rig. This made the map's own rig filter
dropdown ineffective in multi-rig setups.
- Add rig_id field to all decode message structs (AisMessage,
VdesMessage, AprsPacket, CwEvent, Ft8Message, WsprMessage)
- Set rig_id on messages in audio_client before broadcasting, using
the actual rig connection identifier
- Update history collector to prefer message rig_id over the global
active rig fallback
- Pass rig_id through plugin normalize functions (AIS, APRS, VDES,
HF-APRS) so it reaches the map add functions
- Update all map marker functions (aprsMapAddStation, aisMapAddVessel,
vdesMapAddPoint, mapAddLocator) to use the message's rig_id with
fallback to the global picker for backward compatibility
https://claude.ai/code/session_015gC7axHk2jmp7HbFPdbivN
Signed-off-by: Claude <noreply@anthropic.com>
Drop the plugin loading infrastructure (libloading-based dynamic .so/.dylib/.dll
loading) from both trx-server and trx-client. The feature was unused and posed an
unnecessary security risk by executing arbitrary native code from disk.
Removed:
- src/trx-app/src/plugins.rs (plugin discovery, validation, FFI registration)
- examples/trx-plugin-example/ (cdylib example plugin)
- libloading dependency from trx-app
- load_backend_plugins / load_frontend_plugins calls from server and client
- Plugin documentation from README.md and CLAUDE.md
https://claude.ai/code/session_01DTEUpz3XPUeWmz74NeaFgb
Signed-off-by: Claude <noreply@anthropic.com>
Replace all .unwrap() on RwLock/Mutex acquisitions with
.unwrap_or_else(|e| e.into_inner()) to gracefully recover from poisoned
locks instead of panicking. Add lock ordering documentation to the
module header to prevent deadlocks.
https://claude.ai/code/session_01XzurkeuUmamBuhQwxVy7T4
Signed-off-by: Claude <noreply@anthropic.com>
Use a StateWithMeta wrapper struct with #[serde(flatten)] for merging
rig state with frontend meta, replacing the manual string manipulation.
Also add Serialize derive and skip_serializing_if to FrontendMeta.
https://claude.ai/code/session_01XzurkeuUmamBuhQwxVy7T4
Signed-off-by: Claude <noreply@anthropic.com>
Introduce DecodeHistory<T> alias for the repeated
Arc<Mutex<VecDeque<(Instant, Option<String>, T)>>> pattern (9 fields).
Also switch VChanAudioCmd channel senders from UnboundedSender to Sender
to prevent unbounded memory growth under backpressure.
https://claude.ai/code/session_01XzurkeuUmamBuhQwxVy7T4
Signed-off-by: Claude <noreply@anthropic.com>
Decoder bar overlays (AIS, VDES, FT8, APRS, RDS) use backdrop-filter
blur for a frosted-glass look in the browser, but this can't be
replicated on canvas — resulting in opaque blocks covering the spectrum
in screenshots. Cap their background alpha to 0.35 when rendering to
the snapshot canvas.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Press F1 to toggle a help overlay listing available keyboard shortcuts.
Dismiss with F1, Escape, or clicking the backdrop. Refactored the
global keydown handler to route all shortcuts through one listener.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Order bookmark mode and bandwidth updates so WFM bookmarks do\nnot race against the backend mode default.\n\nAlso apply saved bookmark bandwidth in the scheduler path so\nscheduled bookmark replays keep the configured filter width.\n\nTested with:\n- cargo test -p trx-frontend-http\n\nCo-authored-by: OpenAI Codex <codex@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Radio paths now originate from the rig that decoded the message rather
than the currently selected rig. Bookmark locators no longer draw radio
paths. Rig switch no longer tears down decode pipeline since it is
rig-independent. Mobile spectrum controls use flex-wrap for better layout.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add latitude/longitude to /rigs API response. Map now displays receiver
markers for all configured rigs, de-duplicated by location. Decode history
is no longer filtered to the active rig so all remotes contribute to the map.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use the merged bookmarks endpoint instead of separate fetches so general
bookmarks are always visible alongside rig-specific ones.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
WebCodecs AudioDecoder does not support Opus on Safari. Fall back to
opus-decoder WASM library (loaded from CDN) for browsers without
WebCodecs Opus support.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Scope picker filters the bookmarks table for editing. Spectrum and map
always show merged general + active rig bookmarks via bmOverlayList.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Fix clippy warning by adding Default impl for BookmarkStoreMap. Scheduler
bookmark picker now fetches both general and rig-specific bookmarks and
merges them so all available bookmarks are shown.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Two-tier bookmark system: general bookmarks shared across all rigs plus
rig-specific bookmarks with scope picker in the Bookmarks tab. Scheduler
storage split into per-rig files with migration from legacy single file.
Decode history tagged with rig_id and filterable via ?remote= on
/decode/history endpoint. Decode SSE reconnects on rig switch to refresh
filtered history.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
watch::Sender::send() silently discards the value when receiver_count
is zero. The per-rig info channel is created with watch::channel().0
which drops the only Receiver, so the first send(Some(info)) after TCP
connect returns early without storing the value. Later subscribers
see None forever. send_replace() stores unconditionally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use only the per-rig stream info when a remote is specified on the
channel audio path; do not fall back to the global channel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
The audio client was clearing per_rig_info_tx to None every time the
TCP connection dropped, even during transient reconnect cycles. This
caused WebSocket clients subscribing to per-rig audio to stall
indefinitely waiting for stream info that would only arrive after the
next successful reconnect.
Move the None send to the permanent shutdown path only. The last-known
stream info remains valid for the same rig across reconnects.
Also revert the global info_rx fallback from 6e4c5e3 since the root
cause is now fixed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Per-rig info_rx watch channel is transiently None when the rig's audio
TCP connection is between reconnect cycles. This caused the WebSocket
handler to hang indefinitely waiting for stream info that might never
arrive, regressed in 7d76606.
Prefer the per-rig info_rx when it holds a value, otherwise fall back to
the global info channel (the pre-regression behaviour).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Trace per-rig audio subscription lookup, stream info availability, and
session lifecycle to diagnose multi-rig audio regression.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Send the selected remote together with virtual-channel audio requests and use per-rig stream info when /audio is opened with channel_id.
This keeps browser channel audio aligned with the requested remote after the recent multi-rig client changes.
Add coverage for parsing channel_id together with remote.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Create and resume the RX AudioContext from the audio button click so Chromium does not leave playback suspended until a later interaction.
Reuse that context when stream metadata arrives instead of recreating it from the WebSocket message path.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Remove the remaining legacy rig_id aliases across the HTTP frontend and use remote consistently for scheduler and audio requests.
Disable browser caching for the HTML, CSS, and JS assets so clients pick up the parameter rename immediately instead of running stale frontend code.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Parse the renamed `remote` query parameter on `/audio` while keeping
`rig_id` as a compatibility alias, and use per-rig stream info for
rig-scoped audio subscriptions.
Add tests covering both query names.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Keep global state and spectrum updates scoped to the server group that
owns the selected short name, so other servers cannot overwrite the UI
or clear the active spectrum stream.
Add regression tests for cross-server selection and spectrum ownership.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Merge per-server GetRigs updates into the shared remote rig cache
instead of replacing it, so audio tasks from other servers are not
torn down on each poll cycle.
Add a regression test covering the multi-server cache update path.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use full audio endpoint URLs for trx-server audio connections while
preserving server-advertised ports and legacy port-based fallback for
backward compatibility.
Add `server_url` and per-remote `rig_urls` config entries, plus
validation and tests for audio URL parsing and address resolution.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Rename HTTP query params, JSON fields, and scheduler payloads to
use remote names consistently while still accepting legacy `rig_id`
inputs through serde aliases.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Use the remote short name for the HTTP frontend default selection and
keep `default_rig_id` as a serde alias for backward compatibility.
Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Introduce [[remotes]] config array where each entry maps a user-chosen
short name to a (server URL, rig_id) pair. Short names replace raw
rig_ids as the universal key throughout frontends, audio routing, and
state management, allowing rig_ids to safely collide across servers.
Entries sharing the same server URL and token share a single TCP
connection. A request routing dispatcher forwards frontend commands to
the correct per-server channel based on the short name.
Legacy [remote] config and CLI --url are preserved via automatic
fallback to a single-entry remotes list.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Include active rig display name and nearest city in the browser tab
title: "freq - rig name - city, country - trx-rs v<version>".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
Add checkbox to enable/disable NB and number input for threshold (1-100).
Controls are hidden by default and shown when the server reports NB support
via SSE filter state updates.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Stan Grams <sjg@haxx.space>
sjg
Claude