Desktop: complete the QSO when a partner signs off with RR73

[patrickrb]

Summary

The desktop auto-sequencer never completed (or logged) a QSO when the partner signed off with the standard WSJT-X RR73 — it hung retransmitting our R-report indefinitely. Only the non-grid-shaped 73/RRR sign-offs closed a QSO. This is a functional bug affecting normal FT8 QSO flow (Category 2).

Root cause

classify() in desktop/src-tauri/src/qso.rs checked the message's grid field before matching the RR73/RRR/73 sign-offs:

if !msg.grid.is_empty() { return Content::Grid(msg.grid.clone()); }
// ...sign-off match came after...

"RR73" is a valid 4-char Maidenhead-shaped token (R,R in A..R + two digits), so looks_like_grid("RR73") is true and the decoder copies it into the grid field (dsp/decoder.rs:285). classify() therefore returned Content::Grid("RR73") for a standard sign-off. In process_rx, the TxStage::RReport arm only handles Content::Rr73 | Content::Rrr | Content::Bye73; a Grid fell through to _ => {}, so the engine kept resending our R-report and the QSO never advanced to complete().

Fix

Match the RR73/RRR/73 sign-offs on the trimmed extra text before the grid check in classify(). Minimal reorder, no behavior change for genuine grid reports (a real locator like FN42 still classifies as Content::Grid). looks_like_grid is intentionally left untouched — the map/distance code relies on its shape test and rejects RR73 separately.

Testing

Verified by compiling qso.rs standalone (the full Tauri crate needs webkit/dbus system libs unavailable here) via rustc --test with a zig linker, stubbing only serde and the two dependency structs:

• With the fix: all 8 qso tests pass.
• Against the pre-fix classify: the 2 new tests and the 2 de-masked fixtures fail — confirming they capture the bug.

New tests:

• rr73_signoff_is_not_classified_as_a_grid — direct classify() unit test (RR73→Rr73, FN42→Grid).
• rr73_with_decoder_grid_still_completes_qso — full RReport→complete flow with the realistic grid == "RR73" the live decoder emits.

Also corrected the two existing RR73 fixtures, which passed grid == "" (never happens live) and so masked the bug, to grid == "RR73".

Risk

Low. Single-function reorder in a pure state machine; genuine grid reports and the R-report/report paths are unchanged. Preserves WSJT-X interoperability (RR73 is the standard sign-off).

Notes / follow-up

The same root bug exists on iOS (ios/FT8AFKit/Sources/FT8Engine/QsoEngine.swift classify, decoder FT8Decoder.swift), where its tests likewise pass RR73 fixtures with an empty grid. Left for a separate focused PR (one platform per PR).

🤖 Generated with Claude Code

[codecov]

Codecov Report

❌ Patch coverage is 96.42857% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 29.42%. Comparing base (647b12e) to head (0068c02).

Files with missing lines	Patch %	Lines
desktop/src-tauri/src/qso.rs	96.42%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##                dev     #587      +/-   ##
============================================
+ Coverage     29.36%   29.42%   +0.06%     
  Complexity      197      197              
============================================
  Files           179      179              
  Lines         24068    24091      +23     
  Branches       3263     3263              
============================================
+ Hits           7067     7089      +22     
- Misses        16780    16781       +1     
  Partials        221      221              

Flag	Coverage Δ
desktop	61.02% <96.42%> (+0.18%)	⬆️
native	9.93% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
desktop/src-tauri/src/qso.rs	87.50% <96.42%> (+0.53%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Pull request overview

Fixes a desktop auto-sequencer state-machine bug where a partner’s standard FT8 RR73 sign-off was misclassified as a grid (due to RR73 matching the Maidenhead “grid-like” shape), preventing the QSO from ever reaching completion/logging.

Changes:

• Reorders classify() so RR73/RRR/73 sign-offs are matched before treating msg.grid as a grid report.
• Updates existing RR73-related fixtures to reflect live decoder behavior (grid == "RR73").
• Adds regression tests to ensure RR73 is not classified as Content::Grid and that the QSO completes from TxStage::RReport with decoder-realistic inputs.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.