# Story 3.1: Stale State Detection & Dimmed Display Status: done ## Story As a user whose RTL-SDR has temporarily lost signal, I want the display to retain the last known aircraft positions shown as outlines when dump1090 stops delivering fresh data, So that I know the display is stale without a crash or blank screen. ## Acceptance Criteria AC1: **Given** the radar loop is running with a previous successful fetch **When** `HttpFetcher.fetch()` raises `requests.Timeout` **Then** the exception propagates to the loop boundary, which catches it and marks all retained aircraft as `is_stale=True` AC2: **Given** the dump1090 response returns an empty aircraft list when the previous cycle had aircraft **When** the fetcher processes the response **Then** the previous aircraft list is retained with `is_stale=True` on each entry (not replaced with empty) AC3: **Given** aircraft with `is_stale=True` are passed to the renderer **When** `renderer.render()` is called **Then** each stale aircraft is drawn as outline only using `COLOUR_STALE_OUTLINE` **And** heading arrow, label, and trail are still rendered at last known positions AC4: **Given** a stale render cycle **When** the render loop timing is measured **Then** the loop does not crash (stale path is not a crash path) ## Tasks / Subtasks - [x] Task 1: Update `_run_one_cycle()` and `main()` in `src/planemapper/main.py` (AC: #1, #2, #4) - [x] 1.1 Add `last_aircraft: list[Aircraft]` parameter to `_run_one_cycle`; change return type to `list[Aircraft]` - [x] 1.2 Wrap `fetcher.fetch()` in `try/except requests.Timeout` - [x] 1.3 Implement stale logic: use `last_aircraft` with `is_stale=True` when timeout or empty+had-previous - [x] 1.4 Update `main()` to initialise `last: list[Aircraft] = []` and track return value of `_run_one_cycle` - [x] 1.5 Add `import dataclasses` and `import requests` to `main.py` - [x] Task 2: Update `draw_aircraft()` in `src/planemapper/renderer/aircraft.py` (AC: #3) - [x] 2.1 Import `COLOUR_STALE_OUTLINE` from `planemapper.constants` - [x] 2.2 Check `aircraft.is_stale` before calling `_draw_arrow`: if stale, use `COLOUR_STALE_OUTLINE` with forced outline mode - [x] Task 3: Write tests in `tests/test_stale.py` (AC: #1, #2, #3, #4) - [x] 3.1 Test AC1: mock `fetcher.fetch()` to raise `requests.Timeout`; call `_run_one_cycle` with `last_aircraft=[some_aircraft]`; assert returned list has `is_stale=True` - [x] 3.2 Test AC2: mock `fetcher.fetch()` to return `[]`; call with non-empty `last_aircraft`; assert returned list has `is_stale=True` - [x] 3.3 Test AC3: render stale aircraft; assert pixel at aircraft position is not the altitude colour (stale colour is `COLOUR_STALE_OUTLINE` = black) - [x] 3.4 Test AC4: full stale cycle with renderer completes without exception - [x] Task 4: Run quality gates - [x] 4.1 `python -m pytest tests/` — all tests pass - [x] 4.2 `python -m ruff check .` — zero violations - [x] 4.3 `python -m ruff format --check .` — no formatting issues ## Implementation Notes ### Key insight: `is_stale` already exists The `is_stale` field already exists on the `Aircraft` dataclass (defaults to `False`). No model changes are needed. The `is_mlat` field already drives outline rendering in `_draw_arrow` — stale aircraft reuse the same outline path but with a different colour. ### Stale vs MLAT rendering distinction | Condition | Fill | Outline colour | Meaning | |---|---|---|---| | Normal | altitude colour | — | Direct ADS-B, current data | | MLAT (`is_mlat=True`) | None | altitude colour | Uncertain position, current data | | Stale (`is_stale=True`) | None | `COLOUR_STALE_OUTLINE` (black) | Last known position, data age unknown | Stale takes priority over MLAT in the rendering check: if `aircraft.is_stale` is true, always render outline in `COLOUR_STALE_OUTLINE` regardless of `is_mlat`. ### `_run_one_cycle` signature change ```python def _run_one_cycle( renderer: Renderer, fetcher: HttpFetcher, display: DisplayInterface, last_aircraft: list[Aircraft], ) -> list[Aircraft]: ``` Returns the aircraft list used for this cycle (caller passes it back as `last_aircraft` next cycle). ### Stale detection logic in `_run_one_cycle` ```python try: fresh = fetcher.fetch() except requests.Timeout: log.warning("fetch timeout — using stale data") fresh = [] stale_needed = True else: stale_needed = (len(fresh) == 0 and len(last_aircraft) > 0) if stale_needed: aircraft_list = [dataclasses.replace(a, is_stale=True) for a in last_aircraft] else: aircraft_list = fresh ``` `dataclasses.replace()` creates a new `Aircraft` instance with only `is_stale` changed — the original `last_aircraft` entries are not mutated. ### `main()` loop update ```python last: list[Aircraft] = [] while True: last = _run_one_cycle(renderer, fetcher, display, last) time.sleep(REFRESH_INTERVAL_S) ``` Initialise `last` as empty list before the loop. On first cycle, `last_aircraft=[]` means a timeout or empty result produces an empty stale list (no aircraft to retain), which renders a clean empty map — correct behaviour. ### `draw_aircraft()` stale check In `src/planemapper/renderer/aircraft.py`, inside `draw_aircraft()`, before calling `_draw_arrow`: ```python if aircraft.is_stale: _draw_arrow(draw, cx, cy, aircraft.heading, COLOUR_STALE_OUTLINE, is_mlat=True) else: _draw_arrow(draw, cx, cy, aircraft.heading, colour, aircraft.is_mlat) ``` The `is_mlat=True` argument reuses the existing outline-only code path in `_draw_arrow`. The label and trail draw unconditionally at the last known positions — stale state does not suppress them. ### Required imports in `main.py` - `import dataclasses` — for `dataclasses.replace()` - `import requests` — to name `requests.Timeout` in the `except` clause Both should be added to the top-level imports alongside the existing ones. ### Files changed | File | Change | |---|---| | `src/planemapper/main.py` | Add params/return to `_run_one_cycle`; stale detection logic; update `main()` loop; add imports | | `src/planemapper/renderer/aircraft.py` | Import `COLOUR_STALE_OUTLINE`; stale check before `_draw_arrow` | | `tests/test_stale.py` | New test module covering all four ACs |