All 10 failures were tests that had drifted behind the product code, not
regressions in it. Diagnosed each individually:
test_mqtt_handler.py:
- test_checkin_creates_new_node, test_checkin_new_node_defaults_lat_lon:
unpacked 4 positional args from doc_set.call_args[0], but
fstore.doc_set(collection, doc_id, data, merge=False) always passes
merge as a kwarg, so only 3 positional args are ever recorded. Fixed
the unpack to 3.
- test_call_start_creates_call_doc, test_call_start_uses_now_when_started_at_missing:
mocked fstore.doc_get, but _on_call_start looks the node up via the
cached fstore.doc_get_cached (added when Firestore reads were cut to
stay in the free tier). The unmocked doc_get_cached returned a bare
MagicMock, which isn't awaitable. Mocked doc_get_cached instead; also
fixed the same 4-vs-3 positional-arg unpack on doc_set's merge=False call.
- test_call_end_updates_status_and_times, test_call_end_sets_audio_url_when_present:
mocked fstore.doc_update, but _on_call_end now writes via
fstore.doc_set(merge=True) (see the "Fix Upload 404 warning" commit —
doc_update raised "No document to update" when call_end arrived before
call_start). Also calls doc_get_cached to stamp org_id. Mocked
doc_get_cached and asserted against doc_set instead of doc_update.
test_node_sweeper.py:
- test_stale_online_node_marked_offline, test_stale_recording_node_marked_offline,
test_tz_naive_last_seen_is_handled, test_only_stale_nodes_updated_in_batch:
_sweep() now calls app.routers.tokens.release_token(node_id) for every
node it marks offline (added in 2a690ec, the PulseAudio/Discord-token
work). These tests never mocked it, so the module-level
patch("asyncio.to_thread", ...) meant for the node-query call leaked
into release_token's own internal to_thread call, feeding it raw node
dicts where it expected Firestore doc snapshots with .id — hence
"AttributeError: 'dict' object has no attribute 'id'". Patched
app.routers.tokens.release_token directly (it's imported inline inside
_sweep, so patching the source module works); the batch test also now
asserts release_token fires for exactly the two nodes that went offline.
No product code changed — app/internal/mqtt_handler.py, app/routers/tokens.py,
and app/internal/node_sweeper.py all behave as intended. This was pure test
drift across two unrelated feature additions (Firestore-read caching,
Discord-token release-on-offline) that landed without their tests being
updated.
93 passed, 0 failed.
Closes logan/server-26#10.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>