maziggy 5c673620f3 fix(notifications): false-positive Print Stopped on reprint after MQTT reconnect (#1807) ...

Reprints triggered a bogus "Print Stopped" push notification while the print
  kept running, surfaced by the reconciler synthesising a missed PRINT COMPLETE
  on MQTT reconnect.

  bambu_mqtt:3647 mints a fresh subtask_id per dispatch. On reprint, the
  on_print_start expected-archive promotion only wrote subtask_id when the
  stored value was empty (`not archive.subtask_id`) — so the archive kept the
  FIRST run's id. On the next MQTT reconnect, reconcile_stale_active_prints
  (#1542) compared the stale stored id against the printer's live id, found
  a mismatch, and synthesised a status="aborted" PRINT COMPLETE — which fires
  the "Print Stopped" notification.

  Captured cleanly in the reporter's support bundle:

    [RECONCILE] Printer 1: synthesising missed PRINT COMPLETE for archive 31
      — subtask_id changed ('1844213296' → '2103771517')

  immediately followed by gcode_state: RUNNING on the same wire.

  Fix: update archive.subtask_id whenever the new effective id differs from
  the stored one, not only when the stored one is empty. Inequality check
  preserves the noop-on-stable-push behaviour the original guard provided.

  Two places in main.py (expected-print and duplicate-printing-archive
  branches). 3 new unit tests cover the reprint, first-run, and stable-push
  paths. Reconciler itself unchanged — it was doing the right thing given
  the data it had.

2026-06-25 11:07:29 +02:00

..

app

fix(notifications): false-positive Print Stopped on reprint after MQTT reconnect (#1807)

2026-06-25 11:07:29 +02:00

tests

fix(notifications): false-positive Print Stopped on reprint after MQTT reconnect (#1807)

2026-06-25 11:07:29 +02:00

__init__.py

…