dangernoodle-io / breadboard / 29623444175

96%

push

github

feat: add bb_mqtt_client per-instance health snapshot (#914)

- add last_ok_ms/fail_count to each handle's health state (host + espidf)
- fail_count is unconditional on every disconnect (unlike reconnect_count,
  which requires a prior successful connect), so it also captures a failed
  first-connect attempt
- add bb_mqtt_client_health_snap_t + bb_mqtt_client_health_desc
  (bb_serialize descriptor) + bb_mqtt_client_health_fill, mirroring
  bb_tcp_client's per-instance health wire surface
- espidf backend reuses the existing h->lock (already coherent for
  get_stats) rather than adding a new lock; host backend needs none
  (single-writer-per-instance, single-threaded tests)
- bb_mqtt_client_destroy now clears connected without bumping fail_count
  (a caller-initiated close is not a transport failure)
- add deterministic host test hooks (simulate connect success/disconnect
  error/clean close) + mock clock, and cover every new branch


Claude-Session: https://claude.ai/code/session_01VA9XvWjJcbAUGCZpSFfp3H

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

7123 of 7758 branches covered (91.81%)

Branch coverage included in aggregate %.

36 of 36 new or added lines in 2 files covered. (100.0%)

12200 of 12417 relevant lines covered (98.25%)

2335.41 hits per line