Miscellaneous fixes (#23279)

* use monotonic clock for detector inference duration to prevent negative values from wall clock steps

* add ability to set camera's webui_url from camera management pane

* Gemini send thought signature

* Update docs

* copy face and lpr configs from source camera to replay camera

* add guard

* improve dummy camera docs

* remove version number

* fix stale field message after reverting a conditional form field

Routes field-level conditional messages through a dedicated React Context instead of merging them into uiSchema. RJSF's Form keeps state.uiSchema sticky across renders during processPendingChange (formData is updated, uiSchema is not), so a previously injected ui:messages array stays attached to a field even after the triggering condition flips back to false. Context propagation re-runs FieldTemplate directly on every provider value change, sidestepping that staleness.

* add semantic search field message to note that model_size is irrelevant when embeddings provider is selected

---------

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
This commit is contained in:
Josh Hawkins
2026-05-22 07:52:01 -06:00
committed by GitHub
co-authored by Nicolas Mowen
parent a4a592b4e6
commit 0bdf5002a0
14 changed files with 384 additions and 158 deletions
+2 -1
View File
@@ -167,8 +167,9 @@ class DetectorRunner(FrigateProcess):
# detect and send the output
self.start_time.value = datetime.datetime.now().timestamp()
mono_start = time.monotonic()
detections = object_detector.detect_raw(input_frame)
duration = datetime.datetime.now().timestamp() - self.start_time.value
duration = time.monotonic() - mono_start
frame_manager.close(connection_id)
if connection_id not in self.outputs: