* Avoid a context switch in the history api
The history api was creating a job to fetch the
states and another job to convert the states to
json. This can be done in a single job which
decreases the overhead of the operation.
* Ensure there is only one sqlalchemy session created per history
query.
Most queries created three sqlalchemy sessions which was
especially slow with sqlite since it opens and closes the
database.
In testing the UI is noticeably faster at generating history
graphs for entites.
* Add additional coverage
* pass hass first to _states_to_json and _get_significant_states
* Add include_location_attributes
* Add check for different state
* Fix things
* Drop filter
* significant changes only
* Change default behavior
* Remove trailing commas
* Add a commit interval setting to recorder
* Make the default every 1s instead of immediate
* See attached py-spy flamegraphs for why 1s
* This avoids disk thrashing during event storms
* Make Home Assistant significantly more responsive on busy systems
* remove debug
* Add commit forces for tests that expect commits to be immediate
* Add commit forces for tests that expect commits to be immediate
* make sure _trigger_db_commit is in the right place (all effective "wait_recording_done" calls)
* De-duplicate wait_recording_done code
