I seem to have exidentally search and replaced some
needed glob.glob calls when adding dinamic log location,
which should now be repaired.
There also was a mysterious cbpi_dashboard_1.json appearing
in the wrong folder which now inside the dashboard folder.
I also figured out how to run tests locally,
they should now respect the dynamic folder paths.
because cbpi already writes timestamps into the logs
and we dont need two timestamps when exporting.
This commit also reverts to using the same formatter for all logHandlers.
It was wherever 'cbpi setup' was run from.
and if no config folder path is provided its still the same.
This also adds a file based global logger placed in the logs folder.
The sensor loggers respect the new logs folder location.
previously the whirlpool step has been ignored when importing from a kbh database.
on creation the whirlpoolstep would only be on a hardcoded 15 minute timer.
I added the optional timer as argument to create_Whirlpool_Cooldown().
And made use of it for negative hops timer imported from KBH.
For KBH databases this is valid because if the hops timer value is negative it can only be the whirlpool timer
and the value is the time that is set for the brewing device for getting from boil to below 80°C.
I did only test the data and not the functionality,
but it sucessfully replaces the hardcoded "15" timer with the value
from the first whirlpool hop addition while getting rid of the sign.
if accumulated size of log file data exceeds some megabytes. Also chart refresh rate should be greater than some 10 seconds so that chart plots can not block the user interface.
In functions get_data (for dashboard charts) and get_data2 (for chart in analytics) the pandas resample function is reactivated respectively added to further reduce the amount of data to be transferred to the clients.
In function log_data max_bytes and backup_count must be converted to int, when these settings are changed in the settings dialog.