Matt,
How would you prefer bugs be written up? Just posts here, or would it work 
better to create an issue in github?

I noticed the wrap-around portion of _delta_rain() could lead to lost data. 
When a wrap-around occurs (new_rain < last_rain), you return None. But it's 
entirely possible that we could've actually received rain in that interval, 
and we will just be throwing it away... ie at 11:59pm it is raining and 
we've received something like .15" of rain. We happen to be experiencing a 
downpour at that time, and the sensor has logged another .04" of rain. 
Since we go into a new day, the bridges internal counter resets to 0, and 
then reports .03" on the next packet. The interceptor driver will throw the 
.03" away, but then resume operating normally.

I would suggest that when we detect a wrap-around, we substitute 0.0 in 
place of last_rain, and report the delta based on that.

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to