Recent rain variables Build 1085

[BCJKiwi]

Have just discovered something unexpected with the recent rain variables

Have a page with a table of recent values http://silveracorn.co.nz/weather/wxcutrends.php (of course by the time you look at this the values may will have changed).
It is showing 58.0 mm rain for 18 and 24 hours ago. 58mm was the total for the day (50.2 the day before!) and its spring!
However $r24hours is zero as the last rain tip was just over 26 hours ago.

It appears the tag <#RecentRainToday h=18> is reflecting what the graph shows - i.e. as the graph continues with the total for the 'day so far' it continues to retain the maximum value for the day until the daily rollover. It seems the tag is doing the same thing which was not expected. Is this the way it is supposed to be?
I have tried using minutes in the tag instead of hours but the result is the same (as expected) but you never know - worth a try .

It seems only rain is affected in this way, which I presume is because only rain maintains a running total for the day.

[steve]

The 'RecentRainToday' web tag contains the value of 'rain today' at any given time. That's what it's supposed to do anyway, if it's working correctly - perhaps it isn't? If that was unexpected, I don't understand what value you did expect it to contain?

[BCJKiwi]

Well I did not expect it to report the existence of rain at a certain time when there was no rain at that time.
It does not seem to be reporting recent but rather daily rain. But then as you point out, the tag does have 'Today' in it.

To me, the table on the page I referred to earlier is nonsensical in the rain column only, as a 'recent' value for x minutes ago has no relationship a particular 24 hour period called a day.

So I have relabelled the column to indicate it is daily rain but I would have expected the recent tags to represent what would be seen on a graph of the same style as, say, the baro graph.

[steve]

Well I did not expect it to report the existence of rain at a certain time when there was no rain at that time.

If it's showing rain on a day where no rain has fallen, that's a bug.

It does not seem to be reporting recent but rather daily rain. But then as you point out, the tag does have 'Today' in it.

As it says in the help, it's the 'rain so far today'. It's the standard 'rain so far today' figure that Cumulus displays and has always displayed. As it's the 'rain so far today', it resets to zero at midnight (or 0900/1000), as at that time there has been no 'rain so far today'. The word 'Recent' in all of these tags indicates that they are the values of the corresponding 'current' values which occurred recently, i.e up to a week ago. e.g., <#RecentOutsideTemp> gives you the outside temperature values for up to a week ago, and <#RecentRainToday> gives you the 'Rain Today' values for up to a week ago.

To me, the table on the page I referred to earlier is nonsensical in the rain column only, as a 'recent' value for x minutes ago has no relationship a particular 24 hour period called a day.

The 24 hour period that the 'rain so far today' figure refers to is the meteorological day, i.e. midnight to midnight for most users, or 0900/1000 to 0900/1000 for those using that setting.

So I have relabelled the column to indicate it is daily rain but I would have expected the recent tags to represent what would be seen on a graph of the same style as, say, the baro graph.

It's the same figure shown on the Cumulus 'Rainfall today' graph, just as <#RecentOutsideTemp>, for example, is the same value shown on the 'Outside temperature' graph.

[BCJKiwi]

Yes Steve, all understood.

It is just that to the casual uninitiated observer, the table suggests it was raining 18 and 24 hours ago when the rain had stopped 26 hours ago.

The post was not intended to indicate that Cumulus was in error (no error mentioned anywhere by me - specially not in the topic title!) but that the table had unexpected / un-anticipated (if that's a word?) values in it.