Display of daily climatic data - for one year?

[rdstern]

In #3067 there is currently information about a bug in the tabular display of climatic data.

This is a request for a new feature. It is simply to be able to display the table for a single year (or possibly for a set of years).

This may need a checkbox, or perhaps could use the dialogue just as it is. This might need guidance from Danny and David. I am adding Beth to the list, because I think she is also working (with David?) on improvements to the quality control in R-Instat?

There is currently a control labelled Year and this is to include the Year column. We can use the filter (of course) to analyse only a subset of the years. It would be convenient if we could (as an alternative) type in a particular year (possibly a list of years, but OK if just a single year. This could be typed in, or it could be a drop-down, because once you know the station, then you do know the years - though this might involve some tricky code?

The reason is that in our quality control procedures we note dates where the data are odd. It would be good to then be able to look quickly at the data for each odd year - in the nice layout!

[dannyparsons]

We did discuss this before and the decision was not to put it in and just use filters, but I still like the option. You could use this option instead of a global filter, but if you both applied a global filter and typed in years it would apply both "filters" first the global filter, then the selected years. So you may end up with nothing if you think the global filter will be ignored if you type different years. I think as long as users understand how this works its ok.

[rdstern]

In modifying the display daily data there are two other (small?) modifications.

1. There are funny [1] displayed between years. Not a serious problem, but perhaps it could be eliminated.
2. Could there be an option to have no decimals in the table. I found that I recently wanted to look at spell lengths using the display daily. It was fine, but would have been clearer without the decimals. It isn't an option yet in the code, but I think there is just a single place that needs to change. I don't think 2 decimals are needed, so choosing between 0 and 1 is fine. There is a point in the code where it says:

sprintf("%12.1f" and I guess that needs to change. Just below that it does say set to 2 dp, etc, so it may become more complicated. If so, then leave for now. It isn't worth too much time.

[shadrackkibet]

Thanks, I will look into this.