How to generate new columns from points command?

[petercreateParticipant]

I met a new issue that is not able to extract data points from point plot command. The case is the data points with two columns are not complete but with some blanks in either of the columns and they have the third timestamp column. I wanted to tidy them with only valid data rows and so think of ‘from command’. It didn’t work on my DG. Do you have a workaround?

[dgteamModerator]

Here is an example removing blanks in the data table where there is a unique date column.

Here is the data. The dates are in every column but the data has blanks.

Here are the setting for each column that is not input data.

The first is a mask column, the second is an expression column that is sorting the dates, the third column is a mapped column that pulls the data for the filtered date list.

Not sure if this is quite what you are looking for, but perhaps this will help!

[petercreateParticipant]

Thanks. I want to get pairs of AB and AC companied with date_time values. Please see the shot.

date_time A B C
22/08/2018 10:00 1 2
22/08/2018 10:30 2 7
22/08/2018 13:30 3 6 7
23/08/2018 10:00 4 9
23/08/2018 10:30 8
23/08/2018 13:30 5 9 8
24/08/2018 10:00 6 10
24/08/2018 10:30 7 1 8
24/08/2018 13:30 8 3
25/08/2018 10:00 8 8
25/08/2018 10:30 8 8
25/08/2018 13:30 8 5 8

 

[dgteamModerator]

It sounds like you want to have a table that only shows where A & B exist in another table that only shows where A & C exist.

If that is the case, here’s one approach. Use an expression to generate a column of ones, for rows where AB or AC are both there. You could use an if statement for this, but a simple math expression can also do it. When there are missing values by default, nothing gets returned.

Once you have that, you can sort with the column of ones.

Note that we placed the filtered date_time column and its own group. Then you can use the same name “date_time”.

This makes creating the subsequent map columns easy because all you need to do is right-click on the new filtered date_time column, the program will search for other columns with the same name, and you will have the option to map to the original data.

Does this help?

 

[Author]

Posts