Export data action node: Grouping data on multiple properties, or a runtime variable

Hi
I have a datasource consisting of entries of “salarytype” and “hours” for employees. e.g. the same salarytype and employee exists multiple times in the dataset. To give an idea:

Salarylines
employee salarytype hours. date
erik 1929 3,5 18.4.2025
erik 1938 5 18.4.2025
jonas 2991 6,6 19.4.2025
jonas 2991 6,8. 20.4.2025
per 2991 7 20.4.2025

I want to export this to a csv where each combination of employee and salarytype occurs ONCE with the hours summed up. (so the example above would result in 4 lines, grouping line 3 and 4) I have tried to do this with the Export Data action node but I haven´t managed to get there yet… Now I wonder if its even possible without creating a “helper” class which I want to avoid if possible.

My first approach was to just generate a runtime property “employee+salarytype” and then do the row grouping on that. Then I could just add the hours as a static column and use sum. However Im not allowed to sort on anything else than Object classes, so this does not seem to work, since grouping on employee or salary type alone obviously won´t work since the same employee and salary type will be present on multiple rows.

Then I have tried a combination of using data columns, but that seems to generate one column per type of employee/Salarytype which is not what I want.

I could also solve this with a table and do csv export from there, but the same problem here, Im unable to “group” on a runtime property or multiple properties (employee+salary type) for the table, so I get multiple occurences of the same employee+salarytype.

I suppose I could generate a new object class that aggregates the hours into one entry for each combination of employee&salary type, but since this scenario is quite simple I would expect that it was a simple way of doing this… is it?

Appreciate any good ideas.

Hi!

I don’t think you can avoid using a “helper” data source.
My suggestion: use the Aggregate Data action node to group by employee and salarytype, aggregating the sum of hours. Then you can also use the same object class to create the runtime-only target data source of the aggregation. Finally, you export the data from the target data source.

Hope that helps!

1 Like

That did the trick.

I discovered another problem. When download CSV from the table or using the export data action node, I get the headers on the first line in the file. Im unable to find a config to NOT export the header… Is there a trick to do this? Tried to “hide header” in the table before exporting but that did not help…

If I remove the header from alle the columns in the export data action node, it still generates a line with delimiterts

,,,,,,
67af0a809dbbc2b89db1db25,6582b1a84d6fea2d65a9f40e,,47.39,677d2077e7c37fd45ab1b54d,2,2025

Thanks

I found a solution. However it should be possible to avoid getting the header line.

I have registered a feature request to have a setting on Export Data that skips all headers.

1 Like