ds.dataFrameSubset keep.cols / rm.cols parameter should accept character vectors?

[tombisho]

Reviewing the ds-helper package, a function is provided to return the index of a column in a dataframe, because ds.dataFrameSubset requires this for the keep.cols/rm.cols parameter.

It would be easier if ds.dataFrameSubset would accept the column name, because then this helper function would not be needed?

[davraam]

Yes we should modify the ds.dataFrameSubset to accept column names. This is important when the order of the columns is not the same across the different studies.

[tombisho]

That's a good point, I hadn't thought of that!

[tombisho]

Just to revisit this issue, one of our analysts is saying that the removal of ds.subset (where you can use column names) will cause a major issue because the columns are in different orders in each study. So I think ds.subset cannot be removed until this feature is implemented.