Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
-
None
-
None
Description
Given a simple transformapply scenario as shown in the following script
spec = "{ids: false, recode: [ zipcode, district, view ]}"; [X, M] = transformencode(target=F, spec=spec); spec2 = "{ids: false, recode: [ zipcode ]}"; X2 = transformapply(target=F[,1], spec=spec2, meta=M);
currently leads to index out-of-bounds exceptions because the column name zipcode is not found in the column names of the meta data frame. The root cause is a wrong assumption of sorted column names in the underlying implementation.
Attachments
Issue Links
- relates to
-
SYSTEMDS-1824 Wrong transformapply output with col name specification and subset of cols
- Closed