You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
DataFrame merge results in column datatype problem #6127
Alright, the issue is that merge is calling "Clone" on the columns, but Clone returns slightly different types. For example, Instead of returning DoubleDataFrameColumn it is returning PrimitiveDataFrameColumn<double>. DoubleDataFrameColumn does extend PrimitiveDataFrameColumn<double> but they aren't the same type obviously. The problem then is that in the call to GetDoubleColumn, the check if (column is DoubleDataFrameColumn ret) fails because its not actually a DoubleDataFrameColumn anymore.
I not sure the exact best way to fix this from a code standpoint. Probably needs a little bit more investigation to figure out the best way. @luisquintanilla for visiblity.
Activity
michaelgsharp commentedon Mar 18, 2022
Alright, the issue is that merge is calling "Clone" on the columns, but Clone returns slightly different types. For example, Instead of returning
DoubleDataFrameColumnit is returningPrimitiveDataFrameColumn<double>.DoubleDataFrameColumndoes extendPrimitiveDataFrameColumn<double>but they aren't the same type obviously. The problem then is that in the call toGetDoubleColumn, the checkif (column is DoubleDataFrameColumn ret)fails because its not actually aDoubleDataFrameColumnanymore.I not sure the exact best way to fix this from a code standpoint. Probably needs a little bit more investigation to figure out the best way. @luisquintanilla for visiblity.