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
In my use case I can work around the issue because my data is already sorted by the groups. If it wasn't and I had one column that wasn't nullable in with others that were I would then have to write a conditional path that would sort the data myself ahead of time and pull out the columns that don't need null replacement and hope that replace_nulls would not change the order of the data. Would be nice if replace_nulls could just copy the column if needed.
The text was updated successfully, but these errors were encountered:
Can you clarify the issue, please? I dig into the replace_null code and see that nullable columns are already handled thus I'm not sure what's wrong here? As we can see from the code below, all columns are always sorted regardless of their nullability. If a column is not nullable, it will just be copied (after sorted).
harrism
changed the title
[BUG] group_by.replace_nulls fails if the collumn is not nullable
[BUG] group_by.replace_nulls fails if the column is not nullable
Jun 15, 2021
I put this as a bug because it is unexpected.
In my use case I can work around the issue because my data is already sorted by the groups. If it wasn't and I had one column that wasn't nullable in with others that were I would then have to write a conditional path that would sort the data myself ahead of time and pull out the columns that don't need null replacement and hope that
replace_nulls
would not change the order of the data. Would be nice ifreplace_nulls
could just copy the column if needed.The text was updated successfully, but these errors were encountered: