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
Is your feature request related to a problem or challenge? Please describe what you are trying to do.
I'm trying to use the arrow row format in DataFusion. For that it would be helpful to know early (read: before piping any actual payload through physical execution) if a certain data type is supported by the row format or not.
Describe the solution you'd like RowConverter::new should check supported types and return a Result<> if types are not supported. RowConverter::convert_{columns,rows} should never bail out due to unsupported types.
Describe alternatives you've considered
Piping some one-row payload through the converter to check if the types would be supported. This seems to be a hack though.
Additional context
-
The text was updated successfully, but these errors were encountered:
Check supported row format types when creating the converter instead of
during conversion. Also add an additional method
`RowConverter::supports_fields` to check types w/o relying on an error.
Closesapache#3077.
* refactor: remove duplicate code
Decimal types are already handled by `downcast_primitive`.
* refactor: check supported types when creating `RowConverter`
Check supported row format types when creating the converter instead of
during conversion. Also add an additional method
`RowConverter::supports_fields` to check types w/o relying on an error.
Closes#3077.
* Simplify
Co-authored-by: Raphael Taylor-Davies <r.taylordavies@googlemail.com>
Is your feature request related to a problem or challenge? Please describe what you are trying to do.
I'm trying to use the arrow row format in DataFusion. For that it would be helpful to know early (read: before piping any actual payload through physical execution) if a certain data type is supported by the row format or not.
Describe the solution you'd like
RowConverter::new
should check supported types and return aResult<>
if types are not supported.RowConverter::convert_{columns,rows}
should never bail out due to unsupported types.Describe alternatives you've considered
Piping some one-row payload through the converter to check if the types would be supported. This seems to be a hack though.
Additional context
-
The text was updated successfully, but these errors were encountered: