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.
Certain queries may only be interested in the number of rows, and not any actual column data. In theory this could be handled as a projection pushdown that selects no columns, but currently there is no way to create a RecordBatch with no columns and a non-zero row count and so this doesn't work.
Describe the solution you'd like
I would like RecordBatch to store the row_count as a struct member, and allow this to be specified in RecordBatchOptions
Describe alternatives you've considered
We could not make this change
Additional context
This problem can also arise when using Hive style partitioning, where the underlying file data may not contain all columns - apache/datafusion#2000
The text was updated successfully, but these errors were encountered:
)
* Support empty RecordBatch (#1536)
* Placate clippy
* Review feedback
* Fix doc
* Fix create_record_batch_slice_empty_batch test
alamb
changed the title
Create RecordBatch With Non-Zero Row Count But No Columns
Support RecordBatch with No Columns but Non-Zero Row Count
Apr 15, 2022
alamb
changed the title
Support RecordBatch with No Columns but Non-Zero Row Count
Support RecordBatch with zero columns but non zero row count
Apr 15, 2022
Is your feature request related to a problem or challenge? Please describe what you are trying to do.
Certain queries may only be interested in the number of rows, and not any actual column data. In theory this could be handled as a projection pushdown that selects no columns, but currently there is no way to create a RecordBatch with no columns and a non-zero row count and so this doesn't work.
Describe the solution you'd like
I would like
RecordBatch
to store therow_count
as a struct member, and allow this to be specified inRecordBatchOptions
Describe alternatives you've considered
We could not make this change
Additional context
This problem can also arise when using Hive style partitioning, where the underlying file data may not contain all columns - apache/datafusion#2000
The text was updated successfully, but these errors were encountered: