Ryan Caudy
Ryan Caudy
As a user, I would like to be able to work with unsigned long types over Arrow Feather and Flight.
Improve and refactor `PartitioningColumnDataIndex` to support the expected types of location changes that will be introduced by refreshing Iceberg tables: 1. Move responsibility for location table row set shifting to...
Our existing Parquet reading code does not support v2 pages: * `io.deephaven.parquet.ColumnPageReaderImpl#readPageV2` throws `UnsupportedOperationException` * `io.deephaven.parquet.ColumnPageReaderImpl#readKeysFromPageV2` looks incomplete. As far as we can tell, v2 pages are not quite stable,...
Firstly, the over-arching issues: * `ZonedDateTime` instances are not equal to other `ZonedDateTime` instances with different time zones, and so when we *do* convert to a `long` for hashing or...
A number of our `ObjectCodec` implementations are not currently used by default (for example, `BigDecimalCodec`). We typically have round-trip tests, verifying that they can still write and re-read data correctly....
We'd like to be able to allow a single `SelectColumn` to produce multiple outputs. This is a step towards tuple-UDF support, but also will enable `ObjectProcessor` to be used as...
We should be more space efficient in cases like rolling-window row key spaces. This can be done with an update committer that iterates rows in `reclaimed` (`= current - previous`)...
#723 has some useful code snippets. We need to be able to automate this process to make it cheap and reliable.
Test query: ``` input=emptyTable(1_000_000_000).view("A=ii").where("A % 2 == 0") ```
Currently, we regard nested repetition and multi-column fields as uncommon and hard to map into a columnar data table like Deephaven's. This feature request is intended to capture views to...