-
**Describe the bug**
A clear and concise description of what the bug is.
https://github.com/apache/arrow-datafusion/blob/58b43f5c0b629be49a3efa0e37052ec51d9ba3fe/datafusion/core/src/physical_pla…
-
Currently, the connector reports physical column names in the EXAPLIN result. It's a little hard to understand field names especially when dereference pushdown happened with column mapping mode `id` o…
-
# Purpose
This issue is aimed to have a basic optimize rule for `min/max/count` queries on the connectors having accurate `table/partitions/columns statistics`, like `Iceberg` composed of Orc/Parquet…
-
# Goal
In this issue we set up the project to support the Iceberg table format in the Presto native engine - Prestissimo. This project will help the user to be able to read Iceberg tables written by …
-
Pushing predicate through limit might change query semantics (e.g when data is ordered). We don't push predicate through limit in Presto.
CC: @findepi
-
### What happens?
Hello there,
we have a on-prem k8s cluster, with a hive partitioned Minio based bucket containing for test purposes ~ 50 files with total less than 50 MB.
We constructed these…
-
Since 320, predicate pushdown does't work for constants from outside the scope of subquery and it slowdowns specific type of queries significantly.
For example, some queries have a chance to reduce…
-
**Is your feature request related to a problem? Please describe.**
This RFC is meant to propose a heuristic join reordering rule for multi-way inner joins that is meant to:
1. eliminate cross joins …
-
Не во всех ситуациях count() из таблиц с движком Set работает.
Создаем для примера три таблицы на разных движках
>create table internal.test_set (key UInt64) engine = Set();
create table internal…
-
I'd like to read in layers which had an id and use that as a column in the table. `ogr2ogr` has this option as [`preserve_fid`](https://gdal.org/programs/ogr2ogr.html#cmdoption-ogr2ogr-preserve_fid)
…