mkaruza
mkaruza
IndexScan is always calculated when `enable_indexscan` is true. This is postgres GUC variable. Columnar variable GUC is `columnar_index_scan`. Previously columnar was adding *extra cost* for IndexScan path - basically pushing...
Hi @leoyvens thank you for report. Lowering stripe max row count will increase stripes for table and thus related metadata. Part of planning is to read table stripe metadata to...
@japinli hi can you extend this PR with regression test ?
FYI, if first connecton is caching remote file: ``` postgres=# select duckdb.cache('https://datasets.clickhouse.com/hits_compatible/hits.parquet'); ``` Until this file is downloaded all other queries on same remote file will fail as: ``` postgres=#...
Agree with @JelteF this should be fixed asap. Simplest, most straight forward solution would be to fix this once we would have one duckdb singleton wrapper object per process -...
@JelteF we have used this ticket just as indicator and probably at this point doesn't make sense that it is still open. For any future data type support we should...
Closing this issue. We should probably track new supported types by creating independent tickets.
@JelteF agree not critical, but could be approach to get better performances of heap table scans (specifically to fetch table pages in parallel; currently there is bottleneck there as only...
Just to comment on your example. In case of `SELECT *` AS part should have all columns named with expected type. But, for minimum i would expect that columns that...