Jelte Fennema-Nio
Jelte Fennema-Nio
Did some of the initial to support PG18 beta1. It definitely needs another self-review, but seems good to at least put it in a draft PR.
### Description A few things to consider here: 1. Altering schemas within the same default motherduck database 2. Altering schemas within a single non-default (aka `ddb$`) motherduck database 3. Altering...
### Description There seems to be an issue with the background worker syncing. It seems to have something to do with the background worker trying to remove non-existent tables for...
### Description This query returns `0`: ``` select * from duckdb.query($$ select '-170141183460469231731687303715884105728'::hugeint as result $$); ``` But one higher works fine: ``` select * from duckdb.query($$ select '-170141183460469231731687303715884105728'::hugeint +...
### Description ```sql create table t(a numeric); insert into t values (1.8000), (1.346); select round(a, 2) from t; ``` With PG execution this returns: ``` round ─────── 1.80 1.35 ```...
In #281 the function `duckdb.reset_ddb()` was added, which recreates the in memory database. This has as a sideeffect that all temporary duckdb tables are lost. We should also remove the...
### Description It would be good to allow users to call a function like `duckdb.wait_for_motherduck_sync()`. That way they could ensure that DDL changes made outside of pg_duckdb (e.g. from the...
### Description We currently only support calling scalar functions that exist in DuckDB from DuckDB. There are (at least) three situations worth considering here: 1. Converting a PG-only type (e.g....
### Description We added bunch of DuckDB-only functions in this release. Like `time_bucket`, `strftime`, `strptime`. We should add documentation for those so people know they exist. `strftime` and `strptime` should...
### Description The background worker code uses a mix of C and C++, that's a recipe for problems. We should make this code safe.