Thijs

I've made [a PR](https://github.com/duckdb/duckdb/pull/14034) to fix this upstream in DuckDB, it likely won't land until 1.2 even if it passes CI because it's very close to the core and deemed...

I thought about that, but I think I read that's ill-advised? Another thing we could do is make a change to duckdb that allows a callback to be registered for...

@JelteF ```c++ // FIXME: TextDatumGetCstring allocates so it needs a // guard, but it's a macro not a function, so our current gaurd // template does not handle it. ```...

Now that the duckdb connection is a singleton this needs some more work, as it currently assumes the catalog to be thrown away after every query. The created catalog entries...

@JelteF please review and let me know what the remaining blockers are, it's been open quite a while now. I'd like to stop having to fix this with every PR...

when I merge with main I can't resolve a crash in `test/regression/sql/query_filter.sql` `SELECT a, c FROM query_filter_output_column WHERE b = 't1';` is what triggers it, it's a EXC_BAD_ACCESS somewhere in...

> > when I merge with main I can't resolve a crash in test/regression/sql/query_filter.sql > > SELECT a, c FROM query_filter_output_column WHERE b = 't1'; is what triggers it, it's...

[This issue](https://github.com/hydradatabase/pg_quack_internal/issues/43) would be fixed as a side-effect

> [This issue](https://github.com/hydradatabase/pg_quack_internal/issues/43) would be fixed as a side-effect Little update on this: it won't be entirely fixed by this, DuckDB has some limitations on bindings, and with this change...

https://github.com/hydradatabase/pg_quack_internal/issues/67