Jelte Fennema-Nio
Jelte Fennema-Nio
> > I kinda like 2, but I worry there are some cases that won't work with this. > > Yes, we cannot reliably detect such cases by the type...
Yeah I think that would be even better
Re-opening this as we still need to apply the same fix from #371 to the Display derive (#371 only touched the `Debug` derive).
> This PR is stale because it has been open 60 days with no activity. I think the github-actions bot cannot count, it's been only 30 days.
> This seems useful in general, and not specific to pg_duck. How about doing this in the main duckdb repo, or as a separate extension? Yeah, I think it makes...
> Typically duckdb does not need this type of caching because when you use duckdb, you have access to the local filesystem. I don't understand this argument (but I might...
This is now fixed by #130, I now added a test for this in #135
Are there any specific issues you hit when compiling for Greenplum? If so, feel free to list them here and/or create a PR for those.
@mkaruza I realized the initial version of this PR would conflict with your #99 PR, since you're changing the argument types there. Keeping these argument types in sync seems like...
It reproduces a bit inconsistently. But I can get a reproducing case quite easily by adding the following line: ```diff --- a/src/pgduckdb_node.cpp +++ b/src/pgduckdb_node.cpp @@ -71,6 +71,7 @@ ExecuteQuery(DuckdbScanState *state)...