Kirill Müller

Thanks for reaching out! From quickly glancing over it, duckdbfs works today, via dbplyr, and it might take us a while to achieve feature parity here, in particular regarding spatial...

Thanks. We have `duckplyr::duckplyr_df_from_file()` for reading from files. We might even support S3 URIs, worth a try. We translate dplyr to an intermediate representation dubbed "relational API", closer to Codd's...

duckplyr uses duckdb under the hood and inherits all the goodness. We'll need to see what operations we can support without having to `collect()`, but in any case this will...

Thanks, this is helpful. I'll review when I next work on duckplyr.

Thanks for raising this, Robin. Integration with the duckdb spatial extension would be a really cool feature, but also a lot of work. Do we need to figure out how...

Thanks, good question. I see two components: 1. Use in a new project 2. Use in an existing project ## New project No need to bother with connections. Start with...

Thanks, interesting. I wonder what the implementation of `as_duckplyr_df.nanoarrow_array_stream()` would look like?

`duckplyr::as_duckdb_tibble()` is a generic now.

We allow multiple `duckplyr_df` objects in the same con already. The opposite may be a bit trickier. @Tmonster: are there any obstacles combining multiple relational objects that were created from...

If we can't mix and match relational objects from different connections, we should check that they are the same for joins and other operations. We'll also take a look into...