`n_distinct()` does not work with non-trivial windows

[lschneiderbauer]

I cannot run a reprex just now, but the following does not take into account the partition column x which is incorrect.

library(dplyr)
library(duckdb)
library(DBI)

df <- tibble(x = c(1, 1, 2), y = c(1, 1, 2))

con <- DBI::dbConnect(duckdb::duckdb())
duckdb::duckdb_register(con, "df", df)

tbl(con, "df") |> mutate(z = n_distinct(y), .by = x) |> show_query()

I am planning to provide a fix on the upcoming weekend.

[lschneiderbauer]

Side comment: I believe the same is true for the translation of "str_flatten()".

[lschneiderbauer]

Probably this line is the culprit:

duckdb-r/R/backend-dbplyr__duckdb_connection.R

Line 117 in 3ede6f1

win_current_group <- pkg_method("win_current_order", "dbplyr")

[github-actions]

This old thread has been automatically locked. If you think you have found something related to this, please open a new issue and link to this old issue if necessary.