Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
Description
random() generates uniform distributed values. We can probably bind that somehow to runif() in dplyr (will have to ignore the `n` argument). If nothing else, it can be used to implement slice_sample() (see also ARROW-13766, ARROW-13767).
FWIW there's sql_random() in dbplyr, not sure if that's actually exposed meaningfully or just used to implement slice_sample.