jonkeane commented on a change in pull request #12433:
URL: https://github.com/apache/arrow/pull/12433#discussion_r813328095
##########
File path: r/R/dplyr-funcs-datetime.R
##########
@@ -164,4 +164,10 @@ register_bindings_datetime <- function() {
return(semester)
}
})
+ register_binding("date", function(x) {
+ if (inherits(x, "POSIXct")) {
+ x <- vec_to_Array(x, date32())
+ }
+ x$cast(date32())
Review comment:
This implementation made me think of the various `as.*` methods we have
defined [1] (since this is similar to `as.Date()`). Which all use a simpler
setup to create a cast operation. However, I noticed that for those, they are
using `Expression$create(...)` rather than the `build_expr(...)` helper [2].
That `build_expr(...)` here _should_ handle the wrapping of R objects into
Scalars that you're doing in this first bit here. Can you try that out and
confirm that it does what we are looking for with inputs that are both columns
in an arrow table/recordbatch/dataset etc. as well as if it's a single value
from R (i.e. `mutate(term_start = date("2020-01-01")`).
Separately: We should also open a jira if one doesn't exist to clean up
those `as.*` methods to use `build_expr()`
[1]
https://github.com/apache/arrow/blob/a916e6015d7cab397d3b769d760b943a60ea3d97/r/R/dplyr-funcs-type.R#L45-L78
[2]
https://github.com/apache/arrow/blob/a916e6015d7cab397d3b769d760b943a60ea3d97/r/R/expression.R#L195-L259
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
