nealrichardson commented on a change in pull request #9586:
URL: https://github.com/apache/arrow/pull/9586#discussion_r588628931
##########
File path: r/R/dplyr.R
##########
@@ -539,6 +537,13 @@ mutate.arrow_dplyr_query <- function(.data,
if (inherits(results[[new_var]], "try-error")) {
msg <- paste('Expression', as_label(exprs[[i]]), 'not supported in
Arrow')
return(abandon_ship(call, .data, msg))
+ } else if (is_dataset && !inherits(results[[new_var]], "Expression")) {
+ # We need some wrapping to handle literal values
+ if (length(results[[new_var]]) != 1) {
+ msg <- paste0('In ', new_var, " = ", as_label(exprs[[i]]), ", only
values of size one are recycled")
Review comment:
I agree that we should error gracefully, and `mutate(existing_column =
NULL)` seems like a valid thing to support. I expect this would just work with
a Table/RecordBatch now, just not for a dataset with the extra scalar handling
I added.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
