Open asfimport opened 3 years ago
Neal Richardson / @nealrichardson: Here's an example from the taxi dataset that shows a messy string column and three kinds of queries: base R function, stringr, and an arrow_ prefixed compute function that doesn't have an R mapping (afaik)
ds %>%
select(payment_type) %>%
mutate(
allcaps = toupper(payment_type),
chars = str_length(allcaps),
is_upper = arrow_ascii_is_upper(payment_type)
) %>%
head(10) %>%
collect()
Ian Cook / @ianmcook:
We need to fix the type error in payment_type == 3
that Jon describes here: https://issues.apache.org/jira/browse/ARROW-12114?focusedCommentId=17314989&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-17314989
Dragoș Moldovan-Grünfeld / @dragosmg:
I thought it might be a good idea to add some information regarding case_when()
only working inside a mutate() context and not standalone. See ARROW-14362.
Nicola Crane / @thisisnic: We should also include something on the NSE funcs which do have any differences, see https://github.com/apache/arrow/pull/11575#pullrequestreview-796850588 for an example
Will Jones / @wjones127: We may also wish to mention Acero by name as the backend for dplyr queries. We'll likely be discussing dplyr in the upcoming Acero post https://issues.apache.org/jira/browse/ARROW-16632
Reporter: Neal Richardson / @nealrichardson
Related issues:
Note: This issue was originally created as ARROW-12137. Please see the migration documentation for further details.