rok commented on code in PR #13506:
URL: https://github.com/apache/arrow/pull/13506#discussion_r931730491
##########
r/R/dplyr-datetime-helpers.R:
##########
@@ -171,61 +171,35 @@ build_formats <- function(orders) {
# this is needed for 2 reasons:
# 1. strptime does not parse "2022-05" -> we add "-01", thus changing the
format,
# 2. for equivalence to lubridate, which parses `ym` to the first day of the
month
- short_orders <- c("ym", "my")
+ short_orders <- c("ym", "my", "yOm", "Omy")
+ quarter_orders <- c("yq", "qy")
if (any(orders %in% short_orders)) {
orders1 <- setdiff(orders, short_orders)
orders2 <- intersect(orders, short_orders)
orders2 <- paste0(orders2, "d")
orders <- unique(c(orders2, orders1))
}
-
- if (any(orders == "yq")) {
- orders1 <- setdiff(orders, "yq")
- orders2 <- "ymd"
- orders <- unique(c(orders1, orders2))
- }
-
- if (any(orders == "qy")) {
- orders1 <- setdiff(orders, "qy")
- orders2 <- "ymd"
- orders <- unique(c(orders1, orders2))
+ if (any(orders %in% quarter_orders)) {
+ orders <- c(setdiff(orders, quarter_orders), "ymd")
}
+ orders <- unique(orders)
- ymd_orders <- c("ymd", "ydm", "mdy", "myd", "dmy", "dym")
- ymd_hms_orders <- c(
- "ymd_HMS", "ymd_HM", "ymd_H", "dmy_HMS", "dmy_HM", "dmy_H", "mdy_HMS",
- "mdy_HM", "mdy_H", "ydm_HMS", "ydm_HM", "ydm_H"
- )
- # support "%I" hour formats
- ymd_ims_orders <- gsub("H", "I", ymd_hms_orders)
-
- supported_orders <- c(
- ymd_orders,
- ymd_hms_orders,
- gsub("_", " ", ymd_hms_orders), # allow "_", " " and "" as order separators
- gsub("_", "", ymd_hms_orders),
- ymd_ims_orders,
- gsub("_", " ", ymd_ims_orders), # allow "_", " " and "" as order separators
- gsub("_", "", ymd_ims_orders)
- )
-
- unsupported_passed_orders <- setdiff(orders, supported_orders)
- supported_passed_orders <- intersect(orders, supported_orders)
+ formats_list <- map(orders, build_format_from_order)
+ formats_length <- map(map(formats_list, nchar), max)
+ invalid_orders <- formats_length < 6
Review Comment:
Lubridate checks against an example of it's input, which is something we
can't/won't do.
I'd say erroring on invalid formats and returning NA on unsuccessful parsing
would be the thing to do.
--
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]
