ceiling_date() on a daylight savings boundary

[eheinzen]

Both 1 am CST and 1 am CDT result in 1 am CDT when running ceiling_date(, "hour"). I would expect 1 am CST to round to 1 am CST, and 1 am CDT to round to 1 am CDT.

> packageVersion("lubridate")
[1] ‘1.9.3’
> as_datetime("2024-11-03T01:00:00-0600", tz = "America/Chicago") %>% ceiling_date("hour")
Date in ISO8601 format; converting timezone from UTC to "America/Chicago".
[1] "2024-11-03 01:00:00 CDT"
> as_datetime("2024-11-03T01:00:00-0500", tz = "America/Chicago") %>% ceiling_date("hour")
Date in ISO8601 format; converting timezone from UTC to "America/Chicago".
[1] "2024-11-03 01:00:00 CDT"

In other words, I would not expect this to happen:

> x <- as_datetime("2024-11-03T01:00:00-0600", tz = "America/Chicago")
Date in ISO8601 format; converting timezone from UTC to "America/Chicago".
> ceiling_date(x, "hour") >= x
[1] FALSE

[hersheyk]

I don't know if this would solve your problem but a similar issue seems to have been mentioned in Issue #640