Two time points are considered duplicated but are different numeric values

Question

I have a time-series of hourly weather data from 2018. Two time points are showing up as duplicated even though they are different values. The same points are considered duplicates before and after the series is converted to POSIXct. In every case they appear as different numeric values. I am not sure how to create a REPREX since the values are not considered duplicates after I create a dataframe containing the numeric values. I am reading the data into R from an XLSX file. My guess is the problem is somehow created by timezones and daylight saving time, but I would expect duplicated points when an hour is "gained" in October rather than in June. Any advice would be greatly appreciated.

df$Time[which(duplicated(df$Time))]
[1] 43253.50 43253.54

class(df$Time)
[1] "numeric"


df$Time[which(duplicated(df$Time))]
[1] "2018-06-02 12:00:00 UTC" "2018-06-02 13:00:00 UTC"

class(df$Time)
[1] "POSIXct" "POSIXt"

unclass(df$Time[which(duplicated(df$Time))])
[1] 1527940800 1527944400
attr(,"tzone")
[1] "UTC"

Answer 1

duplicated "returns a logical vector indicating which elements (rows) are duplicates". That is, the first occurrence (not yet a duplicate) returns FALSE.

Both of those times appear once more in df$Time

Answer 2

Your use of duplicated() only returns one instance of a duplicate value (but more when there are three or more values):

df <- data.frame(Time = c("Mon", "Mon", "Tue", "Wed", "Wed", "Wed", "Thu", "Fr"))

df$Time[duplicated(df$Time)]

# [1] "Mon" "Wed" "Wed"

You could alternatively use

janitor::get_dupes(df, Time)

#   Time dupe_count
# 1  Mon          2
# 2  Mon          2
# 3  Wed          3
# 4  Wed          3
# 5  Wed          3