Midnight timestamp not returning expected result

[JegorG]

library(lubridate)

dmy_hm("Tuesday, 16 Mar 1999, 00:01")

"1999-03-16 00:01:00 UTC"

dmy_hm("1999-03-16 00:00:00 UTC")

"1999-03-16 UTC"

I expect

"1999-03-16 00:00:00 UTC"

[vspinu]

You are confused about printed representation and actual object. These are not the same. Whenever in doubt use str.

[challisd]

You can get what you want with format(dmy_hm("1999-03-16 00:00:00 UTC"), "%Y-%m-%d %H:%M:%S %Z")

[challisd]

Also I agree with the original poster that this is a bug. I understand the underlying representation is currently the same, but if a user has a string with hours and minutes, parses it with a function called dmy_hm and then converts back to a string it is perfectly reasonable to expect you'll get back a string with hours and minutes. Technical justification aside, this is a bug in that it does not behave the way virtually all users will expect it to.

[vspinu]

you'll get back a string with hours and minutes.

It's not a string that is returned but a POSIXct object. The issue here is that R prints the midnight POSIXct objects without the trailing 00:00:00 for the space efficiency reasons. It's not such a big deal to get used to this, or if you are really troubled by it, you could write your own print.POSIXct method and place it in .Rprofile. In any case it's not lubridate's issue, it's how R prints things.

[JBrocklyAnderson]

So does this mean that the 00:00:00 time value still exists in the cell?

[srlamunion]

So does this mean that the 00:00:00 time value still exists in the cell?

That is my question also. It doesn't appear to be when written to .csv for example. The cell shows just the date and not the trailing 00:00:00 to go with it.