Error in seq.default when trying to use a function after grouping per day

I have a dataframe with some values per time.

library(flux)
library(dplyr)
library(lubridate)

mydf <- tibble(mydate = as.Date(c("2019-05-11 23:55:00", "2019-05-11 23:56:00", "2019-05-11 23:57:00", "2019-05-11 23:58:00",
                                  "2019-05-12 23:55:00", "2019-05-12 23:56:00", "2019-05-12 23:57:00", "2019-05-12 23:58:00")),
               myval = c(101.23, 102.34, 105.43, 106.7, 107.1, 108.2, 109.2, 110.1))

I want to find auc per day. So, I am trying:

mydf <- mydf %>%
    group_by(day = floor_date(mydate, "day")) %>%
    summarise(Valmean = mean(myval, na.rm = TRUE),
              perDayIntegr = auc(time(day), Valmean))

and I am receiving:

Error in seq.default(x[1], x[2], length.out = dens) : 
  'to' must be a finite number

UPDATE:

If I use this dataset:

mydf <- tibble(mydate = as.Date(c("2019-05-11 23:55:00", "2019-05-11 23:56:00", "2019-05-11 23:57:00", "2019-05-11 23:58:00",
                                  "2019-05-12 23:55:00", "2019-05-12 23:56:00", "2019-05-12 23:57:00", "2019-05-12 23:58:00",
                                  "2019-05-13 23:55:00", "2019-05-13 23:56:00", "2019-05-13 23:57:00", "2019-05-14 23:58:00")),
               myval = c(101.23, 102.34, 105.43, 106.7, 107.1, 108.2, 109.2, 110.1, 111, 112, 113, 114))

where in the last line I am using only one row for date "2019-05-14 23:58:00" it still gives me the error.