Trying to convert dates stored as numbers in a column

General
#1

I have this column that is a mixture of "NO" and numbers that are really just badly imported dates.
Is there any way to keep the "NO" but modify the numbers to be the correct date?

# A tibble: 114.167 × 1
   hpv_pre_fecha
   <chr>        
 1 NO           
 2 NO           
 3 NO           
 4 41064        
 5 NO           
 6 NO           
 7 NO           
 8 NO           
 9 NO           
10 41019

I tried this code but it didnt work:

base %>%
  select(hpv_pre_fecha) %>%
  mutate(hpv_pre_fecha2= if_else(str_detect(hpv_pre_fecha,
                                           "[:punct:]"),
                                hpv_pre_fecha %>%
                                  as.numeric() %>%
                                  excel_numeric_to_date() %>%,
                                .))
#2

I know you said you wanted to keep the "NO"s, but in case you are also fine with NAs instead of NOs:

janitor::convert_to_date(data$hpv_pre_fecha, string_conversion_failure = "warning")

This would give you a column with dates converted and NA for anything it could not convert.

#3

Hi, thanks for the answer, but I need to keep the "NO"

#4

OK, in that case just keep on processing:

janitor::convert_to_date(data$hpv_pre_fecha, string_conversion_failure = "warning") |> as.character() |> replace_na("NO")

But the downside is that your dates are now characters...