如何使用 R 中的 DBI 将年月日期时间组合成单个日期时间列？

Question

Here's an example of my source data csv, df:

    Year             : int  2005 2005 2005 2005 2005 2005 2005 2005 2005 2005 ...
    Month            : int  1 1 1 1 1 1 1 1 1 1 ...
    DayofMonth       : int  28 29 30 31 2 3 4 5 6 7 ...
    Time             : int  1605 1605 1610 1605 1900 1900 1900 1900 1900 1900

How can I combine the 4 columns into a single column into a datetime column in my database using SQLite and not dplyr?

I hope the output can be something like: 2005-01-29 09:30:00 so that I can plot graphs.

Answer 1

You could use lubridate::make_datetime :

Year =c(2005, 2005, 2005)
Month = c(1,2,3) 
DayofMonth = c(  28, 28, 30)
Time = c(1605, 1605, 1610)


lubridate::make_datetime(Year,Month,DayofMonth,floor(Time/100),Time%%100)

[1] "2005-01-28 16:05:00 UTC" "2005-02-28 16:05:00 UTC" "2005-03-30 16:10:00 UTC"

Answer 2

Base R, using Waldi's sample data (thanks!):

with(df, as.POSIXct(sprintf("%i-%02i-%i %02i:%02i:00", 
  Year, Month, DayofMonth, Time %/% 100, Time %% 100))
)
# [1] "2005-01-28 16:05:00 EST" "2005-02-28 16:05:00 EST" "2005-03-30 16:10:00 EST"

This is (possibly naïvely) assuming that you never have seconds or fractional minutes or Time values that are not "meaningful" (ie, more than 59 minutes, more than 23 hours).

---

Data

df <- structure(list(Year = c(2005, 2005, 2005), Month = c(1, 2, 3), DayofMonth = c(28, 28, 30), Time = c(1605, 1605, 1610)), class = "data.frame", row.names = c(NA, -3L))