R extract Date and Time Info

问题

I have a data.frame that looks like this:

> df1
   Date         Name    Surname   Amount
2015-07-24      John     Smith     200

I want to extrapolate all the infos out of the Date into new columns, so I can get to this:

> df2
   Date     Year  Month   Day    Day_w      Name    Surname   Amount
2015-07-24  2015    7     24    Friday      John     Smith     200

So now I'd like to have Year, Month, Day and Day of the Week. How can I do that? When I try to first make the variable a date using as.Date the data.frame gets messed up and the Date all become NA (and no new columns). Thanks for your help!

回答1:

Maybe this helps:

df2 <- df1
dates <- strptime(as.character(df1$Date),format="%Y-%m-%d")
df2$Year <- format(dates, "%Y")
df2$Month <- format(dates, "%m")
df2$Day <- format(dates, "%d")
df2$Day_w <- format(dates, "%a")

Afterwards you can rearrange the order of columns in df2as you desire.

回答2:

Here's a simple and efficient solution using the devel version of data.table and its new tstrsplit function which will perform the splitting operation only once and also update your data set in place.

library(data.table)
setDT(df1)[, c("Year", "Month", "Day", "Day_w") := 
             c(tstrsplit(Date, "-", type.convert = TRUE), wday(Date))]
df1
#          Date Name Surname Amount Year Month Day Day_w
# 1: 2015-07-24 John   Smith    200 2015     7  24     6

Note that I've used a numeric representation of the week days because there is an efficient built in wday function for that in the data.table package, but you can easily tweak it if you really need to using format(as.Date(Date), format = "%A") instead.

---

In order to install the devel version use the following

library(devtools)
install_github("Rdatatable/data.table", build_vignettes = FALSE)

来源：https://stackoverflow.com/questions/31648858/r-extract-date-and-time-info