How do I extract only the time parameters from Datetime variable in R?

Question

In a R dataframe I have time variable. The data is in the format %a-%b-%d %H:%M:%S. for e.g.,

2015-03-23 20:00:00

I want to get the following data only

  20:00:00

I have created a table basis the above said variable and trying to make a line graph:

                     Var1 Var2  Freq
    1 2015-03-24 00:00:00   RT   612
    2 2015-03-24 01:00:00   RT    65
    3 2015-03-24 06:00:00   RT    58
    4 2015-03-24 07:00:00   RT  5132
    5 2015-03-24 08:00:00   RT  4483
    6 2015-03-24 09:00:00   RT 11112

I have used the following code to make a ggplot line graph:

   library(ggplot2)
   library(stringr)
   ggplot(rtt, aes(x = as.factor(Var1), y = Freq, colour = Var2, group = Var2)) + geom_line(size = 1) +
    xlab("R Vs T") + geom_point() +
    scale_x_discrete(labels = function(x) str_wrap(x, width = 2)) +
    ggtitle("Number of T Vs R - through the day") +
    theme(plot.title=element_text(size=rel(1.2), lineheight = 1 ))

How do I remove the YMD data from this, because I want only the time and not the data in x-axis and the x-axis in the graph looks completely garbled.

akrun · Accepted Answer · 2015-03-31T03:54:42.353

1

There are a number of options to extract the 'time' part. Some are listed below:

 format(as.POSIXct(str1), '%H:%M:%S')
 [1] "20:00:00"

Or

 sub('[^ ]+ ', '', str1)
 #[1] "20:00:00"

Or

 strftime(str1, format='%H:%M:%S')
 #[1] "20:00:00"

Or

 library(lubridate)
 format(ymd_hms(str1), '%H:%M:%S')
 #[1] "20:00:00"

The ggplot code can be changed to

 library(ggplot2)
 ggplot(rtt, aes(x= factor(strftime(Var1, format='%H:%M:%S')),
     y= Freq, colour=Var2, group=Var2)) +
     xlab("R Vs T") +
     geom_point() + 
     scale_x_discrete(labels = function(x) str_wrap(x, width = 2)) +
     ggtitle("Number of T Vs R - through the day") +
     theme(plot.title=element_text(size=rel(1.2), lineheight = 1 ))

Update

If you need to extract only the 'hour' part

 library(lubridate)
 hour(ymd_hms(str1))
 #[1] 20

data

 str1 <- '2015-03-23 20:00:00'

 rtt <- structure(list(Var1 = c("2015-03-24 00:00:00", 
 "2015-03-24 01:00:00", 
 "2015-03-24 06:00:00", "2015-03-24 07:00:00", "2015-03-24 08:00:00", 
 "2015-03-24 09:00:00"), Var2 = c("RT", "RT", "RT", "RT", "RT", 
 "RT"), Freq = c(612L, 65L, 58L, 5132L, 4483L, 11112L)), 
 .Names = c("Var1", "Var2", "Freq"), class = "data.frame",
  row.names = c(NA, -6L))

edited Mar 31 '15 at 03:54

answered Mar 29 '15 at 06:52

akrun

874,273
37
540
662

Many thanks...akrun......your solution on the ggplot had worked!....your other solutions worked on the variable when i used it creating a dummy variable...but for some reason the date still remained in the table...when I used it on the data frame. But I am sure it has got a lot to do with my df....thank you again.... – Apricot Mar 29 '15 at 07:12
@MageshGovindan Thanks for the reply. I am not sure what you really meant. Perhaps, you need to assign the date to the transformed date. For example. `rtt$Var1 <- strftime(rtt$Var1, format='%H:%M:%S')` – akrun Mar 29 '15 at 07:14
oops....still figuring out the mechanics of stackoverflow's appreciation models.....I was only trying to accept both the answers....thank you for your help again.... – Apricot Mar 31 '15 at 08:01
@MageshGovindan Thanks for the reply. Yes, it may be confusing initially. – akrun Mar 31 '15 at 13:06

score 1 · Answer 2 · answered Mar 29 '15 at 13:44

Since the time consists only of hours:

library(ggplot2)
rtt$hour <- as.POSIXlt(rtt$Var1)$hour
ggplot(rtt, aes(hour, Freq, col = Var2)) + geom_line()

Note: We used this for rtt:

Lines <- "Var1,Var2,Freq
2015-03-24 00:00:00,RT,612
2015-03-24 01:00:00,RT,65
2015-03-24 06:00:00,RT,58
2015-03-24 07:00:00,RT,5132
2015-03-24 08:00:00,RT,4483
2015-03-24 09:00:00,RT,11112"
rtt <- read.csv(text = Lines, as.is = TRUE)

enter image description here

How do I extract only the time parameters from Datetime variable in R?

2 Answers2

Update

data