Apply a function to rows within a data table in R to return time of day

Question

I have a data table, which I have various dates and a "value" with which I need to plot. The values are delivered in a so-called "Delivery Hour" and a "Delivery Interval". An example is below;

enter image description here

I can join a new colum to the right of my table, popultaed (in the first instance) with NAs using

data.table["TIME"]<-NA

enter image description here

This new column (shown below) needs to be populated in each row by the time of day that "value" was delivered, in excel I convert this via the function

=TIME(DELIVERY_HOUR-1, 30*(DELIVERY_INTERVAL-1),0)

So for example, with DELIVERY_HOUR =7 and DELIVERY_INTERVAL=2 this is 06:30.

I'd like to write a function that populates the TIME column row by row, here is my syntax for the function "TIME":

TIME function(x){ time<-(DELIVERY_HOUR-1):(30*(DELIVERY_INTERVAL-1)) result<-paste(time,format="%H%M") return(result) }

I am unsure how to ensure that corresponding row entries are picked and placed into the row entry for TIME.

Any help on this matter would be greatly appreciated.

Just out of curiosity, if you allow tags to R, why not questions about R? — Autolatry, Feb 20 '15 at 11:06
if you are blocked at SO, [system simply won't let question migrate](http://meta.programmers.stackexchange.com/q/7020/31260) — gnat, Feb 20 '15 at 11:12
this is my very first question, I thought it was relevant to this site; there is no reason why I would be blocked at SO. — Autolatry, Feb 20 '15 at 11:13
if this is the case, you can safely flag for migration (flag -> off-topic -> belongs to other site -> ...). Alternatively, you can delete here and reask at the other site — gnat, Feb 20 '15 at 11:16
What's with immediately assuming this user is blocked from asking questions on SO? Chill out, it's Friday. We do, however, need actual data and not pictures of data. Please have a look at http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example. — Rich Scriven, Feb 20 '15 at 17:02
@RichardScriven ["Per stats, Programmers appear to be most popular of SE sites among this category of askers..."](http://meta.programmers.stackexchange.com/a/6991/31260) — gnat, Feb 27 '15 at 08:46

score 1 · Accepted Answer · answered Feb 20 '15 at 17:14

Well if you are actually using the data.table in R you could do the following:

library(data.table)

# minimally reproducible data
dt <- data.table(date=20150202, delivery_date="02/02/2015",
                 delivery_hour = c(rep(7, 4),rep(8,4),9),
                 delivery_interval = c(1,1,2,2,1,1,2,2,1))

dt[, time := paste0(delivery_hour - 1, ":", 30*delivery_interval]
dt
       date delivery_date delivery_hour delivery_interval time
1: 20150202    02/02/2015             7                 1 6:30
2: 20150202    02/02/2015             7                 1 6:30
3: 20150202    02/02/2015             7                 2 6:60
4: 20150202    02/02/2015             7                 2 6:60
5: 20150202    02/02/2015             8                 1 7:30
6: 20150202    02/02/2015             8                 1 7:30
7: 20150202    02/02/2015             8                 2 7:60
8: 20150202    02/02/2015             8                 2 7:60
9: 20150202    02/02/2015             9                 1 8:30

This naturally assumes the 24 hour clock time (delivery hour will not be less than 1 and not exceed 24)

Apply a function to rows within a data table in R to return time of day

1 Answers1