CODEWITHSUNDEEP
rtimeposixintervals
iskandarblue
iskandarblue

Reputation: 7526

splicing time intervals posixct

I have the following time intervals that I would like to split into 10 equally spaced instances.

head(data)
             stoptime           starttime
1 2014-08-19 14:52:04 2014-08-19 15:22:04
2 2014-08-19 16:27:14 2014-08-19 17:17:33
3 2014-08-19 18:05:59 2014-08-19 18:09:12
4 2014-08-19 17:25:35 2014-08-19 17:29:06
5 2014-08-19 18:23:29 2014-08-19 18:57:34
6 2014-08-19 07:39:15 2014-08-19 07:48:49

I am able to take the midpoint using this code

one_day$midtime = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) /2 , origin = '1970-01-01')

however, when I try to extend this code to ten equally spaced instances it goes completely wrong. Why is this happening and how can I fix this code? 

one_day$first = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .1 , origin = '1970-01-01')
one_day$second = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .2, origin = '1970-01-01')
one_day$thrid = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .3, origin = '1970-01-01')
one_day$fourth = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .4, origin = '1970-01-01')
one_day$fifth = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .5, origin = '1970-01-01')
one_day$sixth = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .6, origin = '1970-01-01')
one_day$seventh = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .7, origin = '1970-01-01')
one_day$eighth = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .8, origin = '1970-01-01')
one_day$ninth = as.POSIXct((as.numeric(one_day$stoptime) + as.numeric(one_day$starttime)) * .9, origin = '1970-01-01')

head(one_day)
  diff.time            stoptime           starttime             midtime               first
1      1800 2014-08-19 14:52:04 2014-08-19 15:22:04 2014-08-19 15:07:04 1978-12-05 03:49:24
2      3019 2014-08-19 16:27:14 2014-08-19 17:17:33 2014-08-19 16:52:23 1978-12-05 04:10:28
3       193 2014-08-19 18:05:59 2014-08-19 18:09:12 2014-08-19 18:07:35 1978-12-05 04:25:31
4       211 2014-08-19 17:25:35 2014-08-19 17:29:06 2014-08-19 17:27:20 1978-12-05 04:17:28
5      2045 2014-08-19 18:23:29 2014-08-19 18:57:34 2014-08-19 18:40:31 1978-12-05 04:32:06
6       574 2014-08-19 07:39:15 2014-08-19 07:48:49 2014-08-19 07:44:02 1978-12-05 02:20:48
               second               thrid              fourth               fifth               sixth
1 1987-11-08 12:38:49 1996-10-11 21:28:14 2005-09-15 06:17:39 2014-08-19 15:07:04 2023-07-23 23:56:28
2 1987-11-08 13:20:57 1996-10-11 22:31:26 2005-09-15 07:41:54 2014-08-19 16:52:23 2023-07-24 02:02:52
3 1987-11-08 13:51:02 1996-10-11 23:16:33 2005-09-15 08:42:04 2014-08-19 18:07:35 2023-07-24 03:33:06
4 1987-11-08 13:34:56 1996-10-11 22:52:24 2005-09-15 08:09:52 2014-08-19 17:27:20 2023-07-24 02:44:48
5 1987-11-08 14:04:12 1996-10-11 23:36:18 2005-09-15 09:08:25 2014-08-19 18:40:31 2023-07-24 04:12:37
6 1987-11-08 09:41:36 1996-10-11 17:02:25 2005-09-15 00:23:13 2014-08-19 07:44:02 2023-07-23 15:04:50
              seventh              eighth               ninth
1 2032-06-26 08:45:53 2041-05-30 17:35:18 2050-05-04 02:24:43
2 2032-06-26 11:13:20 2041-05-30 20:23:49 2050-05-04 05:34:18
3 2032-06-26 12:58:37 2041-05-30 22:24:08 2050-05-04 07:49:39
4 2032-06-26 12:02:16 2041-05-30 21:19:44 2050-05-04 06:37:12
5 2032-06-26 13:44:44 2041-05-30 23:16:50 2050-05-04 08:48:56
6 2032-06-25 22:25:38 2041-05-30 05:46:27 2050-05-03 13:07:15

dput(data1)
structure(list(stoptime = structure(c(1408477924, 1408483634, 
1408489559, 1408487135, 1408490609, 1408451955, 1408452727, 1408498708, 
1408486644, 1408454996), class = c("POSIXct", "POSIXt"), tzone = "EST"), 
    starttime = structure(c(1408479724, 1408486653, 1408489752, 
    1408487346, 1408492654, 1408452529, 1408455826, 1408501153, 
    1408488389, 1408458514), class = c("POSIXct", "POSIXt"), tzone = "EST")), .Names = c("stoptime", 
"starttime"), row.names = c(NA, 10L), class = "data.frame")

Upvotes: 2

Views: 316

Answers (2)

SabDeM
SabDeM

Reputation: 7190

1: Seq

First of all you have to convert the columns of your dataframe as POSIXct or POSIXlt class, because the r base function seq has a method for objects of that class.

Just see this semplified code:

library(lubridate)
a <- "2014-08-19 14:52:04"
b <- "2014-08-19 15:22:04"

a <- ymd_hms(a)
b <- ymd_hms(b)

a
[1] "2014-08-19 14:52:04 UTC"
b
[1] "2014-08-19 15:22:04 UTC"

Then you have to just use the seq function and set the parameters length.out with the value of the sequence you are seeking. The code will automatically create a sequence of values from the start to the end equally divided.

seq(a, b, length.out = 10)
[1] "2014-08-19 14:52:04 UTC" "2014-08-19 14:55:24 UTC"
[3] "2014-08-19 14:58:44 UTC" "2014-08-19 15:02:04 UTC"
[5] "2014-08-19 15:05:24 UTC" "2014-08-19 15:08:44 UTC"
[7] "2014-08-19 15:12:04 UTC" "2014-08-19 15:15:24 UTC"
[9] "2014-08-19 15:18:44 UTC" "2014-08-19 15:22:04 UTC"

2: Vectorize step 1

Now that you know how to achieve your goal, it is just a matter of trying how to vectorize it along values.

I bet there are several approaches, here is one. With the mapply function you can loop trough the elements and match the first element (of the first object) with the first element (of the second object) and so on. Keep in mind that you have to specify which parameters are fixed with the MoreArg arguments.

Here is the code:

mapply(seq,
       to = data1$starttime,
       from = data1$stoptime,
       MoreArgs = list(length.out = 10),
       SIMPLIFY = F)

that produces a list of your desired data but not in the desired format sadly:

[[1]]
 [1] "2014-08-19 14:52:04 UTC" "2014-08-19 14:55:24 UTC"
 [3] "2014-08-19 14:58:44 UTC" "2014-08-19 15:02:04 UTC"
 [5] "2014-08-19 15:05:24 UTC" "2014-08-19 15:08:44 UTC"
 [7] "2014-08-19 15:12:04 UTC" "2014-08-19 15:15:24 UTC"
 [9] "2014-08-19 15:18:44 UTC" "2014-08-19 15:22:04 UTC"

[[2]]
 [1] "2014-08-19 16:27:14 UTC" "2014-08-19 16:32:49 UTC"
 [3] "2014-08-19 16:38:24 UTC" "2014-08-19 16:44:00 UTC"
 [5] "2014-08-19 16:49:35 UTC" "2014-08-19 16:55:11 UTC"
 [7] "2014-08-19 17:00:46 UTC" "2014-08-19 17:06:22 UTC"
 [9] "2014-08-19 17:11:57 UTC" "2014-08-19 17:17:33 UTC"

[[3]]
 [1] "2014-08-19 18:05:59 UTC" "2014-08-19 18:06:20 UTC"
 [3] "2014-08-19 18:06:41 UTC" "2014-08-19 18:07:03 UTC"
 [5] "2014-08-19 18:07:24 UTC" "2014-08-19 18:07:46 UTC"
 [7] "2014-08-19 18:08:07 UTC" "2014-08-19 18:08:29 UTC"
 [9] "2014-08-19 18:08:50 UTC" "2014-08-19 18:09:12 UTC"

[[4]]
 [1] "2014-08-19 17:25:35 UTC" "2014-08-19 17:25:58 UTC"
 [3] "2014-08-19 17:26:21 UTC" "2014-08-19 17:26:45 UTC"
 [5] "2014-08-19 17:27:08 UTC" "2014-08-19 17:27:32 UTC"
 [7] "2014-08-19 17:27:55 UTC" "2014-08-19 17:28:19 UTC"
 [9] "2014-08-19 17:28:42 UTC" "2014-08-19 17:29:06 UTC"

[[5]]
 [1] "2014-08-19 18:23:29 UTC" "2014-08-19 18:27:16 UTC"
 [3] "2014-08-19 18:31:03 UTC" "2014-08-19 18:34:50 UTC"
 [5] "2014-08-19 18:38:37 UTC" "2014-08-19 18:42:25 UTC"
 [7] "2014-08-19 18:46:12 UTC" "2014-08-19 18:49:59 UTC"
 [9] "2014-08-19 18:53:46 UTC" "2014-08-19 18:57:34 UTC"

[[6]]
 [1] "2014-08-19 07:39:15 UTC" "2014-08-19 07:40:18 UTC"
 [3] "2014-08-19 07:41:22 UTC" "2014-08-19 07:42:26 UTC"
 [5] "2014-08-19 07:43:30 UTC" "2014-08-19 07:44:33 UTC"
 [7] "2014-08-19 07:45:37 UTC" "2014-08-19 07:46:41 UTC"
 [9] "2014-08-19 07:47:45 UTC" "2014-08-19 07:48:49 UTC"

[[7]]
 [1] "2014-08-19 07:52:07 UTC" "2014-08-19 07:57:51 UTC"
 [3] "2014-08-19 08:03:35 UTC" "2014-08-19 08:09:20 UTC"
 [5] "2014-08-19 08:15:04 UTC" "2014-08-19 08:20:48 UTC"
 [7] "2014-08-19 08:26:33 UTC" "2014-08-19 08:32:17 UTC"
 [9] "2014-08-19 08:38:01 UTC" "2014-08-19 08:43:46 UTC"

[[8]]
 [1] "2014-08-19 20:38:28 UTC" "2014-08-19 20:42:59 UTC"
 [3] "2014-08-19 20:47:31 UTC" "2014-08-19 20:52:03 UTC"
 [5] "2014-08-19 20:56:34 UTC" "2014-08-19 21:01:06 UTC"
 [7] "2014-08-19 21:05:38 UTC" "2014-08-19 21:10:09 UTC"
 [9] "2014-08-19 21:14:41 UTC" "2014-08-19 21:19:13 UTC"

[[9]]
 [1] "2014-08-19 17:17:24 UTC" "2014-08-19 17:20:37 UTC"
 [3] "2014-08-19 17:23:51 UTC" "2014-08-19 17:27:05 UTC"
 [5] "2014-08-19 17:30:19 UTC" "2014-08-19 17:33:33 UTC"
 [7] "2014-08-19 17:36:47 UTC" "2014-08-19 17:40:01 UTC"
 [9] "2014-08-19 17:43:15 UTC" "2014-08-19 17:46:29 UTC"

[[10]]
 [1] "2014-08-19 08:29:56 UTC" "2014-08-19 08:36:26 UTC"
 [3] "2014-08-19 08:42:57 UTC" "2014-08-19 08:49:28 UTC"
 [5] "2014-08-19 08:55:59 UTC" "2014-08-19 09:02:30 UTC"
 [7] "2014-08-19 09:09:01 UTC" "2014-08-19 09:15:32 UTC"
 [9] "2014-08-19 09:22:03 UTC" "2014-08-19 09:28:34 UTC"

At this point I guess it is just a matter of same data manipulation but I can't figure out a way (now).

Upvotes: 4

CPhil
CPhil

Reputation: 927

You can't just multiply the time interval by 0.1, you have to add that 0.1 of the time interval to the earlier time. For example:

one_day$firstexample = one_day$stoptime + 0.1*difftime(one_day$starttime, one_day$stoptime, units = "mins")

As a side note, if you find yourself typing out very similar things multiple times, that's usually a sign that you should turn it into a function.

Upvotes: 1

Related Questions