How to break each element in a dataset into its constituent parts?

Question

So I have a relatively simple problem (and I think that there may be some duplicates of my question out there) but I just can't seem to figure it out and I would really appreciate any and all help.

I have a dataset and in one column, I have multiple rows of different 11-digit numbers. I hope to obtain the last 6 digits of each number and I hope to be able to create a new column in my dataset with the results.

Below is an example:

random_num <- c(11001100100, 11001100300, 11001100400,
                11001100501, 11001100502, 11001100600)
random_stuff <- c(2, 5, 6, 2, 5, 3)
data_frame <- cbind(random_num, random_stuff)

And I hope to get an output that shows something like this:

endresult

So far, this is what I have:

conversion <- function (x) {
  for (i in nrow(x)) 
  {
  c <- as.character[i]
  be <- substring(c, seq(1, nchar(c), 1), seq(1, nchar(c), 1))
  ad <- paste(be[6], be[7], be[8], be[9], be[10], be[11], sep = "")
  final <- as.numeric(ad)
  return(final)
  }
}

finalr <- conversion(data_frame)
finalr

But I either get the error message saying that

'Error in as.character[i]: object of type 'builtin' is not subsettable' or 'Error in mutate_impl(.data, dots) : Evaluation error: 'to' must be of length 1.'

Will really appreciate any advice. Thank you!

`as.character` is a function. You are trying to use that function extract something (`[` — akrun, May 05 '19 at 04:41
You showed a matrix and not a data.frame. Also, `substring/substr` is vectorized, so no need to loop i.e. `substr(as.character(data_frame[,1]), nchar(data_frame[,1])-6, nchar(data_frame[,1]))` — akrun, May 05 '19 at 04:44
something like this would work `substring(df[, 1], nchar(df[, 1]) - 5)` — Ronak Shah, May 05 '19 at 04:45

Tim Biegeleisen · Answer 1 · 2019-05-05T05:17:01.960

1

From what I can, the result column/vector is just the last six digits from the random_num vector. So, we can use the modulus to compute this:

random_num <- c(1100100100, 1100100300, 1100100400,
                1100100501, 1100100502, 1100100600)
result <- random_num %% 1000000
result

[1] 100100 100300 100400 100501 100502 100600

This answer avoids a potentially unnecessary cast from numeric to character.

edited May 05 '19 at 05:17

answered May 05 '19 at 04:41

Tim Biegeleisen

502,043
27
286
360

1

@akrun Try again. The OP had a typo in its sample data. Now the above uses the version which actually matches the expected output. – Tim Biegeleisen May 05 '19 at 05:16
Yeah, it works now, I was scratching my head whether the R version is wrong or not – akrun May 05 '19 at 05:16

score 0 · Accepted Answer · answered May 05 '19 at 04:54

I've converted your dataset in a dataframe:

random_num <- c(11001100100, 11001100300, 11001100400,
                11001100501, 11001100502, 11001100600)
random_stuff <- c(2, 5, 6, 2, 5, 3)
data_frame <- data.frame(random_num, random_stuff)

data_frame$result <- substring(data_frame$random_num, nchar(data_frame$random_num)-6+1)


> data_frame
   random_num random_stuff result
1 11001100100            2 100100
2 11001100300            5 100300
3 11001100400            6 100400
4 11001100501            2 100501
5 11001100502            5 100502
6 11001100600            3 100600

How to break each element in a dataset into its constituent parts?

2 Answers2