4

I have a table called "pipel" that contains more than 10,000 rows. I would like to add an ID column to assign a unique ID for each row. The unique ID must be 30 digits long and starts with "AKM_CC_Test_". I used the code below as a starting point but not sure how to format it to add the prefix and make it 30 digits long.

id <- rownames(pipel)
pipel <- cbind(id=id, pipel)

For example first row ID will need to look like this AKM_CC_Test_000000000000000001

Curious
  • 549
  • 6
  • 23

2 Answers2

7

You could use sprintf(). This creates a 30 character string beginning with "AKM_CC_Test_" and ending in a sequence of 1:nrow(pipel) with leading zeros.

x <- "AKM_CC_Test_"
sprintf("%s%0*d", x, 30 - nchar(x), 1:nrow(pipel))
  • %s inserts x into the string
  • %0*d adds 1:nrow(pipel) with * leading zeros, after x. The * is used to insert 30 - nchar(x) into the format (I did it programatically; you could just insert 18 there if you want)

An example on a simple length 5 (1:5) vector would be 

x <- "AKM_CC_Test_"
sprintf("%s%0*d", x, 30 - nchar(x), 1:5)
# [1] "AKM_CC_Test_000000000000000001" "AKM_CC_Test_000000000000000002"
# [3] "AKM_CC_Test_000000000000000003" "AKM_CC_Test_000000000000000004"
# [5] "AKM_CC_Test_000000000000000005"
Rich Scriven
  • 97,041
  • 11
  • 181
  • 245
  • Thanks the code above worked and created the IDs I was looking for but when I tried to write the new table using write.table(pipel, "pipel9.txt", sep=" ",quote = FALSE, col.names=TRUE) the ID column didn't show in the table. – Curious Oct 10 '16 at 00:43
  • 1
    @Curious - what code did you use to add the column? – Rich Scriven Oct 10 '16 at 07:09
  • @ Rich Scriven- I tried to use: newtable<- cbind(id=X, pipel) – Curious Oct 10 '16 at 23:15
1

You can use : or seq for sequences and you can prepend your leading text with paste or paste0. The heart of the question is on the number padding with leading 0's.

Your options are:

  1. stri_pad from stringi (more intuitive)
  2. str_pad from stringr (more intuitive)
  3. sprintf (no packages needed)
  4. formatC (good if you're familiar with C's printf)

Note that some cases, though not this particular one, necessitate disabling scientific notation for numbers in the sequence. This can be done with options or with_options from devtools.

Please see this popular post for examples of each.

Using formatC:

uid <- paste0("AKM_CC_Test_", formatC(1:10000, width = 18, format = "d", flag = "0"))
head(uid)

[1] "AKM_CC_Test_000000000000000001" "AKM_CC_Test_000000000000000002" "AKM_CC_Test_000000000000000003" "AKM_CC_Test_000000000000000004"
[5] "AKM_CC_Test_000000000000000005" "AKM_CC_Test_000000000000000006"

Using the stringr package:

uid <- paste0("AKM_CC_Test_", str_pad(1:10000, 18, pad = "0")) # remember to load stringr
head(uid)

[1] "AKM_CC_Test_000000000000000001" "AKM_CC_Test_000000000000000002" "AKM_CC_Test_000000000000000003" "AKM_CC_Test_000000000000000004"
[5] "AKM_CC_Test_000000000000000005" "AKM_CC_Test_000000000000000006"

Using sprintf:

head(sprintf("%s%0*d", "AKM_CC_Test_", 18,  1:10000))

[1] "AKM_CC_Test_000000000000000001" "AKM_CC_Test_000000000000000002" "AKM_CC_Test_000000000000000003" "AKM_CC_Test_000000000000000004"
[5] "AKM_CC_Test_000000000000000005" "AKM_CC_Test_000000000000000006"

Using stri_pad from the package stringi:

uid <- paste0("AKM_CC_Test_", stri_pad(1:10000, 18, pad = "0")) # remember to load stringi
head(uid)

[1] "AKM_CC_Test_000000000000000001" "AKM_CC_Test_000000000000000002" "AKM_CC_Test_000000000000000003" "AKM_CC_Test_000000000000000004"
[5] "AKM_CC_Test_000000000000000005" "AKM_CC_Test_000000000000000006"
Community
  • 1
  • 1
Hack-R
  • 22,422
  • 14
  • 75
  • 131
  • 3
    Why did you add my answer to yours? – Rich Scriven Oct 09 '16 at 20:22
  • 1
    @RichScriven What are you talking about? I using the list of padding options on the linked post and I did it differently and comparing/contrasting them. It's just a function Rich, it's not your answer. – Hack-R Oct 09 '16 at 20:23
  • 1
    Oh get real. It's the same thing. – Rich Scriven Oct 09 '16 at 20:24
  • 2
    @RichScriven Wow. All I can say is wow. I answer ahead of you, link a source with a list of options, and apply all of them one by one and you're going to whine and downvote because an option you used is on the famous list? Are there any others I should know are off limits? Can I still use `<-` and `paste`? – Hack-R Oct 09 '16 at 20:25