How to create columns in R Dataframe based on characters stored in a column?

Question

I am looking to create new columns in an R data frame based on characters stored in an existing column. For example, suppose I have the following data frame:

> df = data.frame(retroid = c("loftk001", "vizq001"), pitchcount = c("BBBCCFB", "CCX"))
> df
   retroid pitchcount
1 loftk001    BBBCCFB
2  vizq001        CCX

I want to create new columns, "p1, p2, p3", etc. such that it looks like

   retroid pitchcount p1 p2 p3 p4 p5 p6 p7
1 loftk001    BBBCCFB B  B  B  C  C  F  B
2  vizq001        CCX C  C  X

One of the potential problems I am facing is that the strings stored under column "pitchcount" are of variable length. As in the case above, if the string in "pitchcount" has less characters than the maximum, I just want empty values in the corresponding columns.

Is there a fast way to do this in R?

Thank you in advance!

Rich Scriven · Accepted Answer · 2014-11-19 19:33:53Z

3

You could use cSplit, where df is your original data.

library(splitstackshape)

cs <- cSplit(df[2], "pitchcount", "", stripWhite=FALSE, type.convert=FALSE)
setnames(cs, names(cs), sub(".*_", "p", names(cs)))
cbind(df, cs)
#    retroid pitchcount p1 p2 p3   p4   p5   p6   p7
# 1 loftk001    BBBCCFB  B  B  B    C    C    F    B
# 2  vizq001        CCX  C  C  X <NA> <NA> <NA> <NA>

Another option is to use the new stri_list2matrix function from stringi

library(stringi)

ss <- strsplit(as.character(df$pitchcount), "")
cbind(df, stri_list2matrix(ss, byrow = TRUE, fill = ""))
#    retroid pitchcount 1 2 3 4 5 6 7
# 1 loftk001    BBBCCFB B B B C C F B
# 2  vizq001        CCX C C X

And then you could just paste "p" onto the front of the new columns' names

edited Nov 19, 2014 at 19:33

answered Nov 19, 2014 at 19:16

Rich Scriven

99.8k11 gold badges191 silver badges252 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

user3603486 · Accepted Answer · 2014-11-19 20:01:38Z

0

Another, perhaps a bit clearer:

df <- cbind(df, matrix(nrow=20)) # or however many columns you need

sapply(1:nrow(df), function(x) {
  pc <- df$pitchcount[x]
  y <- 
  df[x, 2 + seq_along(nchar(pc))] <<- strsplit(pc, split="")[[1]]
})

answered Nov 19, 2014 at 20:01

user3603486

Collectives™ on Stack Overflow

How to create columns in R Dataframe based on characters stored in a column?

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related