Returning specific values within a row

user_12233826 0
Viewed 60

I have 1 row of data and 50 columns in the row from a csv which I've put into a dataframe. The data is arranged across the spreadsheet like this:

"FSEG-DFGS-THDG", "SGDG-SGRE-JJDF", "DIDC-DFGS-LEMS"...

How would I select only the middle part of each element (eg, "DFGS" in the 1st one, "SGRE" in the second etc), count their occurances and display the results?

I have tried using the strsplit function but I couldn't get it to work for the entire row of data. I'm thinking a loop of some kind might be what I need

r
2 Answers

You can do unlist(strsplit(x, '-'))[seq(2, length(x)*3, 3)] (assuming your data is consistently of the form A-B-C).

# E.g.
fun <- function(x) unlist(strsplit(x, '-'))[seq(2, length(x)*3, 3)]
fun(c("FSEG-DFGS-THDG", "SGDG-SGRE-JJDF", "DIDC-DFGS-LEMS"))
# [1] "DFGS" "SGRE" "DFGS"

Edit

# Data frame
df <- structure(list(a = "FSEG-DFGS-THDG", b = "SGDG-SGRE-JJDF", c = "DIDC-DFGS-LEMS"), 
                class = "data.frame", row.names = c(NA, -1L))
fun(t(df[1,]))
# [1] "DFGS" "SGRE" "DFGS"
user_85065480

First we create a function strng() and then we apply() it on every column of df. strsplit() splits a string by "-" and strng() returns the second part.

df = data.frame(a = "ab-bc-ca", b = "gn-bc-ca", c = "kj-ll-mn")
strng = function(x) {
  strsplit(x,"-")[[1]][2]
}

# table() outputs frequency of elements in the input
table(apply(df, MARGIN = 2, FUN = strng))

# output: bc ll 
          2  1
user_31725560
Related