Rearranging character vectors in a row of a matrix

Question

Suppose M has the following structure:-

> dput(head(M))
structure(list(SV = structure(c(38L, 21L, 24L, 20L, 54L, 45L), .Label = c("X1 X2 X3", 
"X1 X2 X3 X10 X6 X9", "X1 X2 X3 X10 X8 X9", "X1 X2 X3 X10 X9 X8", 
"X1 X2 X3 X4", "X1 X2 X3 X4 X9 X7", "X1 X2 X3 X5", "X1 X2 X3 X5 X10", 
"X1 X2 X3 X5 X6 X4", "X1 X2 X3 X5 X9", "X1 X2 X3 X6", "X1 X2 X3 X6 X7", 
"X1 X2 X3 X7", "X1 X2 X3 X7 X8 X10", "X1 X2 X3 X8", "X1 X2 X3 X8 X4", 
"X1 X2 X3 X8 X9 X10 X7", "X1 X2 X3 X9", "X1 X2 X3 X9 X10", "X1 X2 X3 X9 X5", 
"X2 X1 X3", "X2 X1 X3 X10", "X2 X1 X3 X10 X5", "X2 X1 X3 X10 X9", 
"X2 X1 X3 X4", "X2 X1 X3 X4 X10", "X2 X1 X3 X4 X6", "X2 X1 X3 X4 X8", 
"X2 X1 X3 X5 X10 X4", "X2 X1 X3 X5 X4", "X2 X1 X3 X5 X7", "X2 X1 X3 X6", 
"X2 X1 X3 X6 X5 X7", "X2 X1 X3 X6 X8", "X2 X1 X3 X6 X8 X9", "X2 X1 X3 X7", 
"X2 X1 X3 X7 X10", "X2 X1 X3 X7 X4", "X2 X1 X3 X7 X5", "X2 X1 X3 X7 X9 X5", 
"X2 X1 X3 X8", "X2 X1 X3 X8 X10 X4", "X2 X1 X3 X8 X10 X5 X9", 
"X2 X1 X3 X8 X6", "X2 X1 X3 X9", "X2 X1 X3 X9 X5", "X2 X1 X3 X9 X5 X4", 
"X2 X1 X3 X9 X6 X7 X10", "X2 X1 X3 X9 X7", "X2 X3 X1 X5 X6", 
"X2 X3 X1 X8", "X3 X1 X2", "X3 X2 X1", "X3 X2 X1 X4"), class = "factor")), row.names = c(NA, 
6L), class = "data.frame")

It can be seen that in some rows they are not consecutive. I need the appearance in a consecutive order X1 X2 X3 X4 X5 X6 X7 in this order. How can I rearrange the rows of M

Just use `dput(head(M))` and copy/paste the output in your question — Sotos, Feb 11 '20 at 14:19
I have edited the question accordingly with the exact structure of ```M```. I need the elements of row to appear in a systematic manner ..first X1, then X2, X3, X4 etc.... — user642080, Feb 11 '20 at 14:46
Not quite a duplicate, but very closely related: [How to sort a character vector where elements contain letters and numbers?](https://stackoverflow.com/q/17531403/903061) — Gregor Thomas, Feb 11 '20 at 15:05

Sotos · Accepted Answer · 2020-02-11T15:07:23.980

You need to split, strip the numbers and order, i.e.

M$SV <- sapply(M$SV, function(i){i1 <- strsplit(as.character(i), ' ')[[1]];  
                                 i2 <- i1[order(as.numeric(gsub('\\D+', '', i1)))]; 
                                 paste(i2, collapse = ' ')})

which gives,

               SV
1  X1 X2 X3 X4 X7
2        X1 X2 X3
3 X1 X2 X3 X9 X10
4  X1 X2 X3 X5 X9
5     X1 X2 X3 X4
6     X1 X2 X3 X9

s_baldur · Answer 2 · 2020-02-11T15:05:15.860

1

M$SV <- as.character(M$SV)
M$SV <- sapply(
  strsplit(M$SV, " "),
  function(x) {
    y <- x[order(as.integer(sub("X", "", x, fixed = TRUE)))]
    paste(y, collapse = " ")
  }
)
M


#                SV
# 1  X1 X2 X3 X4 X7
# 2        X1 X2 X3
# 3 X1 X2 X3 X9 X10
# 4  X1 X2 X3 X5 X9
# 5     X1 X2 X3 X4
# 6     X1 X2 X3 X9

edited Feb 11 '20 at 15:05

answered Feb 11 '20 at 14:59

s_baldur

29,441
4
36
69

Isn't this the same as mine essentially? – Sotos Feb 11 '20 at 15:07
I guess, just different flavour. I think mine should be faster. – s_baldur Feb 11 '20 at 15:09
1

It should since you are not splitting each line – Sotos Feb 11 '20 at 15:09

score 0 · Answer 3 · answered Feb 11 '20 at 14:28

0

try this using transponse and apply:

M <- data.frame(a=c("z1","z1"),
                b=c("z5","z4"),
                c=c("z2","z3"),
                d=c("z6","z2"))

>t(apply(M, 1, sort))
     [,1] [,2] [,3] [,4]
[1,] "z1" "z2" "z5" "z6"
[2,] "z1" "z2" "z3" "z4"

answered Feb 11 '20 at 14:28

norimg

96
3

how can I do it for 2 different rows? and what does 1 mean in the apply function ? – user642080 Feb 11 '20 at 14:48
1

This will miss-sort when the numbers go above 9. See `sort(c("X1", "X2", "X10", "X11"))`. – Gregor Thomas Feb 11 '20 at 15:03

Rearranging character vectors in a row of a matrix

3 Answers3