How can I reorder the rows of a matrix, data.frame or vector according to another one

Question

test1 <- as.matrix(c(1, 2, 3, 4, 5))
row.names(test1) <- c("a", "d", "c", "b", "e") 

test2 <- as.matrix(c(6, 7, 8, 9, 10))
row.names(test2) <- c("e", "d", "c", "b", "a") 

  test1
  [,1]
a    1
d    2
c    3
b    4
e    5

 test2
  [,1]
e    6
d    7
c    8
b    9
a   10

How can I reorder test2 so that the rows are in the same order as test1? e.g.

 test2
  [,1]
a    10
d    7
c    8
b    9
e    6

I tried to use the reorder function with: reorder (test1, test2) but I could not figure out the correct syntax. I see that reorder takes a vector, and I'm here using a matrix. My real data has one character vector and another as a data.frame. I figured that the data structure would not matter too much for this example above, I just need help with the syntax and can adapt it to my real problem.

Solution 1

test2 <- test2[rownames(test1),,drop=FALSE]

Solution 2

After fixing your code snipped to actually generate what your example shows (hint: test1 had names a,b,c,d,e; you meant a,d,c,b,1 as it shows now), this was easier thanks to match():

R> test2[match(row.names(test2), row.names(test1)),1,drop=FALSE]
  [,1]
a   10
d    7
c    8
b    9
e    6
R> 

They key here is that match() does what you want:

R> match(row.names(test2), row.names(test1))
[1] 5 2 3 4 1

Comments

[chimpsarehungry]

Why does this not work for my data? I want to do test2[match(test2$column, test1$column),1,drop=FALSE] because what i am matching by is the values of the columns not the row.names

[chimpsarehungry]

Why does this not work for my data? I want to do test2[match(test2$column, test1$column),1,drop=FALSE] because what i am matching by is the values of the columns not the row.names