rreshape2

Reshape wide to long, preserving ID variable from rownames

I am trying to get the following data into long format in R:

testdata <- data.frame(rnorm(10),rnorm(10),rnorm(10))
rownames(testdata) <- paste0("ID",1:10) # Ids
colnames(testdata) <- c(2001,2002,2003) # Years
testdata

So the columns = time, the rows = IDs. Should not be too hard, but in all the examples I found it was the other way around. How can this be done in datatable or reshape or in any other of the popular dataframe packages? Thanks for any hints. I know one way by transposing my data, but this seems to be a rather inefficient way for this purpose.

Solution

  • Just to turn akrun's comment into a complete answer:

    library(data.table)
melt(setDT(testdata, keep.rownames = TRUE), "rn")

          rn variable       value
 1:  ID1     2001 -0.25265860
 2:  ID2     2001  0.50538399
 3:  ID3     2001  0.68216394
 4:  ID4     2001  0.62203871
 5:  ID5     2001  0.59297019
 6:  ID6     2001  0.69383842
 7:  ID7     2001  1.77900432
 8:  ID8     2001 -1.69010623
 9:  ID9     2001 -2.17762905
10: ID10     2001  0.61463127
11:  ID1     2002  0.42120060
12:  ID2     2002 -0.16148732
...