R help - Jul 2006 - Inflated Array

Hi R-users!

I am trying to create a what I call inflated array (maybe there is
already some other name for that). It is an array that changes
dinamically its dimensions, e.g. the higher the number of third
dimensions, the more rows in the array. So for example the array could
look like the following:

, ,1

1 2 3 4

, , 2

5   6   7  8
9 10 11 12

, , 3

13 14 15 16
17 18 19 20
21 22 23 24

Has anybody a comment on how to build this in the most efficient
manner (I know i could do looping)?

Thanks!

Hadassa

Hadassa Brunschwig <dassybr <at> gmail.com> writes:
> 
> Hi R-users!
> 
> I am trying to create a what I call inflated array (maybe there is
> already some other name for that). It is an array that changes
> dinamically its dimensions, e.g. the higher the number of third
> dimensions, the more rows in the array. 
  base R doesn't have a data structure for this: arrays in R
must not be "ragged" (i.e., every sub-array must have the
same dimensions).

  So you would need to use a list of matrices in your
example.  I don't quite know the logic that you're using
to decide what goes in each sub-table.

The basic example you give could be built 
by hand as follows:

list(1:4,matrix(5:12,byrow=TRUE,ncol=4),
         matrix(13:24,byrow=TRUE,ncol=4))

  if you had used scan() or something else to
get a long, flat vector (x) and also had
a vector (v) that indicated the number of
rows in each sub-table:

x <- 1:24
nrow <- c(1,2,3)
ncol <- 4

ind <- rep(1:length(nrow),ncol*nrow)
lapply(split(x,ind),
    matrix,ncol=ncol,byrow=TRUE)

  seems to work.

  cheers
   Ben Bolker