Getting and setting virtual windows

The virtual window vw function allows one to define a virtual window into an ff_vector or ff_array. The ff object will behave like a smaller array and it is mapped into the specified region of the complete array. This allows for example to execute recursive divide and conquer algorithms that work on parts of the full object, without the need to repeatedly create subfiles.

Usage

vw(x, ...)
vw(x, ...) <- value
# S3 method for class 'ff'
vw(x, ...)
# Default S3 method
vw(x, ...)
# S3 method for class 'ff_vector'
vw(x, ...) <- value
# S3 method for class 'ff_array'
vw(x, ...) <- value

Arguments

x: an ff_vector or ff_array
...: further arguments (not used)
value: a vector or matrix with an Offset, Window and Rest component, see details and examples

Details

Each dimension of an ff array (or vector) is decomposed into three components, an invisible Offset, a visibe Window and an invisible Rest. For each dimension the sum of the vw components must match the dimension (or length). For an ff_vector, vw is simply a vector[1:3], for an array is is a matrix[1:3,seq_along(dim(x))]. vw is a virtual attribute.

Value

NULL or a vw specification, see details

Author

Jens Oehlschlägel

Examples

  x <- ff(1:26, names=letters)
  y <- x
  vw(x) <- c(0, 13, 13)
  vw(y) <- c(13, 13, 0)
  x
#> ff (open) integer offset=0 length=13 totallength=26 (26)
#>  a  b  c  d  e  f  g     h  i  j  k  l  m 
#>  1  2  3  4  5  6  7  :  8  9 10 11 12 13 
  y
#> ff (open) integer offset=13 length=13 totallength=26 (26)
#>  n  o  p  q  r  s  t     u  v  w  x  y  z 
#> 14 15 16 17 18 19 20  : 21 22 23 24 25 26 
  x[1] <- -1
  y[1] <- -2
  vw(x) <- NULL
  x[]
#>  a  b  c  d  e  f  g  h  i  j  k  l  m  n  o  p  q  r  s  t  u  v  w  x  y  z 
#> -1  2  3  4  5  6  7  8  9 10 11 12 13 -2 15 16 17 18 19 20 21 22 23 24 25 26 

  z <- ff(1:24, dim=c(4,6), dimnames=list(letters[1:4], LETTERS[1:6]))
  z
#> ff (open) integer length=24 (24) dim=c(4,6) dimorder=c(1,2)
#>    A  B  C  D  E  F
#> a  1  5  9 13 17 21
#> b  2  6 10 14 18 22
#> c  3  7 11 15 19 23
#> d  4  8 12 16 20 24
  vw(z) <- rbind(c(1,1), c(2,4), c(1,1))
  z
#> ff (open) integer length=8 (24) offset=c(1,1) dim=c(2,4) totaldim=c(4,6) dimorder=c(1,2)
#>    B  C  D  E
#> b  6 10 14 18
#> c  7 11 15 19

  rm(x,y,z); gc()
#>           used (Mb) gc trigger  (Mb) max used  (Mb)
#> Ncells 1272019   68    2239654 119.7  2239654 119.7
#> Vcells 2346474   18    8388608  64.0  8387257  64.0