R/blockSize.R

In raster: Geographic Data Analysis and Modeling

# Author: Robert J. Hijmans
# Date : November 2009
# Version 0.9
# Licence GPL v3


blockSize <- function(x, chunksize, n=nlayers(x), minblocks=4, minrows=1) {

	n <- max(n, 1)
	if (missing(chunksize)) {
		bs <- .chunk()
	} else {
		bs <- chunksize
	}

	blockrows <- try(methods::slot(x@file, 'blockrows'), silent=TRUE)
	if (class(blockrows)[1] == 'try-error') {
		blockrows <- 1
	}
	blockrows <- max(blockrows, 1)


	nr <- nrow(x)
	size <- min(nr, max(1, floor(bs / (ncol(x) * n * 8))))
	# min number of chunks
	if (size > 1) {
		minblocks <- min(nr, max(1, minblocks))
		size <- min(ceiling(nr/minblocks), size)
	}
	size <- min(max(size, minrows), nr)
	size <- max(minrows, blockrows * round(size / blockrows))

	nb <- ceiling(nr / size)
	row <- (0:(nb-1))*size + 1
	nrows <- rep(size, length(row))

	dif = nb * size - nr
	nrows[length(nrows)] = nrows[length(nrows)] - dif

	return(list(row=row, nrows=nrows, n=nb))
}