Generate an index of chunk offsets and sizes for iterating a raster

make_chunk_index() returns a matrix of xchunkoff, ychunkoff, xoff, yoff, xsize, ysize, xmin, xmax, ymin and ymax, i.e., indexing of potentially multi-block chunks defined on block boundaries for iterating I/O operations over a raster. The last four columns are geospatial coordinates of the bounding box. Note that class GDALRaster provides a method of the same name that is more convenient to use with a dataset object.

Usage

make_chunk_index(
  raster_xsize,
  raster_ysize,
  block_xsize,
  block_ysize,
  gt = c(0, 1, 0, 0, 0, 1),
  max_pixels = 0
)

Arguments

raster_xsize

Integer value giving the number of raster columns.

raster_ysize

Integer value giving the number of raster rows.

block_xsize

Integer value giving the horizontal size of a raster block in number of pixels.

block_ysize

Integer value giving the vertical size of a raster block in number of pixels.

gt

A numeric vector of length six containing the affine geotransform for the raster (defaults to c(0, 1, 0, 0, 0, 1)). Required only if geospatial bounding boxes of the chunks are needed in the output.

max_pixels

Numeric value (a whole number), optionally carrying the bit64::integer64 class attribute. Specifies the maximum number of pixels per chunk. Can be set to zero to define chunks as the blocks.

Value

A numeric matrix with number of rows equal to the number of chunks, and named columns: xchunkoff, ychunkoff, xoff, yoff, xsize, ysize, xmin, xmax, ymin, ymax. Offsets are 0-based.

Details

The stand-alone function here supports the general case of chunking/tiling a raster layout without using a dataset object. If the max_pixels argument is set to zero, the chunks are raster blocks (the internal tiles in the case of a tiled format). Otherwise, chunks are defined as the maximum number of consecutive whole blocks containing <= max_pixels, that may span one or multiple whole rows of blocks.

See also

Methods make_chunk_index(), readChunk() and writeChunk() in class GDALRaster.

Usage example in the web article GDAL Block Cache.

Examples

## chunk as one block
blocks <- make_chunk_index(raster_xsize = 156335, raster_ysize = 101538,
                           block_xsize = 256, block_ysize = 256,
                           gt = c(-2362395, 30, 0, 3267405, 0, -30),
                           max_pixels = 0)

nrow(blocks)
#> [1] 242567

head(blocks)
#>      xchunkoff ychunkoff xoff yoff xsize ysize     xmin     xmax    ymin
#> [1,]         0         0    0    0   256   256 -2362395 -2354715 3259725
#> [2,]         1         0  256    0   256   256 -2354715 -2347035 3259725
#> [3,]         2         0  512    0   256   256 -2347035 -2339355 3259725
#> [4,]         3         0  768    0   256   256 -2339355 -2331675 3259725
#> [5,]         4         0 1024    0   256   256 -2331675 -2323995 3259725
#> [6,]         5         0 1280    0   256   256 -2323995 -2316315 3259725
#>         ymax
#> [1,] 3267405
#> [2,] 3267405
#> [3,] 3267405
#> [4,] 3267405
#> [5,] 3267405
#> [6,] 3267405

tail(blocks)
#>           xchunkoff ychunkoff   xoff   yoff xsize ysize    xmin    xmax   ymin
#> [242562,]       605       396 154880 101376   256   162 2284005 2291685 221265
#> [242563,]       606       396 155136 101376   256   162 2291685 2299365 221265
#> [242564,]       607       396 155392 101376   256   162 2299365 2307045 221265
#> [242565,]       608       396 155648 101376   256   162 2307045 2314725 221265
#> [242566,]       609       396 155904 101376   256   162 2314725 2322405 221265
#> [242567,]       610       396 156160 101376   175   162 2322405 2327655 221265
#>             ymax
#> [242562,] 226125
#> [242563,] 226125
#> [242564,] 226125
#> [242565,] 226125
#> [242566,] 226125
#> [242567,] 226125

## chunk as 16 consectutive blocks
chunks <- make_chunk_index(raster_xsize = 156335, raster_ysize = 101538,
                           block_xsize = 256, block_ysize = 256,
                           gt = c(-2362395, 30, 0, 3267405, 0, -30),
                           max_pixels = 256 * 256 * 16)

nrow(chunks)
#> [1] 15483

head(chunks)
#>      xchunkoff ychunkoff  xoff yoff xsize ysize     xmin     xmax    ymin
#> [1,]         0         0     0    0  4096   256 -2362395 -2239515 3259725
#> [2,]         1         0  4096    0  4096   256 -2239515 -2116635 3259725
#> [3,]         2         0  8192    0  4096   256 -2116635 -1993755 3259725
#> [4,]         3         0 12288    0  4096   256 -1993755 -1870875 3259725
#> [5,]         4         0 16384    0  4096   256 -1870875 -1747995 3259725
#> [6,]         5         0 20480    0  4096   256 -1747995 -1625115 3259725
#>         ymax
#> [1,] 3267405
#> [2,] 3267405
#> [3,] 3267405
#> [4,] 3267405
#> [5,] 3267405
#> [6,] 3267405

tail(chunks)
#>          xchunkoff ychunkoff   xoff   yoff xsize ysize    xmin    xmax   ymin
#> [15478,]        33       396 135168 101376  4096   162 1692645 1815525 221265
#> [15479,]        34       396 139264 101376  4096   162 1815525 1938405 221265
#> [15480,]        35       396 143360 101376  4096   162 1938405 2061285 221265
#> [15481,]        36       396 147456 101376  4096   162 2061285 2184165 221265
#> [15482,]        37       396 151552 101376  4096   162 2184165 2307045 221265
#> [15483,]        38       396 155648 101376   687   162 2307045 2327655 221265
#>            ymax
#> [15478,] 226125
#> [15479,] 226125
#> [15480,] 226125
#> [15481,] 226125
#> [15482,] 226125
#> [15483,] 226125