README.md

In hypertidy/granulated: Define Rectangular Regions with Granularity

granulated

The goal of granulated is to provide the basic cell,raster,grid-abstraction tools I’ve become accustomed to with the raster package. I.e. there’s nothing interesting here, it’s very basic core functionality and doesn’t work yet.

There are no examples or tests … this project has floated around for a while but is under active development, stay tuned.

There’s a dummy class “grain” which acts as a placeholder to get all the functionality working, since R has no idea about a raster grid with extent apart from implicit functionality in graphics::image() and graphics::rasterImage().

The very new and experimental grd package provides a more sensible basis for proving this on, and work is currently aimed at that. The raster package has an uncertain future, and no other project we’re aware of aims to capture this functionality that sits somewhere between generic array data and bitmaps in R.

Motivations

• raster (or bitmap) grids in R don’t know about spatial extent although helpers exist in image(),rasterImage()
• native rasters (as.raster(), and the internal ‘nativeRaster’) also don’t have plot helpers for value-scaling (bytes or raw rgb encoding are assumed)
• I have become accustomed to treating raster data in the abstract, there is properties dimension, extent, projection and most work revolves around that - in raster package, data is stored (when it’s materialized) in flat column form, almost never does the actual dimensionality impact the storage of the data
• the raster package provides excellent support for almost everything here (!) but, it is subject to the turbulence and uncertain future of sp et al. 

… grd and granulated aim to capture that abstracted space and the missing pieces in R and we’ll work out the spaces in between.

Despite some heroic efforts (raster, stars, ggspatial, lazyraster, grd, rayvista, …) there’s no easy canvas support in R, ‘here’s my domain fill it with pixels’ and we can do that with GDAL’s warper (e.g. vapour::vapour_warp_raster()) but we need these basic-er tools to enrich the space (because otherwise we are constantly juggling objects, formats, and plumbing when all we really need is dimension, extent, projection as a simple framework). Please see hypertidy/gdalio for some exploration of that idea.

Code of Conduct

Please note that the granulated project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

hypertidy/granulated documentation built on Dec. 20, 2021, 5:54 p.m.