Edzer Pebesma
> I would argue that changing the behavior to allow each pixel to be 'assigned' to multiple geometries probably is more in line with many use cases when by is...
In @alexyshr 's example above, ```r st_interpolate_aw(h_nh_c.stnoxy, mun$geometry, FALSE) ``` also works and gives you a proper mean or sum, similar to how `exactextractr` would do but for the polygon-polygon...
> The raster package will be deprecated in 2023 Could you pls copy that message here? I only see a message from a package it loads, `sp`, which mentions that...
No, from https://cran.r-project.org/web/packages/raster/index.html you can see that `raster` does not depend in any way on `maptools`, `rgdal` or `rgeos`.
For the comparison: `lwgeom` can also branch on spherical Earth, which would be more fair to compare against your spherical distances. I'm primarily interested in both packages giving identical (and...
`lwgeom` doesn't use any, but gets them through the crs passed; parameter values are obtained by `sf` from GDAL, and printed by ```r > sf::st_crs(st_sfc(st_point(), crs = 4326), parameters =...
This is the approach that also raster took. The disadvantages I see are: * it's unlikely that you will keep up with improvements in the libgeographic code base (which is...
Another grouping is found in table 1 of https://journal.r-project.org/archive/2018/RJ-2018-009/RJ-2018-009.pdf
Splitting packages sounds cool, but is a big headache when you want to release things to CRAN: typically changing things in one sub-pkg requires you to change them in all...