In dis-organization/sp_dplyrexpt: Classes and Methods for Spatial Data

I'm very interested to see if we could seamlessly allow these tbl_df objects with existing code, it looks like we can as long as our tests are based on inherits and methods like dim/nrow etc. I'm also interested in next generation rewrites, and keen to chat about it soon. I'll try to get my thoughts down and share

Goals

• more consistent spatial classes
• bbox/extent must have a CRS
• CRS is wkt
• any dim-consistent object allowed (matrix, df, tbl_df)
• flexible build functions, focus on best-guess rather than strict

sp with dplyr

Basic tests.

```r library(sp) library(dplyr) df <- data.frame(x = 1:10, y = rnorm(10), a = letters[1:10]) d_f <- as_data_frame(df)

data(meuse.grid) meuse_grid <- as_data_frame(meuse.grid) m = SpatialPixelsDataFrame(points = meuse_grid[c("x", "y")], data = meuse_grid) meuse_pts <- SpatialPointsDataFrame(SpatialPoints(as.matrix(meuse_grid[, c("x", "y")])), meuse_grid) gridded(meuse_pts) <- TRUE fullgrid(meuse_pts) <- TRUE

new("SpatialPointsDataFrame", SpatialPoints(as.matrix(d_f[,1:2])), data = d_f) new("SpatialGridDataFrame", ) coordinates(d_f) <- c("x", "y")

library(dplyr) library(sp) library(foreign) d_f <- as_data_frame(read.dbf(system.file("vectors", "cities.dbf", package = "rgdal"), as.is = TRUE)) pts <- matrix(rnorm(nrow(d_f) * 2), ncol = 2)

sp_dplyr <- new("SpatialPointsDataFrame", SpatialPoints(pts), data = d_f) ```

dis-organization/sp_dplyrexpt documentation built on May 15, 2019, 8:52 a.m.