Spatial map of vertices measures

[agila5]

Problem

At the moment there is no easy and automatic way to represent network measures estimated for the vertices.

Solution

The algorithm behind SpatialLinesNetwork() defines two attributes for the igraph object named x and y that represent the coordinates of the vertices:

stplanr/R/SpatialLinesNetwork.R

Lines 152 to 156 in bbff85b

	g <- igraph::graph(gdata$pts0, directed = FALSE)
	nodes <- s[gdata$upts + 1, ]
	g$x <- nodes[, 1] # x-coordinate vertex
	g$y <- nodes[, 2] # y-coordinate vertex
	g$n <- as.vector(table(gdata$pts0)) # nr of edges

We can use these coordinates to create an sf object of POINTS and add/plot igraph measures related to vertices. For example:

# packages
library(sf)
#> Linking to GEOS 3.8.0, GDAL 3.0.4, PROJ 6.3.1
library(stplanr)
library(igraph)
library(tmap)

# data
route_network_sf <- rnet_breakup_vertices(route_network_sf)
sln_sf <- SpatialLinesNetwork(route_network_sf)

# Estimate vertices measures
sln_vertices_strength <- strength(sln_sf@g)

# Rebuild the sf object
sln_vertices <- st_as_sf(
  data.frame(
    x = sln_sf@g$x, 
    y = sln_sf@g$y, 
    strength = sln_vertices_strength
  ), 
  coords = c("x", "y"), 
  crs = 4326
)

# Map
tm_shape(sln_sf@sl) + 
  tm_lines() + 
tm_shape(sln_vertices) + 
  tm_dots(size = "strength")

^{Created on 2020-05-19 by the reprex package (v0.3.0)}

IMO the next step is to 1) document this "feature" or 2) create a new function to automatise this process.

[Robinlovelace]

👍 I think that will greatly improve the usefulness of the class for relatively little new code.