Create a toy sfnetwork with arbitrary complexity

[agila5]

The following code can be used to create a toy sfnetwork object with an arbitrary degree of complexity (i.e. number of nodes and edges):

# packages
library(tidygraph)
library(sfnetworks)

# data
set.seed(1)
my_graph <- play_geometry(n = 200, radius = 0.1)
my_sfn <- as_sfnetwork(my_graph, coords = c("x", "y"))
#> Checking if spatial network structure is valid...
#> Spatial network structure is valid

# plot
par(mar = rep(0, 4))
plot(my_sfn)

# a bigger example 
my_graph <- play_geometry(1e6, 1.5e-3)
my_sfn <- as_sfnetwork(my_graph, coords = c("x", "y"))
#> Checking if spatial network structure is valid...
#> Spatial network structure is valid
my_sfn
#> # A sfnetwork with 1000000 nodes and 3526983 edges
#> #
#> # CRS:  NA 
#> #
#> # An undirected simple graph with 1499 components with spatially implicit edges
#> #
#> Registered S3 method overwritten by 'cli':
#>   method     from         
#>   print.boxx spatstat.geom
#> # Node Data:     1,000,000 x 1 (active)
#> # Geometry type: POINT
#> # Dimension:     XY
#> # Bounding box:  xmin: 5.243346e-07 ymin: 1.098961e-07 xmax: 0.9999984 ymax:
#> #   0.9999998
#>                   geometry
#>                    <POINT>
#> 1   (5.243346e-07 0.18507)
#> 2 (1.089182e-06 0.8978485)
#> 3 (1.522014e-06 0.7236908)
#> 4 (1.667067e-06 0.7810175)
#> 5 (2.781628e-06 0.9403087)
#> 6 (3.511552e-06 0.3574057)
#> # ... with 999,994 more rows
#> #
#> # Edge Data: 3,526,983 x 2
#>    from    to
#>   <int> <int>
#> 1     2   242
#> 2     2   446
#> 3     2  1064
#> # ... with 3,526,980 more rows

^{Created on 2021-09-03 by the reprex package (v2.0.1)}

Clearly, the toy objects are not realistic simulations for street or river networks, but I think they could be useful for running (not 100% accurate) benchmarks on large graphs without loading/downloading OSM data. Slightly related to #79.

[luukvdmeer]

In v1.0, this is now implemented in as play_geometric()