Visualizing roads in the cities of Canada

Keywords

julia

Let’s import some packages first.

using GeoMakie
using GeoInterfaceMakie
using GeoInterface
using CairoMakie
using Shapefile
using DataFrames
using DataFramesMeta
using StringEncodings
using Pkg.Artifacts

If you want to run this interactively, you can replace CairoMakie with GLMakie, i.e.

- import CairoMakie
+ import GLMakie

Data

Representations of Canada’s national road network are available from Statistics Canada.

artifact_roadnetwork = Pkg.Artifacts.ensure_artifact_installed("roadnetwork", joinpath(@__DIR__, "Artifacts.toml"))
path = joinpath(artifact_roadnetwork, "lrnf000r21a_e.shp")
@time gdf = DataFrame(Shapefile.Table(path));
@show size(gdf)
first(gdf, 1)

 45.218459 seconds (116.50 M allocations: 7.752 GiB, 3.89% gc time, 2.02% compilation time)
size(gdf) = (2242117, 26)

1×26 DataFrame

Row	geometry	OBJECTID	NGD_UID	NAME	TYPE	DIR	AFL_VAL	ATL_VAL	AFR_VAL	ATR_VAL	CSDDGUID_L	CSDUID_L	CSDNAME_L	CSDTYPE_L	CSDDGUID_R	CSDUID_R	CSDNAME_R	CSDTYPE_R	PRDGUID_L	PRUID_L	PRNAME_L	PRDGUID_R	PRUID_R	PRNAME_R	RANK	CLASS
	Polyline	Int64	String	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String?	String	String?
1	Polyline(Rect(7.65014e6, 1.27149e6, 7.65038e6, 1.2717e6), Int32[0], Shapefile.Point[Point(7.65014e6, 1.27149e6), Point(7.65017e6, 1.2715e6), Point(7.6502e6, 1.27152e6), Point(7.65023e6, 1.27153e6), Point(7.65024e6, 1.27154e6), Point(7.65027e6, 1.27156e6), Point(7.6503e6, 1.27158e6), Point(7.65031e6, 1.27159e6), Point(7.65033e6, 1.27162e6), Point(7.65035e6, 1.27164e6), Point(7.65036e6, 1.27166e6), Point(7.65037e6, 1.27169e6), Point(7.65038e6, 1.2717e6)])	1	5792582	des 60	RANG	missing	195	195	182	194	2021A00052457050	2457050	Saint-Marc-sur-Richelieu	M\xc9	2021A00052457050	2457050	Saint-Marc-sur-Richelieu	M\xc9	2021A000224	24	Quebec / Qu\xe9bec	2021A000224	24	Quebec / Qu\xe9bec	4	23

The documentation says CSDNAME is the Census subdivision name, which seems to map to cities.

Let’s convert it to a proper encoding first:

latin1_to_utf8(s) = decode(Vector{UInt8}(String(coalesce(s, ""))), "Windows-1252")
@time @rtransform! gdf begin
  :CSDNAME_L_UTF8 = latin1_to_utf8(:CSDNAME_L)
  :CSDNAME_R_UTF8 = latin1_to_utf8(:CSDNAME_R)
end

 17.346026 seconds (111.19 M allocations: 9.443 GiB, 17.28% gc time, 5.29% compilation time: 17% of which was recompilation)

Visualizations

We can now create a plot for each city using Makie:

function plot_city(gdf, city_name)
  df = @rsubset gdf (:CSDNAME_L_UTF8 == city_name || :CSDNAME_R_UTF8 == city_name)
  empty_theme = Theme(
    fonts=(; weird="Blackchancery"),
    fontsize=32,
    Axis=(
      backgroundcolor=:transparent,
      leftspinevisible=false,
      rightspinevisible=false,
      bottomspinevisible=false,
      topspinevisible=false,
      xticklabelsvisible=false,
      yticklabelsvisible=false,
      xgridcolor=:transparent,
      ygridcolor=:transparent,
      xminorticksvisible=false,
      yminorticksvisible=false,
      xticksvisible=false,
      yticksvisible=false,
      xautolimitmargin=(0.0, 0.0),
      yautolimitmargin=(0.0, 0.0),
      titlefont=:weird,
    ),
  )
  with_theme(empty_theme) do
    fig = Figure()
    ax = Axis(fig[1, 1])
    poly!.(GeoInterface.convert.(Ref(CairoMakie.GeometryBasics), df[:, :geometry]); strokecolor=:black, color=:black)
    ax.title = city_name
    fig
  end
end;

plot_city(gdf, "Toronto")

plot_city(gdf, "Montréal")

plot_city(gdf, "Vancouver")

plot_city(gdf, "Ottawa")

plot_city(gdf, "Calgary")

plot_city(gdf, "Edmonton")

plot_city(gdf, "Winnipeg")