Here's a simple import of a CSV file using the Tony module's exported
cities() function and a test plot using the plotcities() function.
See also these couple links to the DataFrames.jl and CSV.jl documentation pages.
Note that best shot for Postgres support right now seems to be LibPQ.jl – a Julia wrapper for the PostgreSQL libpq C library.
% julia --project
_
_ _ _(_)_ | Documentation: https://docs.julialang.org
(_) | (_) (_) |
_ _ _| |_ __ _ | Type "?" for help, "]?" for Pkg help.
| | | | | | |/ _` | |
| | |_| | | | (_| | | Version 1.0.3 (2018-12-18)
_/ |\__'_|_|_|\__'_| | Official https://julialang.org/ release
|__/ |
julia> using Tony
[ Info: Recompiling stale cache file /Users/tony/.julia/compiled/v1.0/Tony/iHaKB.ji for Tony [d8886264-044d-11e9-15c8-31bd3d336b1a]
julia> cities()
3228×4 DataFrames.DataFrame
│ Row │ name │ population │ latitude │ longitude │
│ │ Union{Missing, String} │ Int64⍰ │ Float64⍰ │ Float64⍰ │
├──────┼─────────────────────────┼────────────┼──────────┼───────────┤
│ 1 │ New York │ 8287238 │ 40.7306 │ -73.9866 │
│ 2 │ Los Angeles │ 3826423 │ 34.0537 │ -118.243 │
│ 3 │ Chicago │ 2705627 │ 41.8756 │ -87.6244 │
│ 4 │ Houston │ 2129784 │ 29.7589 │ -95.3677 │
│ 5 │ Philadelphia │ 1539313 │ 39.9523 │ -75.1638 │
...
│ 3224 │ Spanish Fort │ 7102 │ 30.6749 │ -87.9153 │
│ 3225 │ Plaquemine │ 7102 │ 30.2891 │ -91.2343 │
│ 3226 │ Milton-Freewater │ 7102 │ 45.9326 │ -118.388 │
│ 3227 │ Benton │ 7096 │ 36.0345 │ -88.1013 │
│ 3228 │ Ocean City │ 7094 │ 39.2776 │ -74.5746 │
julia> names(cities())
4-element Array{Symbol,1}:
:name
:population
:latitude
:longitude
julia>
And we can try this out with this plotcities() function:
function plotcities()
# read cities into a dataframe
df = cities()
# scatter plots
df1 = filter(row -> row[:population] > 900_000, df)
df2 = filter(row -> row[:population] <= 900_000, df)
df3 = filter(row -> haskey(places, row[:name]) , df)
# scatter plot of lat/long - for df2 cities (below 0.9m pop.)
p0 = scatter(df2[:longitude], df2[:latitude],
title = "US Cities",
xlabel = "Longitude",
ylabel = "Latitude",
legend = false,
markershape = :circle,
markercolor = :auto,
# markersize = 4
)
# scatter plot of lat/long - for df3 cities (visited)
scatter!(df3[:longitude], df3[:latitude],
legend = false,
markershape = :circle,
markercolor = :yellow,
markersize = 10
)
# scatter plot of lat/long - for df1 cities (above 0.9m pop.)
scatter!(df1[:longitude], df1[:latitude],
legend = false,
markershape = :rect,
markercolor = :orange,
# markersize = 4
)
savefig("images/plotcities-latlong.png")
# line plots with layout
c = df[1:10, :]
d = df[11:20, :]
p1 = plot([c[:population]],
title = "US Cities (1-10)",
xlabel = "City",
ylabel = "Population",
xticks = (1:10, c[:name]),
xrotation = 60,
legend = false,
markershape = :rect,
markercolor = :orange
)
p2 = plot([d[:population]],
title = "US Cities (11-20)",
xlabel = "City",
ylabel = "Population",
xticks = (1:10, d[:name]),
xrotation = 60,
legend = false,
markershape = :auto,
markercolor = :auto
)
plot(p1,p2,layout=2,legend=false)
savefig("images/plotcities-layout.png")
endThis gives the following plots: a) a scatter plot of US city lat/long positions (orange markers show major cities with pop. of over 0.9m, yellow markers are places I have visited), and b) line plots for US city population distribution
which is comprised of two subplots using the layout composition:
