julia_benchmarking.julia

using CSV
using DataFrames
using Statistics
using Dates

dataset_small = CSV.read("data/dataset_small.csv", DataFrame)
dataset_medium = CSV.read("data/dataset_medium.csv", DataFrame)
dataset_large = CSV.read("data/dataset_large.csv", DataFrame)
secondary_dataset = CSV.read("data/join_dataset.csv", DataFrame)

function sort_dataset(dataset::DataFrame)
    sort(dataset, :Value, rev=true)
end

function filter_dataset(dataset::DataFrame)
    filter(row -> row[:Category] in ['A', 'B'], dataset)
end

function aggregate_dataset(dataset::DataFrame)
    combine(groupby(dataset, :Category), :Value => mean)
end

function join_datasets(main_dataset::DataFrame, secondary_dataset::DataFrame)
    innerjoin(main_dataset, secondary_dataset, on=:ID)
end


function benchmark_task(task_function, dataset::DataFrame, secondary_dataset::DataFrame=DataFrame())
    start_time = time()
    if ncol(secondary_dataset) == 0
        result = task_function(dataset)
    else
        result = task_function(dataset, secondary_dataset)
    end
    end_time = time()
    end_time - start_time
end
results = DataFrame(Task = String[], Size = String[], Time = Float64[], Language = String[])

for (dataset, size) in zip([dataset_small, dataset_medium, dataset_large], ["Small", "Medium", "Large"])
    push!(results, ["Sort", size, benchmark_task(sort_dataset, dataset), "Julia"])
    push!(results, ["Filter", size, benchmark_task(filter_dataset, dataset), "Julia"])
    push!(results, ["Aggregate", size, benchmark_task(aggregate_dataset, dataset), "Julia"])
    push!(results, ["Join", size, benchmark_task(join_datasets, dataset, secondary_dataset), "Julia"])
end

CSV.write("data/benchmark_results_julia.csv", results)