# Process csv data into an RData file to be loaded by the shiny application
library(readr)
library(dplyr)
education_travel <- read_csv("travel-education.csv")
work_travel <- read_csv("travel-work.csv")

work_travel %>% select(
  res_code = SA2_code_usual_residence_address,
  res_name = SA2_name_usual_residence_address,
  res_east = SA2_usual_residence_easting,
  res_north = SA2_usual_residence_northing,
  work_code = SA2_code_workplace_address,
  work_name = SA2_name_workplace_address,
  work_east = SA2_workplace_easting,
  work_north = SA2_workplace_northing,
  private = Drive_a_private_car_truck_or_van,
  passenger = Passenger_in_a_car_truck_van_or_company_bus,
  walk = Walk_or_jog,
  bicycle = Bicycle,
  company = Drive_a_company_car_truck_or_van,
  bus = Public_bus,
  train = Train,
  ferry = Ferry,
  other = Other,
  home = Work_at_home,
  total = Total
) -> work_simp

work_simp %>% 
  group_by(res_code,
                         res_name,
                         res_east,
                         res_north) %>%
  summarise(
    private = sum(ifelse(private < 0, 0, private)),
    passenger = sum(ifelse(passenger < 0, 0, passenger)),
    walk = sum(ifelse(walk < 0, 0, walk)),
    bicycle = sum(ifelse(bicycle < 0, 0, bicycle)),
    company = sum(ifelse(company < 0, 0, company)),
    bus = sum(ifelse(bus < 0, 0, bus)),
    train = sum(ifelse(train < 0, 0, train)),
    ferry = sum(ifelse(ferry < 0, 0, ferry)),
    other = sum(ifelse(other < 0, 0, other)),
    home = sum(ifelse(home < 0, 0, home)),
    total = sum(ifelse(total < 0, 0, total)), .groups="drop"
  ) -> work_from
    

work_simp %>% 
  group_by(work_code,
                         work_name,
                         work_east,
                         work_north) %>%
  summarise(
    private = sum(ifelse(private < 0, 0, private)),
    passenger = sum(ifelse(passenger < 0, 0, passenger)),
    walk = sum(ifelse(walk < 0, 0, walk)),
    bicycle = sum(ifelse(bicycle < 0, 0, bicycle)),
    company = sum(ifelse(company < 0, 0, company)),
    bus = sum(ifelse(bus < 0, 0, bus)),
    train = sum(ifelse(train < 0, 0, train)),
    ferry = sum(ifelse(ferry < 0, 0, ferry)),
    other = sum(ifelse(other < 0, 0, other)),
    home = sum(ifelse(home < 0, 0, home)),
    total = sum(ifelse(total < 0, 0, total)), .groups="drop"
  ) -> work_to

edu_simp <- education_travel %>% select(
  res_code = SA2_code_usual_residence_address,
  res_name = SA2_name_usual_residence_address,
  res_east = SA2_usual_residence_easting,
  res_north = SA2_usual_residence_northing,
  edu_code = SA2_code_educational_address,
  edu_name = SA2_name_educational_address,
  edu_east = SA2_educational_easting,
  edu_north = SA2_educational_northing,
  drive = Drive_a_car_truck_or_van,
  passenger = Passenger_in_a_car_truck_or_van,
  walk = Walk_or_jog,
  bicycle = Bicycle,
  scholbus = School_bus,
  pubbus = Public_bus,
  train = Train,
  ferry = Ferry,
  other = Other,
  home = Study_at_home,
  total = Total
)

edu_simp %>% group_by(res_code, res_name, res_east, res_north) %>%
  summarise(
    drive = sum(ifelse(drive < 0, 0, drive)),
    passenger = sum(ifelse(passenger < 0, 0, passenger)),
    walk = sum(ifelse(walk < 0, 0, walk)),
    bicycle = sum(ifelse(bicycle < 0, 0, bicycle)),
    scholbus = sum(ifelse(scholbus < 0, 0, scholbus)),
    pubbus = sum(ifelse(pubbus < 0, 0, pubbus)),
    train = sum(ifelse(train < 0, 0, train)),
    ferry = sum(ifelse(ferry < 0, 0, ferry)),
    other = sum(ifelse(other < 0, 0, other)),
    home = sum(ifelse(home < 0, 0, home)),
    total = sum(ifelse(total < 0, 0, total)), .groups="drop"
  ) -> edu_from

edu_simp %>% group_by(edu_code, edu_name, edu_east, edu_north) %>%
  summarise(
    drive = sum(ifelse(drive < 0, 0, drive)),
    passenger = sum(ifelse(passenger < 0, 0, passenger)),
    walk = sum(ifelse(walk < 0, 0, walk)),
    bicycle = sum(ifelse(bicycle < 0, 0, bicycle)),
    scholbus = sum(ifelse(scholbus < 0, 0, scholbus)),
    pubbus = sum(ifelse(pubbus < 0, 0, pubbus)),
    train = sum(ifelse(train < 0, 0, train)),
    ferry = sum(ifelse(ferry < 0, 0, ferry)),
    other = sum(ifelse(other < 0, 0, other)),
    home = sum(ifelse(home < 0, 0, home)),
    total = sum(ifelse(total < 0, 0, total)), .groups="drop"
  ) -> edu_to

tencols <-  c("#f85654", "#e31a1c", "#1f78b4", "#6a3d9a", "#b2df8a", 
              "#33a02c", "#fdbf6f", "#ff7f00", "#cab2d6", "#af8ac1")

work_from$MAX <- work_from %>% select(private:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })
work_to$MAX <- work_to %>% select(private:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })
work_simp$MAX <- work_simp %>% select(private:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })

edu_from$MAX <- edu_from %>% select(drive:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })
edu_to$MAX <- edu_to %>% select(drive:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })
edu_simp$MAX <- edu_simp %>% select(drive:home) %>% as.matrix() %>% 
  apply(1, function(x) {
    ifelse(max(x) <= 0, NA, which.max(x))
    })

save(work_simp, work_to, work_from, edu_simp, edu_to, edu_from, tencols, file="viz/datasets.RData")