context("examples")

library(ggplot2)

test_that("examples work", {
  # barchart of state rankings in various categories
  expect_no_error({
    p <- ggplot(state_ranks, aes(variable, rank, fill = variable)) +
      geom_col() +
      coord_flip() +
      facet_geo(~ state) +
      theme_bw()
    print(p)
  })

  # use an alternative US state grid and place
  expect_no_error({
    p <- ggplot(state_ranks, aes(variable, rank, fill = variable)) +
      geom_col() +
      coord_flip() +
      facet_geo(~ state, grid = "us_state_grid2") +
      theme(panel.spacing = unit(0.1, "lines"))
    print(p)
  })

  expect_no_error(grid_preview(us_state_grid2))
  expect_no_error(grid_preview(eu_grid1, label = "name"))

  # custom grid (move Hawaii over more)
  expect_no_error({
    my_grid <- us_state_grid2
    my_grid$col <- my_grid$col + 2
    my_grid$col[my_grid$code == "HI"] <- 1
    grid_preview(my_grid)
  })

  # test to make sure we can have empty columns (since Hawaii is moved over)
  expect_no_error({
    p <- ggplot(state_ranks, aes(variable, rank, fill = variable)) +
      geom_col() +
      coord_flip() +
      facet_geo(~ state, grid = my_grid)
    print(p)
  })

  # use a free x-axis (not a good idea but just to show it works)
  expect_no_error({
    p <- ggplot(state_ranks, aes(variable, rank, fill = variable)) +
      geom_col() +
      coord_flip() +
      facet_geo(~ state, scales = "free_x") +
      theme_bw()
    print(p)
  })

  # plot unemployment rate time series for each state
  expect_no_error({
    p <- ggplot(state_unemp, aes(year, rate)) +
      geom_line() +
      facet_geo(~ state) +
      scale_x_continuous(labels = function(x) paste0("'", substr(x, 3, 4))) +
      ylab("Unemployment Rate (%)")
    print(p)
  })

  # plot the 2016 unemployment rate
  expect_no_error({
    p <- ggplot(subset(state_unemp, year == 2016), aes(factor(year), rate)) +
      geom_col(fill = "steelblue") +
      facet_geo(~ state) +
      theme(
        axis.title.x = element_blank(),
        axis.text.x = element_blank(),
        axis.ticks.x = element_blank()) +
      ylab("Unemployment Rate (%)")
    print(p)
  })

  # plot European Union GDP
  expect_no_error({
    p <- ggplot(eu_gdp, aes(year, gdp_pc)) +
      geom_line(color = "steelblue") +
      geom_hline(yintercept = 100, linetype = 2) +
      facet_geo(~ name, grid = "eu_grid1") +
      scale_x_continuous(labels = function(x) paste0("'", substr(x, 3, 4))) +
      ylab("GDP Per Capita") +
      theme_bw()
    print(p)
  })

  # use a free y-axis to look at just change
  expect_no_error({
    p <- ggplot(eu_gdp, aes(year, gdp_pc)) +
      geom_line(color = "steelblue") +
      facet_geo(~ name, grid = "eu_grid1", scales = "free_y") +
      scale_x_continuous(labels = function(x) paste0("'", substr(x, 3, 4))) +
      ylab("GDP Per Capita in Relation to EU Index (100)") +
      theme_bw()
    print(p)
  })

  # plot European Union annual # of resettled persons
  expect_no_error({
    p <- ggplot(eu_imm, aes(year, persons)) +
      geom_line() +
      facet_geo(~ name, grid = "eu_grid1") +
      scale_x_continuous(labels = function(x) paste0("'", substr(x, 3, 4))) +
      scale_y_sqrt(minor_breaks = NULL) +
      ylab("# Resettled Persons") +
      theme_bw()
    print(p)
  })

  # plot just for 2016
  expect_no_error({
    p <- ggplot(subset(eu_imm, year == 2016), aes(factor(year), persons)) +
      geom_col(fill = "steelblue") +
      geom_text(aes(factor(year), 3000, label = persons), color = "gray") +
      facet_geo(~ name, grid = "eu_grid1") +
      theme(
        axis.title.x = element_blank(),
        axis.text.x = element_blank(),
        axis.ticks.x = element_blank()) +
      ylab("# Resettled Persons in 2016") +
      xlab("Year") +
      theme_bw()
    print(p)
  })

  # plot Australian population
  expect_no_error({
    p <- ggplot(aus_pop, aes(age_group, pop / 1e6, fill = age_group)) +
      geom_col() +
      facet_geo(~ code, grid = "aus_grid1") +
      coord_flip() +
      labs(
        title = "Australian Population Breakdown",
        caption = "Data Source: ABS Labour Force Survey, 12 month average",
        y = "Population [Millions]") +
      theme_bw()
    print(p)
  })

  # South Africa population density by province
  expect_no_error({
    p <- ggplot(sa_pop_dens, aes(factor(year), density, fill = factor(year))) +
      geom_col() +
      facet_geo(~ code, grid = "sa_prov_grid1") +
      labs(title = "South Africa population density by province",
        caption = "Data Source: Statistics SA Census",
        y = "Population density per square km") +
      theme_bw()
    print(p)
  })

  # Use the Afrikaans name stored in the grid, "name_af", as facet labels
  expect_no_error({
    p <- ggplot(sa_pop_dens, aes(factor(year), density, fill = factor(year))) +
      geom_col() +
      facet_geo(~ code, grid = "sa_prov_grid1", label = "name_af") +
      labs(title = "South Africa population density by province",
        caption = "Data Source: Statistics SA Census",
        y = "Population density per square km") +
      theme_bw()
    print(p)
  })

  Sys.setenv(GEOFACET_PKG_TESTING = "TRUE")
  expect_no_error({
    my_grid <- us_state_grid1
    my_grid$col[my_grid$code == "WI"] <- 7
    grid_submit(my_grid, name = "us_grid_tweak_wi",
      desc = "Modified us_state_grid1 to move WI over")
  })

  expect_no_error({
    # edit aus_grid1
    grid_design(data = aus_grid1,
      img = "http://www.john.chapman.name/Austral4.gif")
    # start with a clean slate
    grid_design()
    # arrange the alphabet
    grid_design(data.frame(code = letters))
  })

  Sys.setenv(GEOFACET_PKG_TESTING = "")

  nms <- get_grid_names()
})