tests/testthat/test-read_stats19.R

In ITSLeeds/stats19: Work with Open Road Traffic Casualty Data from Great Britain

context("test-read_stats19")

source("../skip-download.R")

test_that("read_accidents works", {
  # expect error for clean data_dir
  expect_error(read_accidents(year = 2016))
  skip_download()
  # download real data
  acc_2016 = stats19::file_names$dftRoadSafety_Accidents_2016.zip
  dl_stats19(file_name = acc_2016)
  # make sure we have a csv file to read
  path = locate_one_file(
    tempdir(),
    type = "accidents",
    year = 2016,
    filename = sub(".zip", ".csv", acc_2016)
  )
  # read it
  read = read_accidents(
    year = 2016, # make it unique
    data_dir = tempdir()
    # filename = sub(".zip", ".csv", acc_2016)
  )
  raw_read = read.csv(path)
  expect_equal(nrow(read), nrow(raw_read))
  # read with just file name
  read = read_accidents(
    filename = sub(".zip", ".csv", acc_2016)
  )
  expect_equal(nrow(read), nrow(raw_read))
  # read with just file name and format
  read = read_accidents(
    filename = sub(".zip", ".csv", acc_2016),
    format = TRUE
  )
  expect_equal(nrow(read), nrow(raw_read))
  expect_error(read_accidents("junk"))
})
test_that("read_* acc_index works", {
  skip_download()
  veh_2009 = stats19::file_names$DfTRoadSafety_Vehicles_2009.zip
  dl_stats19(file_name = veh_2009)
  cas_2009 = stats19::file_names$DfTRoadSafety_Casualties_2009.zip
  dl_stats19(file_name = cas_2009)
  read_veh = read_vehicles(year = 2009)
  read_cas = read_casualties(year = 2009)
  expect_true(identical(names(read_veh)[1], "accident_index"))
  expect_true(identical(names(read_cas)[1], "accident_index"))

})
test_that("read_vehicles works", {
  skip_download()
  # download real data
  veh_2016 = stats19::file_names$dftRoadSafetyData_Vehicles_2016.zip
  dl_stats19(file_name = veh_2016)

  path = locate_one_file( # need it for raw_read
    data_dir = tempdir(),
    type = "vehicles",
    year = 2016)
  # read it
  read = read_vehicles(
    year = 2016,
    data_dir = tempdir(),
    filename = "Veh.csv"
  )
  raw_read = read.csv(path)
  expect_false(identical(
    class(read$Accident_Index),
    class(raw_read$Accident_Index)
    ))
  read_formatted = read_vehicles(
    year = 2016,
    data_dir = tempdir(),
    filename = "Veh.csv",
    format = FALSE
  )
  expect_false(identical(
    names(read)[1],
    names(read_formatted[1])
  ))
  # read it using file name only IF only one is Veh.csv is downloaded.
  # thefore start from clean
  unlink(tempdir(), recursive = TRUE)
  dir.create(tempdir())
  dl_stats19(file_name = veh_2016)
  read = read_vehicles(
    filename = "Veh.csv"
  )
  expect_false(identical(
    class(read$Accident_Index),
    class(raw_read$Accident_Index)
  ))
  expect_error(read_vehicles("junk"))
  # "Your parameters return identical filenames under different directories."
  dl_stats19(year = 2017, type = 2017)
  # Now 2016 and 2017 file named `Veh.csv` in.
  expect_error(read_accidents(filename = "Veh.csv"))
})

test_that("read_casualties works", {
  skip_download()
  # download real data
  cas_2016 = stats19::file_names$dftRoadSafetyData_Casualties_2016.zip
  dl_stats19(file_name = cas_2016)
  path = locate_one_file(
    type = "Casualties",
    data_dir = tempdir(),
    year = 2016,
    filename = "Cas.csv")
  # read it
  read = read_casualties(
    year = 2016, # make it unique
    data_dir = tempdir(),
    filename = "Cas.csv",
    format = FALSE
  )
  raw_read = read.csv(path)
  expect_false(identical(
    class(read$Accident_Index),
    class(raw_read$Accident_Index)
  ))
  read_formatted = read_casualties(
    year = 2016,
    data_dir = tempdir(),
    format = TRUE
  )
  expect_false(identical(
    names(read)[1],
    names(read_formatted[1])
  ))
  expect_error(read_vehicles("junk"))
})