test_that("directlyStandardiseRates initial check", {
  # example from original dsr package
  # An example of calculating directly standardized rates
  # Data from Table 1, Page 132 of Schoenbach (2000)

  # State specific death counts and fu
  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(5,5)),
                         age=rep(c('00-14','15-24','25-44','45-64','65+'),2),
                         deaths=c(136,57,208,1016,3605,59,18,37,90,81),
                         fu=c(114350,80259,133440,142670,92168,37164,20036,32693,14947,2077))

  # US standard population
  df_ref  <- data.frame(age=c('00-14','15-24','25-44','45-64','65+'),
                        pop=c(23961000,15420000,21353000,19601000,10685000))

  # Directly Standardized Rates (per 100000)
  expect_no_error(my_results <- directlyStandardiseRates(data=df_study,
                                    event="deaths",
                                    denominator="fu",
                                    strata="state",
                                    age = "age",
                                    refdata=df_ref))

})

test_that("using package populations", {

  # european
  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_no_error(directlyStandardiseRates(data = stud_result,
      event = "deaths", denominator = "denominator",
      refdata  = standardPopulation("Europe")))
})

test_that("using strata", {
  stud_result <- data.frame(
    var_1 = c(rep("a", 19),
              rep("b", 19)),
    age_group = rep(c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),2),
    deaths= c(rep(50, 190),rep(100, 190)),
    denominator = rep(1000, 380))

  expect_no_error(directlyStandardiseRates(data = stud_result,
                      event = "deaths", denominator = "denominator",
                      strata = "var_1",
                      refdata  = standardPopulation("Europe")))
})

test_that("using multiple strata", {
  stud_result <- data.frame(
    var_1 = c(rep("a", 19),
              rep("b", 19)),
    var_2 = rep(c("c","d"), 19),
    age_group = rep(c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),2),
    deaths= c(rep(50, 190),rep(100, 190)),
    denominator = rep(1000, 380))

  expect_no_error(directlyStandardiseRates(data = stud_result,
                      event = "deaths", denominator = "denominator",
                      strata = c("var_1", "var_2"),
                      refdata  = standardPopulation("Europe")))
})

test_that("refdata is dataframe", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  refdata = c(
    "0 to 4", "5 to 9", "10 to 14",
    "15 to 19","20 to 24","25 to 29",
    "30 to 34","35 to 39","40 to 44",
    "45 to 49","50 to 54","55 to 59",
    "60 to 64","65 to 69","70 to 74",
    "75 to 79","80 to 84","85 to 89",
    "90 to 150"
  )

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator",
                   refdata  = refdata))

})

test_that("data is dataframe", {

  stud_result <- c(
    "0 to 4","5 to 9","10 to 14",
    "15 to 19","20 to 24","25 to 29",
    "30 to 34","35 to 39","40 to 44",
    "45 to 49","50 to 54","55 to 59",
    "60 to 64","65 to 69","70 to 74",
    "75 to 79","80 to 84","85 to 89",
    "90 to 150"
  )

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator"))

})

test_that("event is column in data", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "death_count", denominator = "denominator"))

})

test_that("denominator is column in data", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator (months)"))

})

test_that("strata is column in data", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator",
                   strata = "sex"))

})

test_that("strata is column in data", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator",
                   strata = "sex"))

})

test_that("pop is column in data", {

  stud_result <- data.frame(
    age_group = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator",
                   pop = "population"))

})

test_that("age is a column in data and refdata", {

  stud_result <- data.frame(
    age_groups = c(
      "0 to 4","5 to 9","10 to 14",
      "15 to 19","20 to 24","25 to 29",
      "30 to 34","35 to 39","40 to 44",
      "45 to 49","50 to 54","55 to 59",
      "60 to 64","65 to 69","70 to 74",
      "75 to 79","80 to 84","85 to 89",
      "90 to 150"
    ),
    deaths= rep(5, 19),
    denominator = rep(100, 19))

  expect_error(directlyStandardiseRates(data = stud_result,
                   event = "deaths", denominator = "denominator",
                   age = "age_group"))

  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(5,5)),
                         age=rep(c('00-14','15-24','25-44','45-64','65+'),2),
                         deaths=c(136,57,208,1016,3605,59,18,37,90,81),
                         fu=c(114350,80259,133440,142670,92168,37164,20036,32693,14947,2077))

  # US standard population
  df_ref  <- data.frame(ages=c('00-14','15-24','25-44','45-64','65+'),
                        pop=c(23961000,15420000,21353000,19601000,10685000))

  expect_error(directlyStandardiseRates(data = df_study,
                   event = "deaths", denominator = "fu",
                   refdata = df_ref,
                   age = "age"))

})

test_that("low outcome counts get warnings (2)",{

  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(5,5)),
                         age=rep(c('00-14','15-24','25-44','45-64','65+'),2),
                         deaths=c(1,1,2,1,0,0,1,3,0,0),
                         fu=c(114350,80259,133440,142670,92168,37164,20036,32693,14947,2077))

  df_ref  <- data.frame(age=c('00-14','15-24','25-44','45-64','65+'),
                        pop=c(23961000,15420000,21353000,19601000,10685000))

  expect_warning(directlyStandardiseRates(data = df_study,
                     event = "deaths",
                     denominator = "fu",
                     refdata = df_ref,
                     age = "age"))
})

test_that("missing age groups are included", {
  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(5,5)),
                         age=rep(c('00-14','15-24','25-44','45-64','65+'),2),
                         deaths=c(150,126,277,111,96,52,148,399,65,33),
                         fu=c(114350,80259,133440,142670,92168,37164,20036,32693,14947,2077))

  df_ref  <- data.frame(age=c('15-24','25-44','45-64','65+'),
                        pop=c(15420000,21353000,19601000,10685000))

  expect_no_error(
    dsr <- directlyStandardiseRates(
    data = df_study,
    event = "deaths",
    denominator = "fu",
    refdata = df_ref,
    age = "age"
  ))

})

test_that("add missing age groups FALSE", {
  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(4,4)),
                         age=rep(c('15-24','25-44','45-64','65+'),2),
                         deaths=c(126,277,111,96,148,399,65,33),
                         fu=c(80259,133440,142670,92168,20036,32693,14947,2077))

  df_ref  <- data.frame(age=c('00-14','15-24','25-44','45-64','65+'),
                        pop=c(1683400, 15420000,21353000,19601000,10685000))

  expect_no_error(
    dsr <- directlyStandardiseRates(
      data = df_study,
      event = "deaths",
      denominator = "fu",
      refdata = df_ref,
      age = "age",
      addMissingGroups = FALSE
    ))

})

test_that("add missing age groups TRUE", {
  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(4,4)),
                         age=rep(c('15-24','25-44','45-64','65+'),2),
                         deaths=c(126,277,111,96,148,399,65,33),
                         fu=c(80259,133440,142670,92168,20036,32693,14947,2077))

  df_ref  <- data.frame(age=c('00-14','15-24','25-44','45-64','65+'),
                        pop=c(1683400, 15420000,21353000,19601000,10685000))

  expect_no_error(
    dsr <- directlyStandardiseRates(
      data = df_study,
      event = "deaths",
      denominator = "fu",
      refdata = df_ref,
      age = "age",
      addMissingGroups = FALSE
    ))

})

test_that("Different age groups used in both datasets", {
  df_study <- data.frame(state=rep(c('Miami',"Alaska"), c(4,4)),
                         age=rep(c('0-24','25-44','45-64','65+'),2),
                         deaths=c(126,277,111,96,148,399,65,33),
                         fu=c(80259,133440,142670,92168,20036,32693,14947,2077))

  df_ref  <- data.frame(age=c('0-25','26-37','38-48','49-67','68+'),
                        pop=c(1683400, 15420000,21353000,19601000,10685000))

  expect_error(
    dsr <- directlyStandardiseRates(
      data = df_study,
      event = "deaths",
      denominator = "fu",
      refdata = df_ref,
      age = "age",
      addMissingGroups = FALSE
    ))

})