library(testthat)

test_that("bootstrap coverage does not trigger the full-data guard", {
  in_id <- seq_len(10)
  expect_true(
    fastml:::fastml_guard_detect_full_analysis(in_id, 10, is_bootstrap = FALSE)
  )
  expect_false(
    fastml:::fastml_guard_detect_full_analysis(in_id, 10, is_bootstrap = TRUE)
  )
})

test_that("guarded resampling errors when split indices are missing", {
  skip_if_not_installed("rsample")

  set.seed(101)
  binary_iris <- iris[iris$Species != "virginica", ]
  binary_iris$Species <- factor(binary_iris$Species)

  idx <- sample(seq_len(nrow(binary_iris)), size = floor(0.7 * nrow(binary_iris)))
  train_split <- binary_iris[idx, , drop = FALSE]
  test_split <- binary_iris[-idx, , drop = FALSE]

  resamples <- rsample::vfold_cv(train_split, v = 3)
  resamples$splits[[1]]$in_id <- NULL

  # The error is now caught, reported as a warning, and tracked in failed_models
  # The final error is "No models were successfully trained"
  expect_error(
    expect_warning(
      fastml(
        train_data = train_split,
        test_data = test_split,
        label = "Species",
        algorithms = "logistic_reg",
        resamples = resamples,
        resampling_method = "cv",
        folds = 3,
        seed = 202
      ),
      "split index information"
    ),
    "No models were successfully trained"
  )
})

test_that("guarded resampling requires consistent index fields", {
  skip_if_not_installed("rsample")

  set.seed(202)
  binary_iris <- iris[iris$Species != "virginica", ]
  binary_iris$Species <- factor(binary_iris$Species)

  idx <- sample(seq_len(nrow(binary_iris)), size = floor(0.7 * nrow(binary_iris)))
  train_split <- binary_iris[idx, , drop = FALSE]
  test_split <- binary_iris[-idx, , drop = FALSE]

  resamples <- rsample::vfold_cv(train_split, v = 3)
  resamples$splits[[1]]$analysis_id <- resamples$splits[[1]]$in_id
  resamples$splits[[1]]$in_id <- NULL

  # The error is now caught, reported as a warning, and tracked in failed_models
  # The final error is "No models were successfully trained"
  expect_error(
    expect_warning(
      fastml(
        train_data = train_split,
        test_data = test_split,
        label = "Species",
        algorithms = "logistic_reg",
        resamples = resamples,
        resampling_method = "cv",
        folds = 3,
        seed = 303
      ),
      "consistent index field"
    ),
    "No models were successfully trained"
  )
})