# tests/testthat/test-gg_kdmap.R

test_that("gg_kdmap handles a very tight KD range (similar affinities)", {
  # Data: all KD values ~100 nM
  test_tight <- data.frame(
    id = paste0("Var", 1:5),
    ka = c(1.2e5, 1.3e5, 1.1e5, 1.25e5, 1.15e5),
    kd = c(0.012, 0.013, 0.011, 0.0125, 0.0115)
  )

  p <- gg_kdmap(test_tight, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure tight-range data plots successfully as a ggplot object
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap handles a very wide KD range (pM to uM)", {
  # Data: KD spans orders of magnitude from 0.1 nM to 10,000 nM
  test_wide <- data.frame(
    id = paste0("Var", 1:6),
    ka = c(1e7, 1e6, 1e5, 1e4, 1e5, 1e6),
    kd = c(0.001, 0.01, 0.1, 1, 10, 100)
  )

  p <- gg_kdmap(test_wide, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure wide-range data (several orders of magnitude) is handled
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap handles ultra-high affinity binders (sub-nM)", {
  # Data: KD in the picomolar / sub-nM range
  test_high_aff <- data.frame(
    id = paste0("Binder", 1:4),
    ka = c(5e6, 8e6, 1e7, 6e6),
    kd = c(0.001, 0.0005, 0.0008, 0.0012)
  )

  p <- gg_kdmap(test_high_aff, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure very small KD values plot correctly
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap handles low-affinity binders (high nM to uM)", {
  # Data: KD in the ~500–2000 nM range
  test_low_aff <- data.frame(
    id = paste0("Weak", 1:4),
    ka = c(1e4, 5e3, 8e3, 1.2e4),
    kd = c(5, 10, 8, 6)
  )

  p <- gg_kdmap(test_low_aff, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure low-affinity interactions (large KD) are plotted
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap works for a single data point", {
  # Data: single binding variant
  test_single <- data.frame(
    id = "Solo",
    ka = 1e5,
    kd = 0.01
  )

  p <- gg_kdmap(test_single, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure function gracefully handles a single row input
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap works for two points in the same order of magnitude", {
  # Data: two similar KD values
  test_two <- data.frame(
    id = c("A", "B"),
    ka = c(1e5, 2e5),
    kd = c(0.01, 0.02)
  )

  p <- gg_kdmap(test_two, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure small datasets with similar KD are supported
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap handles extremely fast on-rates", {
  # Data: very fast association rate constants
  test_extreme_on <- data.frame(
    id = paste0("Fast", 1:3),
    ka = c(1e8, 5e7, 8e7),
    kd = c(0.1, 0.05, 0.08)
  )

  p <- gg_kdmap(test_extreme_on, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure large ka values do not break the plot
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap handles extremely fast off-rates", {
  # Data: very fast dissociation rate constants
  test_extreme_off <- data.frame(
    id = paste0("Unstable", 1:3),
    ka = c(1e4, 5e4, 2e4),
    kd = c(50, 100, 75)
  )

  p <- gg_kdmap(test_extreme_off, iso_n = 15, show_anno = TRUE)

  # Purpose: ensure very large kd values do not break the plot
  expect_s3_class(p, "ggplot")
})

test_that("gg_kdmap validates input parameters correctly", {
  # Setup valid data
  valid_data <- data.frame(
    id = c("A", "B", "C"),
    ka = c(1e5, 2e5, 3e5),
    kd = c(1e-3, 2e-3, 3e-3),
    label_col = c("Label1", "Label2", "Label3"),
    size_col = c(3, 4, 5),
    shape_col = c(21, 22, 23),
    fill_col = c("red", "blue", "green")
  )
  
  # Data validation
  expect_error(
    gg_kdmap(data = "not_a_df"),
    "'data' must be a data frame"
  )
  
  # Labels column validation
  expect_error(
    gg_kdmap(data = valid_data, labels = "missing_col"),
    "Column 'missing_col' not found in data frame"
  )
  
  # ref_id validation
  expect_error(
    gg_kdmap(data = valid_data, ref_id = "NotInData"),
    "Reference ID 'NotInData' not found in data"
  )
  
  # Size validation - neither number nor column
  expect_error(
    gg_kdmap(data = valid_data, size = "missing_size_col"),
    "'size' must be a number or column name in data"
  )
  
  # Shape validation - invalid number
  expect_error(
    gg_kdmap(data = valid_data, shape = 99),
    "'shape' must be between 0 and 25"
  )
  
  # Shape validation - not column name
  expect_error(
    gg_kdmap(data = valid_data, shape = "missing_shape"),
    "'shape' must be a valid shape number or column name in data"
  )
  
  # ref_shape validation
  expect_error(
    gg_kdmap(data = valid_data, ref_shape = -1),
    "'ref_shape' must be between 0 and 25"
  )
  
  # Color/fill validation - neither valid color nor column
  expect_error(
    gg_kdmap(data = valid_data, fill = "not_a_color_or_column"),
    "'fill' must be a valid color or column name in data"
  )
  
  expect_error(
    gg_kdmap(data = valid_data, color = "invalid_color_xyz"),
    "'color' must be a valid color or column name in data"
  )
  
  # rep_lines validation
  expect_error(
    gg_kdmap(data = valid_data, rep_lines = "yes"),
    "'rep_lines' must be TRUE or FALSE"
  )
  
  # iso_alpha validation
  expect_error(
    gg_kdmap(data = valid_data, iso_alpha = 1.5),
    "'iso_alpha' must be between 0 and 1"
  )
  
  expect_error(
    gg_kdmap(data = valid_data, iso_alpha = -0.1),
    "'iso_alpha' must be between 0 and 1"
  )
  
  # iso_width validation
  expect_error(
    gg_kdmap(data = valid_data, iso_width = 0),
    "'iso_width' must be a positive number"
  )
  
  # iso_type validation
  expect_error(
    gg_kdmap(data = valid_data, iso_type = "invalid_type"),
    "'iso_type' must be one of"
  )
  
  # iso_n validation
  expect_error(
    gg_kdmap(data = valid_data, iso_n = 0),
    "'iso_n' must be a positive number"
  )
  
  # text_padding validation
  expect_error(
    gg_kdmap(data = valid_data, text_padding = -1),
    "'text_padding' must be a non-negative number"
  )
  
  # show_anno validation
  expect_error(
    gg_kdmap(data = valid_data, show_anno = "no"),
    "'show_anno' must be TRUE or FALSE"
  )
})

test_that("gg_kdmap handles valid column mappings", {
  test_data <- data.frame(
    id = c("A", "B", "C"),
    ka = c(1e5, 2e5, 3e5),
    kd = c(1e-3, 2e-3, 3e-3),
    pt_size = c(3, 4, 5),
    pt_shape = c(21, 22, 23),
    pt_fill = c("red", "blue", "green")
  )
  
  # Valid size as column
  expect_s3_class(
    gg_kdmap(data = test_data, size = "pt_size"),
    "ggplot"
  )
  
  # Valid shape as column
  expect_s3_class(
    gg_kdmap(data = test_data, shape = "pt_shape"),
    "ggplot"
  )
  
  # Valid fill as column
  expect_s3_class(
    gg_kdmap(data = test_data, fill = "pt_fill"),
    "ggplot"
  )
})

# --------------------------------------------------------------------
# Additional feature tests
# --------------------------------------------------------------------

test_that("gg_kdmap respects show_anno argument", {
  # Reuse a moderate dataset
  test_wide <- data.frame(
    id = paste0("Var", 1:6),
    ka = c(1e7, 1e6, 1e5, 1e4, 1e5, 1e6),
    kd = c(0.001, 0.01, 0.1, 1, 10, 100)
  )

  p_anno    <- gg_kdmap(test_wide, iso_n = 15, show_anno = TRUE)
  p_no_anno <- gg_kdmap(test_wide, iso_n = 15, show_anno = FALSE)

  # Purpose: both variants should produce valid ggplot objects
  expect_s3_class(p_anno, "ggplot")
  expect_s3_class(p_no_anno, "ggplot")
})

test_that("gg_kdmap accepts different iso_n values without error", {
  # Reuse a representative dataset
  test_tight <- data.frame(
    id = paste0("Var", 1:5),
    ka = c(1.2e5, 1.3e5, 1.1e5, 1.25e5, 1.15e5),
    kd = c(0.012, 0.013, 0.011, 0.0125, 0.0115)
  )

  # Purpose: ensure iso_n parameter is flexible and does not cause failures
  expect_s3_class(gg_kdmap(test_tight, iso_n = 5,  show_anno = TRUE), "ggplot")
  expect_s3_class(gg_kdmap(test_tight, iso_n = 10, show_anno = TRUE), "ggplot")
  expect_s3_class(gg_kdmap(test_tight, iso_n = 20, show_anno = TRUE), "ggplot")
})

test_that("gg_kdmap errors when required columns are missing", {
  # Start from a valid dataset
  valid <- data.frame(
    id = c("A", "B"),
    ka = c(1e5, 2e5),
    kd = c(0.01, 0.02)
  )

  # Purpose: missing kd column should produce an error
  expect_error(
    gg_kdmap(valid[, c("id", "ka")]),
    regexp = ""
  )

  # Purpose: missing ka column should also produce an error
  expect_error(
    gg_kdmap(valid[, c("id", "kd")]),
    regexp = ""
  )
})

test_that("gg_kdmap works when id is factor or character", {
  base_df <- data.frame(
    id = c("A", "B", "C"),
    ka = c(1e5, 2e5, 1.5e5),
    kd = c(0.01, 0.02, 0.015)
  )

  df_char <- base_df
  df_factor <- base_df
  df_factor$id <- factor(df_factor$id)

  # Purpose: support both character and factor id types
  p_char   <- gg_kdmap(df_char,   iso_n = 15, show_anno = TRUE)
  p_factor <- gg_kdmap(df_factor, iso_n = 15, show_anno = TRUE)

  expect_s3_class(p_char, "ggplot")
  expect_s3_class(p_factor, "ggplot")
})