test_that("minimize_vl options behave as expected", {
  CM <- c(0, 4, 7)
  DM <- c(2, 6, 9)
  expect_equal(minimize_vl(CM, DM), matrix(c(2, -3, 2, -2, 2, -1), nrow=2))
  expect_equal(minimize_vl(CM, DM, no_ties=TRUE), c(2, 2, 2))
  expect_equal(minimize_vl(CM, DM, method="euclidean"), c(2, 2, 2))
  expect_equal(minimize_vl(CM, DM, edo=10), c(-1, -2, -1))
  expect_error(minimize_vl(CM, DM, method="badmethod"))
  expect_error(minimize_vl(c(0, 3, 7), c(0, 3, 7, 10)))
})

test_that("minimize_vl hamming distance works", {
  guido_c <- c(0, 2, 4, 5, 7, 9)
  guido_g <- c(0, 2, 4, 7, 9, 11)
  expect_equal(minimize_vl(guido_c, guido_g, method="hamming"),
               c(0, 0, 0, 6, 0, 0))

  expect_equal(minimize_vl(j(dia), j(1, 2, 3, 4, 5, 6, 7), method="h"),
               c(0, 0, 0, 0, 0, 0, 0))

  two_options_mat <- matrix(c(0, 0, 3, 0, 0, 0, 0, 0, 3, 0),
                            nrow=2, byrow=TRUE)
  expect_equal(minimize_vl(c(0, 2, 4, 4, 5), 
                           c(0, 2, 4, 5, 7), 
                           method="hamming"),
               two_options_mat)
  expect_equal(minimize_vl(c(0, 2, 2), c(0, 2, 2), method="hamming"),
               c(0, 0, 0))

  expect_equal(minimize_vl(c(0, 1, 1, 2, 6), 
                           c(11, 0, 1, 1, 5), 
                           method="hamming"),
               c(0, 0, 0, -3, -1))

  wt_1 <- c(0, 2, 2)
  wt_2 <- c(0, 0, 2)
  res_mat <- matrix(c(0, -2, 0, 0, 0, -2), nrow=2, byrow=TRUE)
  expect_equal(minimize_vl(wt_1, wt_2, method="h"), res_mat)
  expect_equal(minimize_vl(wt_1, wt_2, method="h", no_ties=TRUE),
               c(0, -2, 0))
})

test_that("whichmodebest works", {
  expect_equal(whichmodebest(c(0, 4, 7), c(7, 0, 4)), 2)
  expect_equal(whichmodebest(c(7, 0, 4), c(0, 4, 7)), 3)
  expect_equal(whichmodebest(c(0, 4, 7), c(2, 6, 9)), c(1, 3))
  expect_equal(whichmodebest(c(0, 4, 7), c(2, 6, 9), no_ties=TRUE), 1)
  expect_equal(whichmodebest(c(0, 4, 7), c(2, 6, 9),
                             method="euclidean"), 
               1)
  expect_equal(whichmodebest(c(0, 4, 6), c(10, 0, 4),
                             method="chebyshev"),
               c(1, 2))


  pyth_locrian <- 12 * log2(c(1, 256/243, 32/27, 4/3, 1024/729, 128/81, 16/9))
  pyth_lydian  <- 12 * log2(c(1, 9/8, 81/64, 729/512, 3/2, 27/16, 243/128))
  expect_equal(whichmodebest(pyth_locrian, pyth_lydian), 7)
  expect_equal(whichmodebest(pyth_locrian, pyth_lydian, method="hamming"), 1)

  expect_equal(whichmodebest(c(0, 6, 11), c(6, 12, 17), edo=19), 3)
})

test_that("vl_dist works", {
  c_dom7 <- c(0, 4, 7, 10)
  a_dom7 <- c(1, 4, 7, 9)
  expect_equal(vl_dist(c_dom7, c_dom7), 0)
  expect_equal(vl_dist(c_dom7, a_dom7), 2)
  expect_equal(vl_dist(a_dom7, c_dom7), 2)
  expect_equal(vl_dist(c_dom7, a_dom7, method="euclidean"), sqrt(2))
  expect_equal(vl_dist(c_dom7, a_dom7, method="chebyshev"), 1)
  expect_equal(vl_dist(c_dom7, a_dom7, method="hamming"), 2)
})

test_that("flex_points works", {
  expect_equal(flex_points(c(0, 4, 7)), c(2, 6, 10))
  expect_equal(flex_points(c(0, 1, 6)), c(2.5, 6, 9.5))
  expect_equal(flex_points(c(0, 4, 7), method="euclidean"), 
               c(2.08, 6.00, 9.91))
  expect_equal(flex_points(c(0, 3, 6, 10, 13), edo=17),
               c(1.7, 5.1, 8.5, 11.9, 15.3))
  expect_equal(flex_points(c(0, 2, 3, 5, 7, 9, 11), subdivide=2),
               c(0.5, 2.5, 4.5, 6.0, 7.5, 9.5, 11.0))
})

test_that("tndists works", {
  m7_tndists <- function() tndists(c(0, 3, 5, 8))
  vdiffr::expect_doppelganger("tndists plot for minor 7th chord",
                              m7_tndists)
})

test_that("tndists goal param works", {
  plr_tndists <- function() tndists(c(0,4,7), c(0,3,7))
  vdiffr::expect_doppelganger("tndists plots for M-m vls",
                              plr_tndists)
})