test-io.R
In PlotFTIR: Plot FTIR Spectra

test_that("reading csv works", {
  # Create a temporary CSV with wavenumber and absorbance columns
  data <- data.frame(
    wavenumber = 1000:1500,
    absorbance = biodiesel$absorbance[1:501]
  )
  temp_file <- withr::local_tempfile(fileext = ".csv")
  tmppath <- dirname(temp_file)
  tmpfile <- basename(temp_file)
  write.csv(data, file = temp_file, row.names = FALSE)

  # Read the data using read_ftir
  result <- read_ftir(path = tmppath, file = tmpfile)

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], tools::file_path_sans_ext(tmpfile))
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$wavenumber)
  expect_equal(round(result$absorbance, 4), round(data$absorbance, 4))

  # Make sure single file paths work
  expect_equal(result, read_ftir(path = file.path(tmppath, tmpfile), file = NA))

  # Create a temporary CSV with misnamed wavenumber column
  data <- data.frame(
    "row" = 1000:1500,
    absorbance = biodiesel$absorbance[1:501]
  )
  write.csv(data, file = temp_file, row.names = FALSE)

  # Read the data using read_ftir
  expect_message(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "has deduced that input data",
    fixed = TRUE
  )
  suppressMessages(result <- read_ftir(path = tmppath, file = tmpfile))

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], tools::file_path_sans_ext(tmpfile))
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$row)
  expect_equal(round(result$absorbance, 2), round(data$absorbance, 2))

  # do it backwards
  data <- data.frame("absorbance" = data$absorbance, "row" = 1000:1500)
  write.csv(data, file = temp_file, row.names = FALSE)
  expect_equal(
    result$wavenumber,
    read_ftir(path = tmppath, file = tmpfile)$wavenumber
  )

  # do it with a name match
  data <- data.frame("energy" = 1000:1500, "absorbance" = data$absorbance)
  write.csv(data, file = temp_file, row.names = FALSE)
  expect_equal(
    result$wavenumber,
    read_ftir(path = tmppath, file = tmpfile)$wavenumber
  )

  # Create a temporary CSV with misnamed energy column (absorbance)
  data <- data.frame(
    "wavenumber" = 1000:1500,
    "energy" = biodiesel$absorbance[1:501]
  )
  write.csv(data, file = temp_file, row.names = FALSE)

  # Read the data using read_ftir
  expect_message(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "has deduced that input data",
    fixed = TRUE
  )
  suppressMessages(result <- read_ftir(path = tmppath, file = tmpfile))

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], tools::file_path_sans_ext(tmpfile))
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$wavenumber)
  expect_equal(round(result$absorbance, 4), round(data$energy, 4))

  # Create a temporary CSV with misnamed energy column (transmittance)
  data <- data.frame(
    "wavenumber" = 1000:1500,
    "energy" = 100 - (biodiesel$absorbance[1:501] * 20)
  )
  write.csv(data, file = temp_file, row.names = FALSE)

  # Read the data using read_ftir
  expect_message(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "has deduced that input data",
    fixed = TRUE
  )
  suppressMessages(
    result <- read_ftir(path = tmppath, file = tmpfile, sample_name = "test")
  )

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "transmittance", "sample_id"))
  expect_equal(result$sample_id[1], "test")
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$wavenumber)
  expect_equal(round(result$transmittance, 2), round(data$energy, 2))

  data <- data.frame(
    "wavenumber" = 1000:1500,
    "absorbance" = biodiesel$absorbance[1:501],
    sample_id = "test"
  )
  write.csv(data, file = temp_file, row.names = FALSE)
  expect_error(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "Input file has too many columns",
    fixed = TRUE
  )

  data <- data.frame("row" = 1000:1500, "col" = 2000:2500)
  write.csv(data, file = temp_file, row.names = FALSE)
  expect_error(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "Could not confidently determine which column contains wavenumber",
    fixed = TRUE
  )
})

test_that("read_ftir handles invalid arguments", {
  expect_error(
    read_ftir(path = NULL, file = "file.csv"),
    regexp = "must be a single string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = "path", file = NULL),
    regexp = "must be a single string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = c("path1", "path2"), file = "file.csv"),
    regexp = "must be a single string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = "path", file = c("file1.csv", "file2.csv")),
    regexp = "must be a single string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = ".", file = "file.csv", sample_name = c("name1", "name2")),
    regexp = "must be a single string value or single",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = ".", file = "file.csv", sample_name = 123),
    regexp = "must be a string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir(path = ".", file = "nonexistent_file.csv"),
    regexp = 'nonexistent_file.csv" does not appear to exist',
    fixed = TRUE
  )
  tempfile <- withr::local_tempfile(fileext = ".docx")
  file.create(tempfile)
  expect_error(
    read_ftir(path = dirname(tempfile), file = basename(tempfile)),
    regexp = "could not be processed",
    fixed = TRUE
  )
  tempfile <- withr::local_tempfile(fileext = ".a2r")
  file.create(tempfile)
  expect_error(
    read_ftir(path = dirname(tempfile), file = basename(tempfile)),
    regexp = "PlotFTIR is not (yet) able to read .a2r files",
    fixed = TRUE
  )
  tempfile <- withr::local_tempfile(fileext = ".spc")
  file.create(tempfile)
  expect_error(
    read_ftir(path = dirname(tempfile), file = basename(tempfile)),
    regexp = "PlotFTIR is not (yet) able to read .spc files",
    fixed = TRUE
  )
})

test_that("reading asp works", {
  # create data and write file
  data <- data.frame(
    wavenumber = 1000:1500,
    absorbance = biodiesel$absorbance[1:501]
  )
  temp_file <- withr::local_tempfile(fileext = ".asp")
  tmppath <- dirname(temp_file)
  tmpfile <- basename(temp_file)

  write(
    c(
      nrow(data),
      max(data$wavenumber),
      min(data$wavenumber),
      1,
      2,
      4,
      rev(data$absorbance)
    ),
    temp_file,
    ncolumns = 1
  )

  # Read the data using read_ftir
  expect_message(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "has deduced that input data",
    fixed = TRUE
  )
  suppressMessages(result <- read_ftir(path = tmppath, file = tmpfile))

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], tools::file_path_sans_ext(tmpfile))
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$wavenumber)
  expect_equal(round(result$absorbance, 2), round(data$absorbance, 2))

  # Create a temporary CSV with misnamed energy column (transmittance)
  data <- data.frame(
    "wavenumber" = 1000:1500,
    "transmittance" = 100 - (biodiesel$absorbance[1:501] * 20)
  )
  write(
    c(
      nrow(data),
      max(data$wavenumber),
      min(data$wavenumber),
      1,
      2,
      4,
      rev(data$transmittance)
    ),
    temp_file,
    ncolumns = 1
  )

  # Read the data using read_ftir
  expect_message(
    read_ftir(path = tmppath, file = tmpfile),
    regexp = "has deduced that input data",
    fixed = TRUE
  )
  suppressMessages(
    result <- read_ftir(path = tmppath, file = tmpfile, sample_name = "test")
  )

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "transmittance", "sample_id"))
  expect_equal(result$sample_id[1], "test")
  expect_equal(nrow(result), nrow(data))
  expect_equal(result$wavenumber, data$wavenumber)
  expect_equal(round(result$transmittance, 2), round(data$transmittance, 2))
})

test_that("reading .jdx works", {
  if (!requireNamespace("readJDX", quietly = TRUE)) {
    expect_error(
      read_ftir_jdx(data.frame("testdata" = LETTERS)),
      regexp = "requires readJDX package installation for this function.",
      fixed = TRUE
    )
    testthat::skip("readJDX not available for testing interface")
  }

  jdx_ftir <- read_ftir(system.file("extdata", "SBO.jdx", package = "readJDX"))

  # read the .jdx file using readJDX
  jdx_jdx <- readJDX::readJDX(system.file(
    "extdata",
    "SBO.jdx",
    package = "readJDX"
  ))

  expect_equal(names(jdx_jdx)[4], unique(jdx_ftir$sample_id))
  expect_equal(nrow(jdx_ftir), nrow(jdx_jdx[[4]]))
  expect_true('transmittance' %in% colnames(jdx_ftir))
  expect_message(
    read_ftir(
      path = system.file("extdata", "SBO.jdx", package = "readJDX"),
      sample_name = 'test_sample'
    ),
    "does not match that contained in the .jdx file"
  )

  #More IR Data
  jdx_ir2 <- read_ftir(
    path = system.file("extdata", "MiniDIFDUP.JDX", package = "readJDX")
  )
  expect_equal(unique(jdx_ir2$sample_id), "Demo IR Spectrum")

  # NMR Data
  expect_error(
    read_ftir(
      path = system.file("extdata", "PCRF.jdx", package = "readJDX")
    ),
    "Could not confirm `infrared` data file."
  )
  # 2D NMR Data
  expect_error(
    read_ftir(
      path = system.file("extdata", "isasspc1.dx", package = "readJDX")
    ),
    "Could not confirm `infrared` data file."
  )
})

# Check reading multiple files
test_that("Reading multiple files works", {
  # Prep some files
  data <- data.frame(
    wavenumber = 1000:1500,
    absorbance = biodiesel$absorbance[1:501]
  )
  tmppath <- withr::local_tempdir()
  temp_file1 <- withr::local_tempfile(tmpdir = tmppath, fileext = ".csv")
  temp_file2 <- withr::local_tempfile(tmpdir = tmppath, fileext = ".csv")
  tmpfile1 <- basename(temp_file1)
  tmpfile2 <- basename(temp_file2)
  write.csv(data, file = file.path(tmppath, tmpfile1), row.names = FALSE)
  write.csv(data, file = file.path(tmppath, tmpfile2), row.names = FALSE)

  # Read the data using read_ftir
  result <- read_ftir_directory(
    path = tmppath,
    files = c(tmpfile1, tmpfile2),
    sample_names = c("one", "two")
  )

  # Check the result
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], "one")
  expect_equal(result$sample_id[nrow(result)], "two")
  expect_equal(nrow(result), nrow(data) * 2)
  expect_equal(result$wavenumber, rep(data$wavenumber, 2))
  expect_equal(round(result$absorbance, 4), rep(round(data$absorbance, 4), 2))

  # Read the data using read_ftir
  result <- read_ftir_directory(path = tmppath, files = c(tmpfile1, tmpfile2))

  # Check the result (no sample names)
  expect_equal(colnames(result), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(result$sample_id[1], tools::file_path_sans_ext(tmpfile1))
  expect_equal(
    result$sample_id[nrow(result)],
    tools::file_path_sans_ext(tmpfile2)
  )
  expect_equal(nrow(result), nrow(data) * 2)
  expect_equal(result$wavenumber, rep(data$wavenumber, 2))
  expect_equal(round(result$absorbance, 4), rep(round(data$absorbance, 4), 2))

  # Checking for issues
  expect_error(
    suppressWarnings(read_ftir_directory(
      path = tmppath,
      files = c("fake.csv", "fake2.csv")
    )),
    regexp = "No spectral data was read from files",
    fixed = TRUE
  )
  expect_warning(
    read_ftir_directory(
      path = tmppath,
      files = c(tmpfile1, tmpfile2, "fake.csv")
    ),
    regexp = 'fake.csv" does not appear to exist',
    fixed = TRUE
  )
  suppressWarnings(
    result2 <- read_ftir_directory(
      path = tmppath,
      files = c(tmpfile1, tmpfile2, "fake.csv")
    )
  )

  expect_equal(result, result2)

  expect_error(
    read_ftir_directory(
      path = tmppath,
      files = c(tmpfile1, tmpfile2),
      sample_names = c("One", "Two", "Extra")
    ),
    regexp = "You provided 3 `sample_names` and 2 `files`",
    fixed = TRUE
  )

  expect_error(
    read_ftir_directory(
      path = c(tmppath, tmppath),
      files = c(tmpfile1, tmpfile2)
    ),
    regexp = "must be a single string value",
    fixed = TRUE
  )
  expect_error(
    read_ftir_directory(
      path = tmppath,
      files = c(tmpfile1, as.data.frame(tmpfile2))
    ),
    regexp = "must be a vector of string values",
    fixed = TRUE
  )
})

test_that("plot saves", {
  if (!require("ggplot2", quietly = TRUE)) {
    # Of course, we can't generate a plot to feed to the manipulations.
    # This means that we can pass any value, the `ggplot` presence is tested first.

    expect_error(
      save_plot(123),
      "requires ggplot2 package installation",
      fixed = TRUE
    )
    testthat::skip("ggplot2 not available for testing file saving")
  }

  temp_file <- withr::local_tempfile(fileext = ".png")

  expect_false(file.exists(temp_file))
  save_plot(plot_ftir(biodiesel), filename = temp_file)
  expect_true(file.exists(temp_file))

  # test arg checks.
  expect_error(
    save_plot("abc", filename = temp_file),
    "`ftir_spectra_plot` must be a ggplot object. You provided a string",
    fixed = TRUE
  )
})

test_that("interface to ir is ok", {
  if (!requireNamespace("ir", quietly = TRUE)) {
    expect_error(
      ir_to_plotftir(data.frame("testdata" = LETTERS)),
      regexp = "requires ir package installation for this function.",
      fixed = TRUE
    )
    expect_error(
      ir_to_df(data.frame("testdata" = LETTERS)),
      regexp = "requires ir package installation for this function.",
      fixed = TRUE
    )
    expect_error(
      plotftir_to_ir(biodiesel),
      regexp = "requires ir package installation for this function.",
      fixed = TRUE
    )
    testthat::skip("ir not available for testing interface")
  }

  irdata <- ir::ir_sample_data
  # Param checks
  expect_error(
    ir_to_plotftir(biodiesel),
    regexp = "must be of class <ir>, produced by the ir package.",
    fixed = TRUE
  )
  expect_error(
    ir_to_df(biodiesel),
    regexp = "must be of class <ir>, produced by the ir package.",
    fixed = TRUE
  )
  expect_error(
    ir_to_plotftir(irdata, what = c(1, "two")),
    regexp = "must contain the row numbers of sample spectra to extract, or exact names matching what is in `ir_data$id_sample`",
    fixed = TRUE
  )
  expect_error(
    ir_to_plotftir(irdata, what = c(1, 1e6)),
    regexp = "must contain the row numbers of sample spectra to extract, or exact names matching what is in `ir_data$id_sample`",
    fixed = TRUE
  )

  expect_error(
    plotftir_to_ir(biodiesel, metadata = "bob"),
    regexp = "must be either `NA` or a <data.frame>"
  )

  allir <- ir_to_plotftir(irdata)
  expect_equal(length(unique(allir$sample_id)), nrow(irdata))
  expect_equal(colnames(allir), c("wavenumber", "absorbance", "sample_id"))

  irnum <- ir_to_plotftir(irdata, what = c(1:5))
  expect_equal(length(unique(irnum$sample_id)), 5)

  irname <- ir_to_plotftir(
    irdata,
    what = c("GN 11-389", "GN 11-400", "GN 11-407")
  )
  expect_equal(length(unique(irname$sample_id)), 3)

  plotir <- plotftir_to_ir(biodiesel)

  expect_equal(nrow(plotir), length(unique(biodiesel$sample_id)))

  plotirmeta <- plotftir_to_ir(
    biodiesel,
    data.frame(
      "biodiesel_content" = c(0, 0.25, 0.5, 1, 2.5, 5, 7.5, 10, 0.5, 5, NA)
    )
  )

  expect_equal(nrow(plotirmeta), length(unique(biodiesel$sample_id)))
  expect_true("biodiesel_content" %in% colnames(plotirmeta))

  expect_true("ggplot" %in% suppressWarnings(class(plot_ftir(irdata))))
})

test_that("Interface to ChemoSpec is ok", {
  if (!requireNamespace("R.utils", quietly = TRUE)) {
    expect_error(
      plotftir_to_chemospec(biodiesel),
      regexp = "requires R.utils package installation for this function.",
      fixed = TRUE
    )
    testthat::skip("R.utils not available for testing interface")
  }

  if (!requireNamespace("ChemoSpec", quietly = TRUE)) {
    expect_error(
      chemospec_to_plotftir(data.frame("testdata" = LETTERS)),
      regexp = "requires ChemoSpec package installation for this function.",
      fixed = TRUE
    )
    expect_error(
      plotftir_to_chemospec(biodiesel),
      regexp = "requires ChemoSpec package installation for this function.",
      fixed = TRUE
    )
    testthat::skip("ChemoSpec not available for testing interface")
  }

  data("SrE.IR", package = "ChemoSpec", envir = environment())
  data("SrE.NMR", package = "ChemoSpec", envir = environment())

  expect_error(
    chemospec_to_plotftir(SrE.NMR),
    regexp = "must be of IR spectra, this data appears to be from another instrument.",
    fixed = TRUE
  )
  expect_error(
    chemospec_to_plotftir(data.frame("A" = LETTERS)),
    regexp = "must be of class <Spectra>, produced by the ChemoSpec package. You provided ",
    fixed = TRUE
  )

  csftir <- chemospec_to_plotftir(SrE.IR)

  expect_equal(colnames(csftir), c("wavenumber", "absorbance", "sample_id"))
  expect_equal(length(unique(csftir$sample_id)), length(SrE.IR$names))

  expect_error(
    plotftir_to_chemospec(biodiesel, group_colours = "blue"),
    regexp = ", or a vector of the same length as group_crit",
    fixed = TRUE
  )
  expect_error(
    plotftir_to_chemospec(
      biodiesel,
      group_crit = c("biodiesel", "unknown"),
      group_colours = c("orange", "green", "blue")
    ),
    regexp = ", or a vector of the same length as group_crit",
    fixed = TRUE
  )
  expect_message(
    plotftir_to_chemospec(
      biodiesel,
      group_crit = c("biodiesel", "unknown"),
      group_colours = c("red", "blue"),
      description = "This is a very long description with 57 characters in it."
    ),
    regexp = "ChemoSpec advises that description is 40 characters or less. Your description is 57 characters",
    fixed = TRUE
  )

  expect_message(
    plotftir_to_chemospec(biodiesel),
    regexp = " to ensure enough colours available for groups.",
    fixed = TRUE
  )
  expect_error(
    plotftir_to_chemospec(rbind(biodiesel, sample_spectra)),
    regexp = " has to make 12 or less groups for ChemoSpec to be happy",
    fixed = TRUE
  )

  csdata <- plotftir_to_chemospec(
    biodiesel,
    group_crit = c("biodiesel", "unknown")
  )

  expect_equal(class(csdata), "Spectra")
  expect_type(csdata$data, "double")

  csdata2 <- chemospec_to_plotftir(csdata)

  expect_equal(
    csdata2[csdata2$sample_id == "biodiesel_0", ],
    biodiesel[biodiesel$sample_id == "biodiesel_0", ],
    ignore_attr = TRUE
  )

  expect_true("ggplot" %in% suppressWarnings(class(plot_ftir(SrE.IR))))
})