path: root/jni/ruby/test/csv/test_features.rb

#!/usr/bin/env ruby -w
# encoding: UTF-8

# tc_features.rb
#
#  Created by James Edward Gray II on 2005-10-31.
#  Copyright 2005 James Edward Gray II. You can redistribute or modify this code
#  under the terms of Ruby's license.

begin
  require "zlib"
rescue LoadError
end

require_relative "base"
require "tempfile"

class TestCSV::Features < TestCSV
  extend DifferentOFS

  TEST_CASES = [ [%Q{a,b},               ["a", "b"]],
                 [%Q{a,"""b"""},         ["a", "\"b\""]],
                 [%Q{a,"""b"},           ["a", "\"b"]],
                 [%Q{a,"b"""},           ["a", "b\""]],
                 [%Q{a,"\nb"""},         ["a", "\nb\""]],
                 [%Q{a,"""\nb"},         ["a", "\"\nb"]],
                 [%Q{a,"""\nb\n"""},     ["a", "\"\nb\n\""]],
                 [%Q{a,"""\nb\n""",\nc}, ["a", "\"\nb\n\"", nil]],
                 [%Q{a,,,},              ["a", nil, nil, nil]],
                 [%Q{,},                 [nil, nil]],
                 [%Q{"",""},             ["", ""]],
                 [%Q{""""},              ["\""]],
                 [%Q{"""",""},           ["\"",""]],
                 [%Q{,""},               [nil,""]],
                 [%Q{,"\r"},             [nil,"\r"]],
                 [%Q{"\r\n,"},           ["\r\n,"]],
                 [%Q{"\r\n,",},          ["\r\n,", nil]] ]

  def setup
    super
    @sample_data = <<-END_DATA.gsub(/^ +/, "")
    line,1,abc
    line,2,"def\nghi"

    line,4,jkl
    END_DATA
    @csv = CSV.new(@sample_data)
  end

  def test_col_sep
    [";", "\t"].each do |sep|
      TEST_CASES.each do |test_case|
        assert_equal( test_case.last.map { |t| t.tr(",", sep) unless t.nil? },
                      CSV.parse_line( test_case.first.tr(",", sep),
                                      col_sep: sep ) )
      end
    end
    assert_equal([",,,", nil], CSV.parse_line(",,,;", col_sep: ";"))
  end

  def test_row_sep
    assert_raise(CSV::MalformedCSVError) do
        CSV.parse_line("1,2,3\n,4,5\r\n", row_sep: "\r\n")
    end
    assert_equal( ["1", "2", "3\n", "4", "5"],
                  CSV.parse_line(%Q{1,2,"3\n",4,5\r\n}, row_sep: "\r\n"))
  end

  def test_quote_char
    TEST_CASES.each do |test_case|
      assert_equal( test_case.last.map { |t| t.tr('"', "'") unless t.nil? },
                    CSV.parse_line( test_case.first.tr('"', "'"),
                                    quote_char: "'" ) )
    end
  end

  def test_bug_8405
    TEST_CASES.each do |test_case|
      assert_equal( test_case.last.map { |t| t.tr('"', "|") unless t.nil? },
                    CSV.parse_line( test_case.first.tr('"', "|"),
                                    quote_char: "|" ) )
    end
  end

  def test_csv_char_readers
    %w[col_sep row_sep quote_char].each do |reader|
      csv = CSV.new("abc,def", reader.to_sym => "|")
      assert_equal("|", csv.send(reader))
    end
  end

  def test_row_sep_auto_discovery
    ["\r\n", "\n", "\r"].each do |line_end|
      data       = "1,2,3#{line_end}4,5#{line_end}"
      discovered = CSV.new(data).row_sep
      assert_equal(line_end, discovered)
    end

    assert_equal("\n", CSV.new("\n\r\n\r").row_sep)

    assert_equal($/, CSV.new("").row_sep)

    assert_equal($/, CSV.new(STDERR).row_sep)
  end

  def test_lineno
    assert_equal(5, @sample_data.lines.to_a.size)

    4.times do |line_count|
      assert_equal(line_count, @csv.lineno)
      assert_not_nil(@csv.shift)
      assert_equal(line_count + 1, @csv.lineno)
    end
    assert_nil(@csv.shift)
  end

  def test_readline
    test_lineno

    @csv.rewind

    test_lineno
  end

  def test_unknown_options
    assert_raise_with_message(ArgumentError, /unknown/) {
      CSV.new(@sample_data, unknown: :error)
    }
  end

  def test_skip_blanks
    assert_equal(4, @csv.to_a.size)

    @csv  = CSV.new(@sample_data, skip_blanks: true)

    count = 0
    @csv.each do |row|
      count += 1
      assert_equal("line", row.first)
    end
    assert_equal(3, count)
  end

  def test_csv_behavior_readers
    %w[ unconverted_fields return_headers write_headers
        skip_blanks        force_quotes ].each do |behavior|
      assert_not_predicate(CSV.new("abc,def"), "#{behavior}?", "Behavior defaulted to on.")
      csv = CSV.new("abc,def", behavior.to_sym => true)
      assert_predicate(csv, "#{behavior}?", "Behavior change now registered.")
    end
  end

  def test_converters_reader
    # no change
    assert_equal( [:integer],
                  CSV.new("abc,def", converters: [:integer]).converters )

    # just one
    assert_equal( [:integer],
                  CSV.new("abc,def", converters: :integer).converters )

    # expanded
    assert_equal( [:integer, :float],
                  CSV.new("abc,def", converters: :numeric).converters )

    # custom
    csv = CSV.new("abc,def", converters: [:integer, lambda {  }])
    assert_equal(2, csv.converters.size)
    assert_equal(:integer, csv.converters.first)
    assert_instance_of(Proc, csv.converters.last)
  end

  def test_header_converters_reader
    # no change
    hc = :header_converters
    assert_equal([:downcase], CSV.new("abc,def", hc => [:downcase]).send(hc))

    # just one
    assert_equal([:downcase], CSV.new("abc,def", hc => :downcase).send(hc))

    # custom
    csv = CSV.new("abc,def", hc => [:symbol, lambda {  }])
    assert_equal(2, csv.send(hc).size)
    assert_equal(:symbol, csv.send(hc).first)
    assert_instance_of(Proc, csv.send(hc).last)
  end

  # reported by Kev Jackson
  def test_failing_to_escape_col_sep_bug_fix
    assert_nothing_raised(Exception) { CSV.new(String.new, col_sep: "|") }
  end

  # reported by Chris Roos
  def test_failing_to_reset_headers_in_rewind_bug_fix
    csv = CSV.new("forename,surname", headers: true, return_headers: true)
    csv.each {|row| assert_predicate row, :header_row?}
    csv.rewind
    csv.each {|row| assert_predicate row, :header_row?}
  end

  # reported by Dave Burt
  def test_leading_empty_fields_with_multibyte_col_sep_bug_fix
    data = <<-END_DATA.gsub(/^\s+/, "")
    <=><=>A<=>B<=>C
    1<=>2<=>3
    END_DATA
    parsed = CSV.parse(data, col_sep: "<=>")
    assert_equal([[nil, nil, "A", "B", "C"], ["1", "2", "3"]], parsed)
  end

  def test_gzip_reader_bug_fix
    zipped = nil
    assert_nothing_raised(NoMethodError) do
      zipped = CSV.new(
                 Zlib::GzipReader.open(
                   File.join(File.dirname(__FILE__), "line_endings.gz")
                 )
               )
    end
    assert_equal("\r\n", zipped.row_sep)
  ensure
    zipped.close
  end if defined?(Zlib::GzipReader)

  def test_gzip_writer_bug_fix
    Tempfile.create(%w"temp .gz") {|tempfile|
      tempfile.close
      file = tempfile.path
      zipped = nil
      assert_nothing_raised(NoMethodError) do
        zipped = CSV.new(Zlib::GzipWriter.open(file))
      end
      zipped << %w[one two three]
      zipped << [1, 2, 3]
      zipped.close

      assert_include(Zlib::GzipReader.open(file) {|f| f.read},
                     $INPUT_RECORD_SEPARATOR, "@row_sep did not default")
    }
  end if defined?(Zlib::GzipWriter)

  def test_inspect_is_smart_about_io_types
    str = CSV.new("string,data").inspect
    assert_include(str, "io_type:StringIO", "IO type not detected.")

    str = CSV.new($stderr).inspect
    assert_include(str, "io_type:$stderr", "IO type not detected.")

    Tempfile.create(%w"temp .csv") {|tempfile|
      tempfile.close
      path = tempfile.path
      File.open(path, "w") { |csv| csv << "one,two,three\n1,2,3\n" }
      str  = CSV.open(path) { |csv| csv.inspect }
      assert_include(str, "io_type:File", "IO type not detected.")
    }
  end

  def test_inspect_shows_key_attributes
    str = @csv.inspect
    %w[lineno col_sep row_sep quote_char].each do |attr_name|
      assert_match(/\b#{attr_name}:[^\s>]+/, str)
    end
  end

  def test_inspect_shows_headers_when_available
    CSV.new("one,two,three\n1,2,3\n", headers: true) do |csv|
      assert_include(csv.inspect, "headers:true", "Header hint not shown.")
      csv.shift  # load headers
      assert_match(/headers:\[[^\]]+\]/, csv.inspect)
    end
  end

  def test_inspect_encoding_is_ascii_compatible
    CSV.new("one,two,three\n1,2,3\n".encode("UTF-16BE")) do |csv|
      assert_send([Encoding, :compatible?,
                   Encoding.find("US-ASCII"), csv.inspect.encoding],
                  "inspect() was not ASCII compatible.")
    end
  end

  def test_version
    assert_not_nil(CSV::VERSION)
    assert_instance_of(String, CSV::VERSION)
    assert_predicate(CSV::VERSION, :frozen?)
    assert_match(/\A\d\.\d\.\d\Z/, CSV::VERSION)
  end

  def test_accepts_comment_skip_lines_option
    assert_nothing_raised(ArgumentError) do
      CSV.new(@sample_data, :skip_lines => /\A\s*#/)
    end
  end

  def test_accepts_comment_defaults_to_nil
    c = CSV.new(@sample_data)
    assert_nil(c.skip_lines)
  end

  class RegexStub
  end

  def test_requires_skip_lines_to_call_match
    regex_stub = RegexStub.new
    assert_raise_with_message(ArgumentError, /skip_lines/) do
      CSV.new(@sample_data, :skip_lines => regex_stub)
    end
  end

  def test_comment_rows_are_ignored
    sample_data = "line,1,a\n#not,a,line\nline,2,b\n   #also,no,line"
    c = CSV.new sample_data, :skip_lines => /\A\s*#/
    assert_equal [["line", "1", "a"], ["line", "2", "b"]], c.each.to_a
  end

  def test_quoted_skip_line_markers_are_ignored
    sample_data = "line,1,a\n\"#not\",a,line\nline,2,b"
    c = CSV.new sample_data, :skip_lines => /\A\s*#/
    assert_equal [["line", "1", "a"], ["#not", "a", "line"], ["line", "2", "b"]], c.each.to_a
  end

  def test_string_works_like_a_regexp
    sample_data = "line,1,a\n#(not,a,line\nline,2,b\n   also,#no,line"
    c = CSV.new sample_data, :skip_lines => "#"
    assert_equal [["line", "1", "a"], ["line", "2", "b"]], c.each.to_a
  end

end