Koans AboutRegularExpressions

class AboutRegularExpressions < Neo::Koan
  def test_a_pattern_is_a_regular_expression
    assert_equal Regexp, /pattern/.class
  end

  def test_a_regexp_can_search_a_string_for_matching_content
    assert_equal "match", "some matching content"[/match/]
  end

  def test_a_failed_match_returns_nil
    assert_equal nil, "some matching content"[/missing/]
  end

Regular Expressions, or Regexp, are usually indicated with /.../ and test whether a string contains a given pattern, or extract the portions that match. If no match is found, nil is returned.

def test_question_mark_means_optional
    assert_equal "ab", "abbcccddddeeeee"[/ab?/]
    assert_equal "a", "abbcccddddeeeee"[/az?/]
  end

  def test_plus_means_one_or_more
    assert_equal "bccc", "abbcccddddeeeee"[/bc+/]
  end

  def test_asterisk_means_zero_or_more
    assert_equal "abb", "abbcccddddeeeee"[/ab*/]
    assert_equal "a", "abbcccddddeeeee"[/az*/]
    assert_equal "", "abbcccddddeeeee"[/z*/]
  end

?, +, and * are repetition metacharacters called "quantifiers" and are used to specify how many times a construct needs to occur. ? indicates that it is optional, + means one or more, and * means zero or more.

  def test_the_left_most_match_wins
    assert_equal "a", "abbccc az"[/az*/]
  end

  def test_character_classes_give_options_for_a_character
    animals = ["cat", "bat", "rat", "zat"]
    assert_equal ["cat", "bat", "rat"], animals.select { |a| a[/[cbr]at/] }
  end

Instead of expressing every character you'd like to match, you may instead provide options for the construct to return.

  def test_slash_d_is_a_shortcut_for_a_digit_character_class
    assert_equal "42", "the number is 42"[/[0123456789]+/]
    assert_equal "42", "the number is 42"[/\d+/]
  end

  def test_character_classes_can_include_ranges
    assert_equal "42", "the number is 42"[/[0-9]+/]
  end

  def test_slash_s_is_a_shortcut_for_a_whitespace_character_class
    assert_equal " \t\n", "space: \t\n"[/\s+/]
  end

  def test_slash_w_is_a_shortcut_for_a_word_character_class
    # NOTE:  This is more like how a programmer might define a word.
    assert_equal "variable_1", "variable_1 = 42"[/[a-zA-Z0-9_]+/]
    assert_equal "variable_1", "variable_1 = 42"[/\w+/]
  end

  def test_period_is_a_shortcut_for_any_non_newline_character
    assert_equal "abc", "abc\n123"[/a.+/]
  end

  def test_a_character_class_can_be_negated
    assert_equal "the number is ", "the number is 42"[/[^0-9]+/]
  end

  def test_shortcut_character_classes_are_negated_with_capitals
    assert_equal "the number is ", "the number is 42"[/\D+/]
    assert_equal "space:", "space: \t\n"[/\S+/]
    # ... a programmer would most likely do
    assert_equal " = ", "variable_1 = 42"[/[^a-zA-Z0-9_]+/]
    assert_equal " = ", "variable_1 = 42"[/\W+/]
  end

You can return matches for only digits and the shorthand for that instead of expressing every digit is to write \d+. If you'd like to return a matching digit in a certain range, you may do that as well ex. [0-9]. You can also opt to negate what should be returned using ^ which returns everything in the string except what you indicated to negate. For character classes, you can use a capital W to negate `ex. /\W/.

def test_slash_a_anchors_to_the_start_of_the_string
    assert_equal "start", "start end"[/\Astart/]
    assert_equal nil, "start end"[/\Aend/]
  end

  def test_slash_z_anchors_to_the_end_of_the_string
    assert_equal "end", "start end"[/end\z/]
    assert_equal nil, "start end"[/start\z/]
  end

  def test_caret_anchors_to_the_start_of_lines
    assert_equal "2", "num 42\n2 lines"[/^\d+/]
  end

  def test_dollar_sign_anchors_to_the_end_of_lines
    assert_equal "42", "2 lines\nnum 42"[/\d+$/]
  end

  def test_slash_b_anchors_to_a_word_boundary
    assert_equal "vines", "bovine vines"[/\bvine./]
  end

Anchors are metacharacters that match positions between characters and anchor the match to a specific position. For example \A matches to the beginning of a string,\Z\ matches to the end of a string, ^ matches to the start of a line, $ matches to the end of a line, and \b anchors to a word boundary.

  def test_parentheses_group_contents
    assert_equal "hahaha", "ahahaha"[/(ha)+/]
  end

  def test_parentheses_also_capture_matched_content_by_number
    assert_equal "Gray", "Gray, James"[/(\w+), (\w+)/, 1]
    assert_equal "James", "Gray, James"[/(\w+), (\w+)/, 2]
  end

  def test_variables_can_also_be_used_to_access_captures
    assert_equal "Gray, James", "Name:  Gray, James"[/(\w+), (\w+)/]
    assert_equal "Gray", $1
    assert_equal "James", $2
  end

  def test_a_vertical_pipe_means_or
    grays = /(James|Dana|Summer) Gray/
    assert_equal "James Gray", "James Gray"[grays]
    assert_equal "Summer", "Summer Gray"[grays, 1]
    assert_equal nil, "Jim Gray"[grays, 1]
  end

  def test_scan_is_like_find_all
    assert_equal ["one", "two", "three"], "one two-three".scan(/\w+/)
  end

  def test_sub_is_like_find_and_replace
    assert_equal "one t-three", "one two-three".sub(/(t\w*)/) { $1[0, 1] }
  end

  def test_gsub_is_like_find_and_replace_all
    assert_equal "one t-t", "one two-three".gsub(/(t\w*)/) { $1[0, 1] }
  end
end

Using parentheses can be used to match a group of characters and can even allow you to match by position number. The vertical bar metacharacter combines multiple expressions into a single one that matches either of the expressions. Subs and GSubs may also be used to not return specific portions of a match.