src/about_regular_expressions.rb

   1 # -*- coding: utf-8 -*-
   2 require File.expand_path(File.dirname(__FILE__) + '/edgecase')
   3
   4 class AboutRegularExpressions < EdgeCase::Koan
   5   def test_a_pattern_is_a_regular_expression
   6     assert_equal __(Regexp), /pattern/.class
   7   end
   8
   9   def test_a_regexp_can_search_a_string_for_matching_content
  10     assert_equal __("match"), "some matching content"[/match/]
  11   end
  12
  13   def test_a_failed_match_returns_nil
  14     assert_equal __(nil), "some matching content"[/missing/]
  15   end
  16
  17   # ------------------------------------------------------------------
  18
  19   def test_question_mark_means_optional
  20     assert_equal __("ab"), "abbcccddddeeeee"[/ab?/]
  21     assert_equal __("a"), "abbcccddddeeeee"[/az?/]
  22   end
  23
  24   def test_plus_means_one_or_more
  25     assert_equal __("bccc"), "abbcccddddeeeee"[/bc+/]
  26   end
  27
  28   def test_asterisk_means_zero_or_more
  29     assert_equal __("abb"), "abbcccddddeeeee"[/ab*/]
  30     assert_equal __("a"), "abbcccddddeeeee"[/az*/]
  31     assert_equal __(""), "abbcccddddeeeee"[/z*/]
  32
  33     # THINK ABOUT IT:
  34     #
  35     # When would * fail to match?
  36   end
  37
  38   # THINK ABOUT IT:
  39   #
  40   # We say that the repetition operators above are "greedy."
  41   #
  42   # Why?
  43
  44   # ------------------------------------------------------------------
  45
  46   def test_the_left_most_match_wins
  47     assert_equal __("a"), "abbccc az"[/az*/]
  48   end
  49
  50   # ------------------------------------------------------------------
  51
  52   def test_character_classes_give_options_for_a_character
  53     animals = ["cat", "bat", "rat", "zat"]
  54     assert_equal __(["cat", "bat", "rat"]), animals.select { |a| a[/[cbr]at/] }
  55   end
  56
  57   def test_slash_d_is_a_shortcut_for_a_digit_character_class
  58     assert_equal __("42"), "the number is 42"[/[0123456789]+/]
  59     assert_equal __("42"), "the number is 42"[/\d+/]
  60   end
  61
  62   def test_character_classes_can_include_ranges
  63     assert_equal __("42"), "the number is 42"[/[0-9]+/]
  64   end
  65
  66   def test_slash_s_is_a_shortcut_for_a_whitespace_character_class
  67     assert_equal __(" \t\n"), "space: \t\n"[/\s+/]
  68   end
  69
  70   def test_slash_w_is_a_shortcut_for_a_word_character_class
  71     # NOTE:  This is more like how a programmer might define a word.
  72     assert_equal __("variable_1"), "variable_1 = 42"[/[a-zA-Z0-9_]+/]
  73     assert_equal __("variable_1"), "variable_1 = 42"[/\w+/]
  74   end
  75
  76   def test_period_is_a_shortcut_for_any_non_newline_character
  77     assert_equal __("abc"), "abc\n123"[/a.+/]
  78   end
  79
  80   def test_a_character_class_can_be_negated
  81     assert_equal __("the number is "), "the number is 42"[/[^0-9]+/]
  82   end
  83
  84   def test_shortcut_character_classes_are_negated_with_capitals
  85     assert_equal __("the number is "), "the number is 42"[/\D+/]
  86     assert_equal __("space:"), "space: \t\n"[/\S+/]
  87     # ... a programmer would most likely do
  88     assert_equal __(" = "), "variable_1 = 42"[/[^a-zA-Z0-9_]+/]
  89     assert_equal __(" = "), "variable_1 = 42"[/\W+/]
  90   end
  91
  92   # ------------------------------------------------------------------
  93
  94   def test_slash_a_anchors_to_the_start_of_the_string
  95     assert_equal __("start"), "start end"[/\Astart/]
  96     assert_equal __(nil), "start end"[/\Aend/]
  97   end
  98
  99   def test_slash_z_anchors_to_the_end_of_the_string
 100     assert_equal __("end"), "start end"[/end\z/]
 101     assert_equal __(nil), "start end"[/start\z/]
 102   end
 103
 104   def test_caret_anchors_to_the_start_of_lines
 105     assert_equal __("2"), "num 42\n2 lines"[/^\d+/]
 106   end
 107
 108   def test_dollar_sign_anchors_to_the_end_of_lines
 109     assert_equal __("42"), "2 lines\nnum 42"[/\d+$/]
 110   end
 111
 112   def test_slash_b_anchors_to_a_word_boundary
 113     assert_equal __("vines"), "bovine vines"[/\bvine./]
 114   end
 115
 116   # ------------------------------------------------------------------
 117
 118   def test_parentheses_group_contents
 119     assert_equal __("hahaha"), "ahahaha"[/(ha)+/]
 120   end
 121
 122   # ------------------------------------------------------------------
 123
 124   def test_parentheses_also_capture_matched_content_by_number
 125     assert_equal __("Gray"), "Gray, James"[/(\w+), (\w+)/, 1]
 126     assert_equal __("James"), "Gray, James"[/(\w+), (\w+)/, 2]
 127   end
 128
 129   def test_variables_can_also_be_used_to_access_captures
 130     assert_equal __("Gray, James"), "Name:  Gray, James"[/(\w+), (\w+)/]
 131     assert_equal __("Gray"), $1
 132     assert_equal __("James"), $2
 133   end
 134
 135   # ------------------------------------------------------------------
 136
 137   def test_a_vertical_pipe_means_or
 138     grays = /(James|Dana|Summer) Gray/
 139     assert_equal __("James Gray"), "James Gray"[grays]
 140     assert_equal __("Summer"), "Summer Gray"[grays, 1]
 141     assert_equal __(nil), "Jim Gray"[grays, 1]
 142   end
 143
 144   # THINK ABOUT IT:
 145   #
 146   # Explain the difference between a character class ([...]) and alternation (|).
 147
 148   # ------------------------------------------------------------------
 149
 150   def test_scan_is_like_find_all
 151     assert_equal __(["one", "two", "three"]), "one two-three".scan(/\w+/)
 152   end
 153
 154   def test_sub_is_like_find_and_replace
 155     assert_equal __("one t-three"), "one two-three".sub(/(t\w*)/) { $1[0, 1] }
 156   end
 157
 158   def test_gsub_is_like_find_and_replace_all
 159     assert_equal __("one t-t"), "one two-three".gsub(/(t\w*)/) { $1[0, 1] }
 160   end
 161 end