Merge pull request #40 from tom-lord/use_regexp_property_values

Use `regexp_property_values` gem for accurate property mapping

Author: tom-lord

.ruby-version

@@ -1 +1 @@
-3.0.0
+3.1.2

Gemfile

@@ -3,7 +3,6 @@ source 'https://rubygems.org'
 group :test do
   gem 'rspec'
   gem 'coveralls', require: false
-  gem 'pry'
 end
 
 # Specify your gem's dependencies in regexp-examples.gemspec

lib/regexp-examples.rb

@@ -1,4 +1,3 @@
-require_relative 'regexp-examples/unicode_char_ranges'
 require_relative 'regexp-examples/chargroup_parser'
 require_relative 'regexp-examples/config'
 require_relative 'regexp-examples/char_sets'
@@ -8,6 +7,5 @@
 require_relative 'regexp-examples/helpers'
 require_relative 'regexp-examples/parser'
 require_relative 'regexp-examples/repeaters'
-require_relative 'regexp-examples/unicode_char_ranges'
 require_relative 'regexp-examples/version'
 require_relative 'core_extensions/regexp/examples'

lib/regexp-examples/char_sets.rb

@@ -53,7 +53,5 @@ module CharSets
       'word'   => Word,
       'ascii'  => Any
     }.freeze
-
-    NamedPropertyCharMap = UnicodeCharRanges.instance
   end.freeze
 end

lib/regexp-examples/parser_helpers/charset_negation_helper.rb

@@ -2,7 +2,7 @@
 module RegexpExamples
   module CharsetNegationHelper
     def negate_if(charset, is_negative)
-      is_negative ? (CharSets::Any.dup - charset) : charset
+      is_negative ? (CharSets::Any.dup - charset.to_a) : charset
     end
   end
 end

lib/regexp-examples/parser_helpers/parse_after_backslash_group_helper.rb

@@ -1,3 +1,5 @@
+require 'regexp_property_values'
+
 module RegexpExamples
   # A collection of related helper methods, utilised by the `Parser` class
   module ParseAfterBackslashGroupHelper
@@ -91,13 +93,19 @@ def parse_backslash_unicode_sequence(full_hex_sequence)
     end
 
     def parse_backslash_named_property(p_negation, caret_negation, property_name)
-      @current_position += (caret_negation.length + # 0 or 1, of '^' is present
+      @current_position += (caret_negation.length + # 0 or 1, if '^' is present
                             property_name.length +
                             2) # Length of opening and closing brackets (always 2)
       # Beware of double negatives! E.g. /\P{^Space}/
       is_negative = (p_negation == 'P') ^ (caret_negation == '^')
       CharGroup.new(
-        negate_if(CharSets::NamedPropertyCharMap[property_name.downcase], is_negative),
+        negate_if(
+          RegexpPropertyValues[property_name]
+            .matched_codepoints
+            .lazy
+            .filter_map { |cp| cp.chr('utf-8') unless cp.between?(0xD800, 0xDFFF) },
+          is_negative
+        ),
         @ignorecase
       )
     end

lib/regexp-examples/unicode_char_ranges.rb

@@ -14,9 +14,10 @@ Gem::Specification.new do |s|
   s.test_files       = s.files.grep(/^(test|spec|features)\//)
   s.require_paths    = ['lib']
   s.homepage         = 'http://rubygems.org/gems/regexp-examples'
-  s.add_development_dependency 'bundler', '> 1.7'
-  s.add_development_dependency 'rake', '~> 12.0'
-  s.add_development_dependency 'pry', '~> 0.12.0'
+  s.add_dependency 'regexp_property_values', '~> 1.5'
+  s.add_development_dependency 'bundler', '~> 2.4'
+  s.add_development_dependency 'rake', '~> 13.0'
+  s.add_development_dependency 'pry'
   s.add_development_dependency 'warning', '~> 0.10.0'
   s.license          = 'MIT'
   s.required_ruby_version = '>= 2.4.0'

regexp-examples.gemspec

@@ -190,37 +190,24 @@ def self.examples_are_empty(*regexps)
         /\P{Ll}/, # Negation syntax type 2
         /\P{^Ll}/ # Double negation!! (Should cancel out)
       )
-      # An exhaustive set of tests for all named properties!!! This is useful
-      # for verifying the PStore contains correct values for all ruby versions
-      %w[
-        Alnum Alpha Blank Cntrl Digit Graph Lower Print Punct Space Upper XDigit
-        Word ASCII Any Assigned L Ll Lm Lo Lt Lu M Mn Mc Me N Nd Nl No P Pc Pd
-        Ps Pe Pi Pf Po S Sm Sc Sk So Z Zs Zl Zp C Cc Cf Cn Co Arabic Armenian
-        Balinese Bengali Bopomofo Braille Buginese Buhid Canadian_Aboriginal
-        Cham Cherokee Common Coptic Cyrillic Devanagari Ethiopic Georgian
-        Glagolitic Greek Gujarati Gurmukhi Han Hangul Hanunoo Hebrew Hiragana
-        Inherited Kannada Katakana Kayah_Li Khmer Lao Latin Lepcha Limbu Malayalam
-        Mongolian Myanmar New_Tai_Lue Nko Ogham Ol_Chiki Oriya Phags_Pa Rejang
-        Runic Saurashtra Sinhala Sundanese Syloti_Nagri Syriac Tagalog Tagbanwa
-        Tai_Le Tamil Telugu Thaana Thai Tibetan Tifinagh Vai Yi
-      ].each do |property|
-        it "examples for /\p{#{property}}/" do
-          regexp_examples = /\p{#{property}}/.examples(max_group_results: 99_999)
-          expect(regexp_examples)
-            .not_to be_empty,
-                    "No examples were generated for regexp: /\p{#{property}}/"
-          # Just do one big check, for test system performance (~30% faster)
-          # (Otherwise, we're doing up to 128 checks on 123 properties!!!)
-          expect(regexp_examples.join('')).to match(/\A\p{#{property}}+\z/)
-        end
-      end
 
-      # The following seem to genuinely have no matching examples (!!??!!?!)
-      %w[
-        Cs Carian Cuneiform Cypriot Deseret Gothic Kharoshthi Linear_B Lycian
-        Lydian Old_Italic Old_Persian Osmanya Phoenician Shavian Ugaritic
-      ].each do |property|
-        examples_are_empty(/\p{#{property}}/)
+      expected_empty_properties = %w[surrogate inlowsurrogates inhighsurrogates inhighprivateusesurrogates]
+
+      RegexpPropertyValues.all_for_current_ruby.map(&:identifier).each do |property|
+        if(expected_empty_properties).include?(property)
+          examples_are_empty(/\p{#{property}}/)
+        else
+          it "examples for /\p{#{property}}/" do
+            regexp_examples = /\p{#{property}}/.examples(max_group_results: 99_999)
+
+            expect(regexp_examples)
+              .not_to be_empty,
+            "No examples were generated for regexp: /\p{#{property}}/"
+              # Just do one big check, for test system performance (~30% faster)
+              # (Otherwise, we're potentially doing 99999 checks on 123 properties!!!)
+              expect(regexp_examples.join('')).to match(/\A\p{#{property}}+\z/)
+          end
+        end
       end
     end