regexp_property_values 0.2.1 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: bfe93f83643ea9d85f998e3c0cfed5ae3580211a0509a140ef66f99749ea7e83
-  data.tar.gz: e2c7cbda1f656ba625e84b60fe1ac94dd07bea0199342c033b5d7e35057d3a36
+  metadata.gz: 20ea749474673385c6a98a9cfb2a6f4265d84ce688f69d3068ee3ba351ddb18a
+  data.tar.gz: 8cc47f0492ac633689a75bbf92a3b50fa7c71c3e5b36dd72be0d2c50f6b375d6
 SHA512:
-  metadata.gz: 5b3bb49079c6d969db2c32ef682bd33bb7bc8ca73eded912d27dd3b146735de8e1c239bb2b7657901cce7d5172a3c3b47f0e04d97221b64b83c9e7107b0fc12f
-  data.tar.gz: 1c871837a405a1b3944a00e3ab256872a3e22b21f66a7237801b3efe4fb7783286127b59a6587c7e99e76e02dc14f27f13e854ceae0b848f3c5fc23e4542b018
+  metadata.gz: 6dd88b94bc75382f973bfe7fc402bcf52183dcc93ed05630b02c9f2890bce40bc01ce07a9e98e02b10f592da3394e2faceeb173a7a7b3995f1a0f14f1698d9e6
+  data.tar.gz: cfa31e9a424a1dc385bc8719605566ace765fa4fbaf5ab594d7fe09c09028db5e1b2aea5a6eaaef11e579626ff849f89f5c2e88a44e368567295712be487fef9

data/.gitignore

@@ -1,3 +1,4 @@
+*.bundle
 *.gem
 *.iml
 *.stTheme.cache

data/.travis.yml

@@ -1,9 +1,10 @@
 sudo: false
 language: ruby
 rvm:
-  - 1.9.3
-  - 2.0.0
-  - 2.2.0
-  - 2.5.0
-  - 2.6.0
-before_install: gem install bundler -v 1.16.1
+  - 2.0
+  - 2.4
+  - 2.5
+  - 2.6
+before_install:
+  - gem update --system
+  - gem install bundler

data/README.md

@@ -3,9 +3,9 @@
 [![Gem Version](https://badge.fury.io/rb/regexp_property_values.svg)](http://badge.fury.io/rb/regexp_property_values)
 [![Build Status](https://travis-ci.org/janosch-x/regexp_property_values.svg?branch=master)](https://travis-ci.org/janosch-x/regexp_property_values)
 
-This microlibrary lets you see which property values are supported by the regular expression engine of the Ruby version you are running.
+This small library lets you see which property values are supported by the regular expression engine of the Ruby version you are running and directly reads out their codepoint ranges from there.
 
-That is, it determines all supported values for `\p{value}` expressions.
+That is, it determines all supported values for `\p{value}` expressions and what they match.
 
 ## Usage
 
@@ -17,10 +17,7 @@ require 'regexp_property_values'
 PV = RegexpPropertyValues
 
 PV.all # => ["Alpha", "Blank", "Cntrl", ...]
-PV.all.sort # => ["AHex", "ASCII", "Adlam", "Adlm", "Age=1.1", ...]
-
 PV.by_category # => {"POSIX brackets" => ["Alpha", ...], "Special" => ...}
-
 PV.short_and_long_names # => [["M", "Grek", ...], ["Mark", "Greek", ...]]
 ```
 
@@ -28,23 +25,27 @@ PV.short_and_long_names # => [["M", "Grek", ...], ["Mark", "Greek", ...]]
 
 ```ruby
 PV.all_for_current_ruby # => ["Alpha", "Blank", "Cntrl", ...]
-PV.all_for_current_ruby.include?('Newline') # => false
 
 PV.by_category.map { |k, v| [k, v.select(&:supported_by_current_ruby?] }
 
-PV.short_and_long_names.map { |a| a.select(&:supported_by_current_ruby?) }
+# etc.
 ```
 
-##### Utility methods
+##### Inspect properties
 
 ```ruby
-PV.supported_by_current_ruby?('alpha') # => true
-PV.supported_by_current_ruby?('foobar') # => false
+PV['alpha'].supported_by_current_ruby? # => true
+PV['foobar'].supported_by_current_ruby? # => false
 
-# this one takes a second
-PV.matched_characters('AHex') # => %w[0 1 2 3 4 5 6 7 8 9 A B C ...]
+PV['AHex'].matched_characters # => %w[0 1 2 3 4 5 6 7 8 9 A B C ...]
+PV['AHex'].matched_codepoints # => [48, 49, 50, ...]
+PV['AHex'].matched_ranges # => [48..57, 65..70, 97..102]
+```
 
-# this one takes a minute or two
+##### Utility methods
+
+```ruby
+# This one takes a few seconds (or minutes, without the C extension)
 PV.alias_hash # => {"M" => "Mark", "Grek" => "Greek", ...}
 
 # download the latest list of possible properties

data/Rakefile

@@ -4,3 +4,12 @@ require "rspec/core/rake_task"
 RSpec::Core::RakeTask.new(:spec)
 
 task :default => :spec
+
+require 'rake/extensiontask'
+
+Rake::ExtensionTask.new('regexp_property_values') do |ext|
+  ext.lib_dir = 'lib/regexp_property_values'
+end
+
+# recompile before running specs
+task(:spec).enhance([:compile])

data/ext/regexp_property_values/extconf.rb

@@ -0,0 +1,5 @@
+require 'mkmf'
+
+name = 'regexp_property_values'
+
+create_makefile("#{name}/#{name}")

data/ext/regexp_property_values/regexp_property_values.c

@@ -0,0 +1,56 @@
+#include "ruby.h"
+#include "ruby/encoding.h"
+#include "ruby/oniguruma.h" // still in recent rubies f. backwards compatibility
+
+static int prop_name_to_ctype(char* name, rb_encoding *enc) {
+  UChar *uname;
+  int ctype;
+
+  uname = (UChar*)name;
+  ctype = ONIGENC_PROPERTY_NAME_TO_CTYPE(enc, uname, uname + strlen(name));
+  if (ctype < 0) rb_raise(rb_eArgError, "Unknown property name `%s`", name);
+
+  return ctype;
+}
+
+VALUE onig_ranges_to_rb(const OnigCodePoint *onig_ranges) {
+  unsigned int range_count, i;
+  VALUE result, sub_range;
+
+  range_count = onig_ranges[0];
+  result = rb_ary_new2(range_count); // rb_ary_new_capa not avail. in Ruby 2.0
+
+  for (i = 0; i < range_count; i++) {
+    sub_range = rb_range_new(INT2FIX(onig_ranges[(i * 2) + 1]),
+                             INT2FIX(onig_ranges[(i * 2) + 2]),
+                             0);
+    rb_ary_store(result, i, sub_range);
+  }
+
+  return result;
+}
+
+VALUE rb_prop_ranges(char* name, rb_encoding *enc) {
+  int ctype;
+  const OnigCodePoint *onig_ranges;
+  OnigCodePoint sb_out;
+
+  ctype = prop_name_to_ctype(name, enc);
+  ONIGENC_GET_CTYPE_CODE_RANGE(enc, ctype, &sb_out, &onig_ranges);
+  return onig_ranges_to_rb(onig_ranges);
+}
+
+VALUE method_matched_ranges(VALUE self, VALUE arg) {
+  char *prop_name;
+  rb_encoding *enc;
+
+  prop_name = StringValueCStr(arg);
+  enc = rb_enc_get(arg);
+  return rb_prop_ranges(prop_name, enc);
+}
+
+void Init_regexp_property_values() {
+  VALUE module;
+  module = rb_define_module("OnigRegexpPropertyHelper");
+  rb_define_singleton_method(module, "matched_ranges", method_matched_ranges, 1);
+}

data/lib/regexp_property_values/extension.rb

@@ -0,0 +1,53 @@
+module RegexpPropertyValues
+  module Extension
+    def supported_by_current_ruby?
+      !!regexp
+    rescue RegexpError, SyntaxError
+      false
+    end
+
+    def regexp
+      @regexp ||= /\p{#{self}}/u
+    end
+
+    if const_defined?(:OnigRegexpPropertyHelper)
+      # C extension loaded
+
+      def matched_codepoints
+        matched_ranges.flat_map(&:to_a)
+      end
+
+      def matched_ranges
+        OnigRegexpPropertyHelper.matched_ranges(self)
+      end
+
+      def matched_characters
+        matched_codepoints.map { |cp| cp.chr('utf-8') }
+      end
+    else
+      # Ruby fallback - this stuff is slow as hell, and it wont get much faster
+
+      def matched_codepoints
+        matched_characters.map(&:ord)
+      end
+
+      def matched_ranges
+        require 'set'
+        matched_characters
+          .to_set(SortedSet)
+          .divide { |i, j| (i - j).abs == 1 }
+          .map { |s| a = s.to_a; a.first..a.last }
+      end
+
+      def matched_characters
+        regexp.respond_to?(:match?) ||
+          regexp.define_singleton_method(:match?) { |str| !!match(str) }
+
+        @@characters ||= ((0..0xD7FF).to_a + (0xE000..0x10FFFF).to_a)
+                         .map { |cp_number| [cp_number].pack('U') }
+
+        @@characters.select { |char| regexp.match?(char) }
+      end
+    end
+  end
+end

data/lib/regexp_property_values/version.rb

@@ -1,3 +1,3 @@
 module RegexpPropertyValues
-  VERSION = '0.2.1'
+  VERSION = '0.3.0'
 end

data/lib/regexp_property_values.rb

@@ -1,4 +1,9 @@
-require 'regexp_property_values/value_extension'
+begin
+  require 'regexp_property_values/regexp_property_values'
+rescue LoadError
+  warn 'regexp_property_values could not load C extension, using slower Ruby'
+end
+require 'regexp_property_values/extension'
 require 'regexp_property_values/version'
 
 module RegexpPropertyValues
@@ -26,21 +31,20 @@ module RegexpPropertyValues
   end
 
   def by_category
-    result = File.foreach(file_path).inject({}) do |hash, line|
+    result = File.foreach(file_path).each_with_object({}) do |line, hash|
       if /^\* (?<category>\S.+)/ =~ line
         @current_category = category
         hash[@current_category] ||= []
       elsif /^ {4}(?<value_name>\S.*)/ =~ line
-        hash[@current_category] << value(value_name)
+        hash[@current_category] << value_name.extend(Extension)
       end
-      hash
     end
     add_oniguruma_properties(result)
     result
   end
 
   def add_oniguruma_properties(props_by_category)
-    props_by_category['Special'] << value('Newline')
+    props_by_category['Special'] << 'Newline'.extend(Extension)
   end
 
   def alias_hash
@@ -48,11 +52,10 @@ module RegexpPropertyValues
     return {} if short_names.empty?
 
     long_names -= by_category['POSIX brackets']
-    by_matched_characters.each_value.inject({}) do |hash, props|
-      next hash if props.count < 2
+    by_matched_codepoints.each_value.each_with_object({}) do |props, hash|
+      next if props.count < 2
       long_name = (props & long_names)[0] || fail("no long name for #{props}")
       (props & short_names).each { |short_name| hash[short_name] = long_name }
-      hash
     end
   end
 
@@ -60,27 +63,17 @@ module RegexpPropertyValues
     short_name_categories = ['Major and General Categories',
                              'PropertyAliases',
                              'PropertyValueAliases (Script)']
-    by_category.inject([[], []]) do |(short, long), (cat_name, props)|
+    by_category.each_with_object([[], []]) do |(cat_name, props), (short, long)|
       (short_name_categories.include?(cat_name) ? short : long).concat(props)
-      [short, long]
     end
   end
 
-  def by_matched_characters
-    puts 'Establishing property characters, this may take a bit ...'
-    all_for_current_ruby.group_by(&:matched_characters)
-  end
-
-  def matched_characters(prop)
-    value(prop).matched_characters
-  end
-
-  def supported_by_current_ruby?(prop)
-    value(prop).supported_by_current_ruby?
+  def by_matched_codepoints
+    puts 'Establishing property codepoints, this may take a bit ...'
+    all_for_current_ruby.group_by(&:matched_codepoints)
   end
 
-  def value(prop)
-    prop.singleton_class.send(:include, ValueExtension)
-    prop
+  def [](prop)
+    prop.extend(Extension)
   end
 end

data/regexp_property_values.gemspec

@@ -8,11 +8,10 @@ Gem::Specification.new do |s|
   s.authors       = ['Janosch Müller']
   s.email         = ['janosch84@gmail.com']
 
-  s.summary       = "Lists property values supported by Ruby's regex engine"
-  s.description   = 'This microlibrary lets you see which property values are '\
-                    'supported by the regular expression engine of the Ruby '\
-                    'version you are running. That is, it determines all '\
-                    'supported values for `\p{value}` expressions.'
+  s.summary       = "Inspect property values supported by Ruby's regex engine"
+  s.description   = 'This small library lets you see which property values '\
+                    'are supported by the regular expression engine of the '\
+                    'Ruby version you are running, and what they match.'
   s.homepage      = 'https://github.com/janosch-x/regexp_property_values'
   s.license       = 'MIT'
 
@@ -21,7 +20,12 @@ Gem::Specification.new do |s|
   end
   s.require_paths = ['lib']
 
+  s.extensions = %w[ext/regexp_property_values/extconf.rb]
+
+  s.required_ruby_version = '>= 2.0.0'
+
   s.add_development_dependency 'bundler', '~> 1.16'
   s.add_development_dependency 'rake', '~> 10.0'
+  s.add_development_dependency 'rake-compiler', '~> 1.0'
   s.add_development_dependency 'rspec', '~> 3.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: regexp_property_values
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.0
 platform: ruby
 authors:
 - Janosch Müller
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-05-13 00:00:00.000000000 Z
+date: 2018-07-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -38,6 +38,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '10.0'
+- !ruby/object:Gem::Dependency
+  name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -52,13 +66,14 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '3.0'
-description: This microlibrary lets you see which property values are supported by
-  the regular expression engine of the Ruby version you are running. That is, it determines
-  all supported values for `\p{value}` expressions.
+description: This small library lets you see which property values are supported by
+  the regular expression engine of the Ruby version you are running, and what they
+  match.
 email:
 - janosch84@gmail.com
 executables: []
-extensions: []
+extensions:
+- ext/regexp_property_values/extconf.rb
 extra_rdoc_files: []
 files:
 - ".gitignore"
@@ -70,9 +85,11 @@ files:
 - Rakefile
 - bin/console
 - bin/setup
+- ext/regexp_property_values/extconf.rb
+- ext/regexp_property_values/regexp_property_values.c
 - lib/UnicodeProps.txt
 - lib/regexp_property_values.rb
-- lib/regexp_property_values/value_extension.rb
+- lib/regexp_property_values/extension.rb
 - lib/regexp_property_values/version.rb
 - regexp_property_values.gemspec
 homepage: https://github.com/janosch-x/regexp_property_values
@@ -87,7 +104,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '0'
+      version: 2.0.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
@@ -98,5 +115,5 @@ rubyforge_project:
 rubygems_version: 2.7.6
 signing_key: 
 specification_version: 4
-summary: Lists property values supported by Ruby's regex engine
+summary: Inspect property values supported by Ruby's regex engine
 test_files: []

data/lib/regexp_property_values/value_extension.rb

@@ -1,20 +0,0 @@
-module RegexpPropertyValues
-  def self.characters
-    @characters ||= ((0..55_295).to_a + (57_344..1_114_111).to_a)
-                    .map { |cp_number| [cp_number].pack('U') }
-  end
-
-  module ValueExtension
-    def supported_by_current_ruby?
-      begin !!regexp; rescue RegexpError, SyntaxError; false end
-    end
-
-    def matched_characters
-      RegexpPropertyValues.characters.select { |char| regexp.match(char) }
-    end
-
-    def regexp
-      @regexp ||= /\p{#{self}}/u
-    end
-  end
-end