regexp_property_values 0.3.5 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 26eb46070df0c2f41fe773860140f0b9fea4f9512284a49f2dcac8d77d7055e5
-  data.tar.gz: d7f31346d39f7d00d0cc3d837b0f32685c2eec56f4e784b641428959134506c2
+  metadata.gz: 3295d2f712a9e6aca9af41f2d4b98775dfc39cec0c9dc026fcd50999d4757dd2
+  data.tar.gz: 2203633248c4cfb39489edede3be46fc3f935f8bbb6d958c9bd962ecf268a593
 SHA512:
-  metadata.gz: 5f425e9a3d570bc612b75f1b113a2077cfeb35e106406655344a13d7c9cd6b74cdc0ce59552cd6017f3b6db3b654877f0cfe6ef02ef60d463c83158a7f51b23a
-  data.tar.gz: bdf9f9fc7c2a8077af610f2285672fcc4af78fb774542c8cfcf61c8a1ccf03d4894dd79729119c5084ccae752171c92c96430f7d1a35a4214857389f59e9eba9
+  metadata.gz: eb72ccf725571146439af6ebaaec6a38420b39b19348827f08f3ca230af326a32653f05b4820f2fd66d98c5e7f7b394a4790e03badb95530311e1a5def209e2d
+  data.tar.gz: 26d720ddd4c306223e38e944b7b08cc4c7170b39c120b906766d599b77fc402d560aa1dcbd8921bad30301536800b29cbfc39d525318827561efe162343a08a5

data/.gitignore

@@ -20,6 +20,7 @@ binstubs/*
 bundler_stubs/*/.yardoc
 Gemfile.lock
 /.bundle/
+/.vscode/
 /_yardoc/
 /coverage/
 /doc/

data/.travis.yml

@@ -6,6 +6,3 @@ rvm:
   - 2.5
   - 2.6
   - jruby-9.1.9.0
-before_install:
-  - gem update --system
-  - gem install bundler

data/CHANGELOG.md

@@ -0,0 +1,19 @@
+# Changelog
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
+and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
+
+## [1.0.0] - 2019-06-16
+
+### Changed
+- removed `::by_category`, `::by_matched_codepoints`, `::short_and_long_names`
+- return values are now always of a custom `Value` class, no longer extended `Strings`
+- unknown properties now raise `RegexpPropertyValues::Error`, no longer an `ArgumentError`
+
+### Added
+- `Value#identifier`
+- `Value#full_name`
+
+### Fixed
+- better codepoint determination speed for non-C Rubies (still slow)

data/README.md

@@ -1,7 +1,7 @@
 # RegexpPropertyValues
 
 [![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)
+[![Build Status](https://travis-ci.org/jaynetics/regexp_property_values.svg?branch=master)](https://travis-ci.org/jaynetics/regexp_property_values)
 
 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.
 
@@ -16,22 +16,16 @@ require 'regexp_property_values'
 
 PV = RegexpPropertyValues
 
-PV.all # => ["Alpha", "Blank", "Cntrl", ...]
-PV.by_category # => {"POSIX brackets" => ["Alpha", ...], "Special" => ...}
-PV.short_and_long_names # => [["M", "Grek", ...], ["Mark", "Greek", ...]]
+PV.all # => [<Value name='Alpha'>, <Value name='Blank'>, ...]
 ```
 
 ##### Browse property values supported by the Ruby you are running
 
 ```ruby
-PV.all_for_current_ruby # => ["Alpha", "Blank", "Cntrl", ...]
-
-PV.by_category.map { |k, v| [k, v.select(&:supported_by_current_ruby?] }
-
-# etc.
+PV.all_for_current_ruby # => [<Value name='Alpha'>, <Value name='Blank'>, ...]
 ```
 
-##### Inspect properties
+##### Inspect property values
 
 ```ruby
 PV['alpha'].supported_by_current_ruby? # => true
@@ -42,7 +36,7 @@ PV['AHex'].matched_codepoints # => [48, 49, 50, ...]
 PV['AHex'].matched_ranges # => [48..57, 65..70, 97..102]
 ```
 
-If [`character_set`](https://github.com/janosch-x/character_set) is installed, you can also do this:
+If [`character_set`](https://github.com/jaynetics/character_set) is installed, you can also do this:
 
 ```ruby
 PV['AHex'].character_set # => #<CharacterSet: {48, 49...} (size: 22)>
@@ -51,9 +45,9 @@ PV['AHex'].character_set # => #<CharacterSet: {48, 49...} (size: 22)>
 ##### Utility methods
 
 ```ruby
-# This one takes a few seconds (or minutes, without the C extension)
-PV.alias_hash # => {"M" => "Mark", "Grek" => "Greek", ...}
+# get a Hash of aliases for property names
+PV.alias_hash # => { <Value name='M'> => <Value name='Mark'>, ... }
 
-# download the latest list of possible properties
+# download a list of possible properties for the running Ruby version
 PV.update
 ```

data/Rakefile

@@ -17,6 +17,8 @@ namespace :java do
   java_gemspec.platform = 'java'
   java_gemspec.extensions = []
 
+  java_gemspec.add_runtime_dependency 'range_compressor', '~> 1.0'
+
   Gem::PackageTask.new(java_gemspec) do |pkg|
     pkg.need_zip = true
     pkg.need_tar = true

data/ext/regexp_property_values/regexp_property_values.c

@@ -2,25 +2,29 @@
 #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) {
+static int prop_name_to_ctype(char *name, rb_encoding *enc)
+{
   UChar *uname;
   int ctype;
 
-  uname = (UChar*)name;
+  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);
+  if (ctype < 0)
+    rb_raise(rb_eArgError, "Unknown property name `%s`", name);
 
   return ctype;
 }
 
-VALUE onig_ranges_to_rb(const OnigCodePoint *onig_ranges) {
+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++) {
+  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);
@@ -30,26 +34,28 @@ VALUE onig_ranges_to_rb(const OnigCodePoint *onig_ranges) {
   return result;
 }
 
-VALUE rb_prop_ranges(char* name, rb_encoding *enc) {
+VALUE rb_prop_ranges(char *name)
+{
   int ctype;
   const OnigCodePoint *onig_ranges;
   OnigCodePoint sb_out;
+  rb_encoding *enc;
+  enc = rb_utf8_encoding();
 
   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) {
+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);
+  return rb_prop_ranges(prop_name);
 }
 
-void Init_regexp_property_values() {
+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/aliases

@@ -0,0 +1,233 @@
+AHex;ASCII_Hex_Digit
+Adlm;Adlam
+Aghb;Caucasian_Albanian
+Arab;Arabic
+Armi;Imperial_Aramaic
+Armn;Armenian
+Avst;Avestan
+Bali;Balinese
+Bamu;Bamum
+Bass;Bassa_Vah
+Batk;Batak
+Beng;Bengali
+Bhks;Bhaiksuki
+Bidi_C;Bidi_Control
+Bopo;Bopomofo
+Brah;Brahmi
+Brai;Braille
+Bugi;Buginese
+Buhd;Buhid
+C;Other
+CI;Case_Ignorable
+CWCF;Changes_When_Casefolded
+CWCM;Changes_When_Casemapped
+CWL;Changes_When_Lowercased
+CWT;Changes_When_Titlecased
+CWU;Changes_When_Uppercased
+Cakm;Chakma
+Cans;Canadian_Aboriginal
+Cari;Carian
+Cc;Control
+Cf;Format
+Cher;Cherokee
+Cn;Unassigned
+Co;Private_Use
+Combining_Mark;Mark
+Copt;Coptic
+Cprt;Cypriot
+Cs;Surrogate
+Cyrl;Cyrillic
+DI;Default_Ignorable_Code_Point
+Dep;Deprecated
+Deva;Devanagari
+Dia;Diacritic
+Dogr;Dogra
+Dsrt;Deseret
+Dupl;Duployan
+Egyp;Egyptian_Hieroglyphs
+Elba;Elbasan
+Elym;Elymaic
+Ethi;Ethiopic
+Ext;Extender
+Geor;Georgian
+Glag;Glagolitic
+Gong;Gunjala_Gondi
+Gonm;Masaram_Gondi
+Goth;Gothic
+Gr_Base;Grapheme_Base
+Gr_Ext;Grapheme_Extend
+Gr_Link;Grapheme_Link
+Gran;Grantha
+Grek;Greek
+Gujr;Gujarati
+Guru;Gurmukhi
+Hang;Hangul
+Hani;Han
+Hano;Hanunoo
+Hatr;Hatran
+Hebr;Hebrew
+Hex;Hex_Digit
+Hira;Hiragana
+Hluw;Anatolian_Hieroglyphs
+Hmng;Pahawh_Hmong
+Hmnp;Nyiakeng_Puachue_Hmong
+Hung;Old_Hungarian
+IDC;ID_Continue
+IDS;ID_Start
+IDSB;IDS_Binary_Operator
+IDST;IDS_Trinary_Operator
+Ideo;Ideographic
+Ital;Old_Italic
+Java;Javanese
+Join_C;Join_Control
+Kali;Kayah_Li
+Kana;Katakana
+Khar;Kharoshthi
+Khmr;Khmer
+Khoj;Khojki
+Knda;Kannada
+Kthi;Kaithi
+L;Letter
+LC;Cased_Letter
+LOE;Logical_Order_Exception
+Lana;Tai_Tham
+Laoo;Lao
+Latn;Latin
+Lepc;Lepcha
+Limb;Limbu
+Lina;Linear_A
+Linb;Linear_B
+Ll;Lowercase_Letter
+Lm;Modifier_Letter
+Lo;Other_Letter
+Lt;Titlecase_Letter
+Lu;Uppercase_Letter
+Lyci;Lycian
+Lydi;Lydian
+M;Mark
+Mahj;Mahajani
+Maka;Makasar
+Mand;Mandaic
+Mani;Manichaean
+Marc;Marchen
+Mc;Spacing_Mark
+Me;Enclosing_Mark
+Medf;Medefaidrin
+Mend;Mende_Kikakui
+Merc;Meroitic_Cursive
+Mero;Meroitic_Hieroglyphs
+Mlym;Malayalam
+Mn;Nonspacing_Mark
+Mong;Mongolian
+Mroo;Mro
+Mtei;Meetei_Mayek
+Mult;Multani
+Mymr;Myanmar
+N;Number
+NChar;Noncharacter_Code_Point
+Nand;Nandinagari
+Narb;Old_North_Arabian
+Nbat;Nabataean
+Nd;Decimal_Number
+Nkoo;Nko
+Nl;Letter_Number
+No;Other_Number
+Nshu;Nushu
+OAlpha;Other_Alphabetic
+ODI;Other_Default_Ignorable_Code_Point
+OGr_Ext;Other_Grapheme_Extend
+OIDC;Other_ID_Continue
+OIDS;Other_ID_Start
+OLower;Other_Lowercase
+OMath;Other_Math
+OUpper;Other_Uppercase
+Ogam;Ogham
+Olck;Ol_Chiki
+Orkh;Old_Turkic
+Orya;Oriya
+Osge;Osage
+Osma;Osmanya
+P;Punctuation
+PCM;Prepended_Concatenation_Mark
+Palm;Palmyrene
+Pat_Syn;Pattern_Syntax
+Pat_WS;Pattern_White_Space
+Pauc;Pau_Cin_Hau
+Pc;Connector_Punctuation
+Pd;Dash_Punctuation
+Pe;Close_Punctuation
+Perm;Old_Permic
+Pf;Final_Punctuation
+Phag;Phags_Pa
+Phli;Inscriptional_Pahlavi
+Phlp;Psalter_Pahlavi
+Phnx;Phoenician
+Pi;Initial_Punctuation
+Plrd;Miao
+Po;Other_Punctuation
+Prti;Inscriptional_Parthian
+Ps;Open_Punctuation
+QMark;Quotation_Mark
+Qaac;Coptic
+Qaai;Inherited
+RI;Regional_Indicator
+Rjng;Rejang
+Rohg;Hanifi_Rohingya
+Runr;Runic
+S;Symbol
+SD;Soft_Dotted
+STerm;Sentence_Terminal
+Samr;Samaritan
+Sarb;Old_South_Arabian
+Saur;Saurashtra
+Sc;Currency_Symbol
+Sgnw;SignWriting
+Shaw;Shavian
+Shrd;Sharada
+Sidd;Siddham
+Sind;Khudawadi
+Sinh;Sinhala
+Sk;Modifier_Symbol
+Sm;Math_Symbol
+So;Other_Symbol
+Sogd;Sogdian
+Sogo;Old_Sogdian
+Sora;Sora_Sompeng
+Soyo;Soyombo
+Sund;Sundanese
+Sylo;Syloti_Nagri
+Syrc;Syriac
+Tagb;Tagbanwa
+Takr;Takri
+Tale;Tai_Le
+Talu;New_Tai_Lue
+Taml;Tamil
+Tang;Tangut
+Tavt;Tai_Viet
+Telu;Telugu
+Term;Terminal_Punctuation
+Tfng;Tifinagh
+Tglg;Tagalog
+Thaa;Thaana
+Tibt;Tibetan
+Tirh;Tirhuta
+UIdeo;Unified_Ideograph
+Ugar;Ugaritic
+VS;Variation_Selector
+Vaii;Vai
+WSpace;White_Space
+Wara;Warang_Citi
+Wcho;Wancho
+XIDC;XID_Continue
+XIDS;XID_Start
+Xpeo;Old_Persian
+Xsux;Cuneiform
+Yiii;Yi
+Z;Separator
+Zanb;Zanabazar_Square
+Zinh;Inherited
+Zl;Line_Separator
+Zp;Paragraph_Separator
+Zs;Space_Separator
+Zyyy;Common
+Zzzz;Unknown

data/lib/regexp_property_values.rb

@@ -3,77 +3,35 @@ begin
 rescue LoadError
   warn 'regexp_property_values could not load C extension, using slower Ruby'
 end
-require 'regexp_property_values/extension'
+require 'regexp_property_values/updater'
+require 'regexp_property_values/value'
 require 'regexp_property_values/version'
 
 module RegexpPropertyValues
-  module_function
+  Error = Class.new(StandardError)
 
-  LIST_URL = 'https://raw.githubusercontent.com/k-takata/Onigmo/master/doc/UnicodeProps.txt'
+  VALUES_PATH  = File.join(__dir__, 'values')
+  ALIASES_PATH = File.join(__dir__, 'aliases')
 
-  def update
-    puts "Downloading #{LIST_URL}"
-    require 'open-uri'
-    File.open(file_path, 'w') { |f| IO.copy_stream(open(LIST_URL), f) }
-    puts 'Done!'
+  def self.[](name)
+    Value.new(name)
   end
 
-  def file_path
-    File.expand_path('../UnicodeProps.txt', __FILE__)
+  def self.all_for_current_ruby
+    @all_for_current_ruby ||= all.select(&:supported_by_current_ruby?)
   end
 
-  def all
-    by_category.values.flatten
+  def self.all
+    @all ||= File.readlines(VALUES_PATH).map { |line| Value.new(line.chomp) }
   end
 
-  def all_for_current_ruby
-    all.select(&:supported_by_current_ruby?)
+  def self.alias_hash
+    @alias_hash ||= File.readlines(ALIASES_PATH).map do |line|
+      line.chomp.split(';').map { |name| Value.new(name) }
+    end.to_h
   end
 
-  def by_category
-    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_name.extend(Extension)
-      end
-    end
-    add_oniguruma_properties(result)
-    result
-  end
-
-  def add_oniguruma_properties(props_by_category)
-    props_by_category['Special'] << 'Newline'.extend(Extension)
-  end
-
-  def alias_hash
-    short_names, long_names = short_and_long_names
-    return {} if short_names.empty?
-
-    long_names -= by_category['POSIX brackets']
-    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 }
-    end
-  end
-
-  def short_and_long_names
-    short_name_categories = ['Major and General Categories',
-                             'PropertyAliases',
-                             'PropertyValueAliases (Script)']
-    by_category.each_with_object([[], []]) do |(cat_name, props), (short, long)|
-      (short_name_categories.include?(cat_name) ? short : long).concat(props)
-    end
-  end
-
-  def by_matched_codepoints
-    puts 'Establishing property codepoints, this may take a bit ...'
-    all_for_current_ruby.group_by(&:matched_codepoints)
-  end
-
-  def [](prop)
-    prop.extend(Extension)
+  def self.update
+    Updater.call
   end
 end