charwidth 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 0b348719803e3e762056267c4f562cc2b88cd45d
-  data.tar.gz: 7f6367623657410b4deeb6b3927bfd983ae2d5c6
+  metadata.gz: 4d80ef9c4f6abc327fc91b3551c412da1be6f0f2
+  data.tar.gz: e8f1793a2de0a8edb0f71a4c50f1f3732801ebb7
 SHA512:
-  metadata.gz: 37bbf39dc106e747f4c0a8339e8c1bf58d9de55bea754680e719cbe149570783197cecdfec7a4a5fe5da03d0adf4b3efa6cda28915462d3263eb795bf253d99a
-  data.tar.gz: 73fe1f134bb17213ce8dd8859c4d96a4801593fa52445876eaaac0d407cbe178d8f159e834e21bbb05d362e3da6eaa746394a7d62afa10829008434d863b6e6a
+  metadata.gz: 5408b2545d686f3fc9fab6bb327ceae284bcdc9d05389dde12ac9a593c1b7f1a7d5de77666b6db2e5cc3626eab8e7b85f9c2d362ff9ab5c919dc8ad41858f74d
+  data.tar.gz: 31053172fbda051a8516b7f5e6e41fbfb98852340c9bd0d9f97f38213dd059a33e643133bf0a2e32977ddf7704537f329a72f1fb4a96e2d9c3888ab766405df3

data/README.md

@@ -6,7 +6,7 @@ Normalize Unicode fullwidth / halfwidth (zenkaku / hankaku) characters.
 
 Add this line to your application's Gemfile:
 
-    gem 'charwidth_template'
+    gem 'charwidth'
 
 
 And then execute:
@@ -15,7 +15,7 @@ And then execute:
 
 Or install it yourself as:
 
-    $ gem install charwidth_template
+    $ gem install charwidth
 
 # Usage
 

data/charwidth.gemspec

@@ -21,6 +21,7 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "bundler", "~> 1.3"
   spec.add_development_dependency "rake"
   spec.add_development_dependency "rspec"
+  spec.add_development_dependency "rspec-its"
   spec.add_development_dependency "simplecov"
   spec.add_development_dependency "activerecord", ">= 3"
   spec.add_development_dependency "sqlite3"

data/lib/charwidth.rb

@@ -6,6 +6,41 @@ module Charwidth
   autoload :CLI, "charwidth/cli"
 
   module ClassMethods
+    HALFWIDTH_TO_FULLWIDTH = {
+      ascii: [
+        Characters::ASCII_PUNCTUATION_AND_SYMBOLS,
+        Characters::FULLWIDTH_ASCII_VARIANTS,
+      ],
+      white_parenthesis: [
+        Characters::WHITE_PARENTHESIS,
+        Characters::FULLWIDTH_BRACKETS,
+      ],
+      cjk_punctuation: [
+        Characters::HALFWIDTH_CJK_PUCTUATION,
+        Characters::CJK_SYMBOLS_AND_PUNCTUATION,
+      ],
+      katakana: [
+        Characters::HALFWIDTH_KATAKANA_VARIANTS,
+        Characters::KATAKANA,
+      ],
+      hangul: [
+        Characters::HALFWIDTH_HANGUL_VARIANTS,
+        Characters::HANGUL,
+      ],
+      latin_1_punctuation_and_symbols: [
+        Characters::LATIN_1_PUNCTUATION_AND_SYMBOLS,
+        Characters::FULLWIDTH_SYMBOL_VARIANTS,
+      ],
+      mathematical_symbols: [
+        Characters::HALFWIDTH_SYMBOL_VARIANTS,
+        Characters::MATHEMATICAL_SYMBOLS,
+      ],
+      space: [
+        Characters::SPACE,
+        Characters::IDEOGRAPHIC_SPACE,
+      ],
+    }
+
     # Normalize Unicode fullwidth / halfwidth (zenkaku / hankaku) characters
     # options: {
     #   only: [:ascii, :white_parenthesis, :cjk_punctuation, :katakana, :space],
@@ -21,6 +56,24 @@ module Charwidth
       normalize_charwidth!(string, options)
     end
 
+    def to_full_width(string)
+      to_full_width!(string.dup)
+    end
+
+    def to_full_width!(src)
+      unify_voiced_katakana!(src)
+
+      before, after = "", ""
+      HALFWIDTH_TO_FULLWIDTH.each_value do |half, full|
+        before << half
+        after << full
+      end
+
+      escape_for_tr!(before)
+      escape_for_tr!(after)
+      src.tr!(before, after) || src
+    end
+
     private
     TYPES = [
       :ascii, :white_parenthesis, :cjk_punctuation, :katakana, :hangul,
@@ -51,34 +104,23 @@ module Charwidth
       before, after = "", ""
       types.each do |type|
         case type
-        when :ascii
-          before << Characters::FULLWIDTH_ASCII_VARIANTS
-          after << Characters::ASCII_PUNCTUATION_AND_SYMBOLS
-        when :white_parenthesis
-          before << Characters::FULLWIDTH_BRACKETS
-          after << Characters::WHITE_PARENTHESIS
-        when :cjk_punctuation
-          before << Characters::HALFWIDTH_CJK_PUCTUATION
-          after << Characters::CJK_SYMBOLS_AND_PUNCTUATION
-        when :katakana
-          before << Characters::HALFWIDTH_KATAKANA_VARIANTS
-          after << Characters::KATAKANA
-        when :hangul
-          before << Characters::HALFWIDTH_HANGUL_VARIANTS
-          after << Characters::HANGUL
-        when :latin_1_punctuation_and_symbols
-          before << Characters::FULLWIDTH_SYMBOL_VARIANTS
-          after << Characters::LATIN_1_PUNCTUATION_AND_SYMBOLS
-        when :mathematical_symbols
-          before << Characters::HALFWIDTH_SYMBOL_VARIANTS
-          after << Characters::MATHEMATICAL_SYMBOLS
-        when :space
-          before << Characters::IDEOGRAPHIC_SPACE
-          after << Characters::SPACE
+        when :ascii, :white_parenthesis, :latin_1_punctuation_and_symbols, :space
+          # convert fullwidth to halfwidth
+          HALFWIDTH_TO_FULLWIDTH[type].tap{|half, full|
+            before << full
+            after << half
+          }
+        when :cjk_punctuation, :katakana, :hangul, :mathematical_symbols
+          # convert halfwidth to fullwidth
+          HALFWIDTH_TO_FULLWIDTH[type].tap{|half, full|
+            before << half
+            after << full
+          }
         end
       end
 
-      after.sub!('\\', '\\\\\\\\') # escape for tr
+      escape_for_tr!(before)
+      escape_for_tr!(after)
       src.tr!(before, after) || src
     end
 
@@ -94,6 +136,15 @@ module Charwidth
         str.gsub!(h, f) || str
       end
     end
+
+    def escape_for_tr!(s)
+      s.gsub!('\\', '\\\\')
+      s.gsub!('-', '\\-')
+      s.gsub!('^', '\\^')
+      s.gsub!('[', '\\[')
+      s.gsub!(']', '\\]')
+      s
+    end
   end
 
   extend ClassMethods

data/lib/charwidth/string.rb

@@ -8,6 +8,14 @@ module Charwidth
     def normalize_charwidth!(options = {})
       Charwidth.normalize!(self, options)
     end
+
+    def to_full_width
+      dup.to_full_width!
+    end
+
+    def to_full_width!
+      Charwidth.to_full_width!(self)
+    end
   end
 end
 

data/lib/charwidth/version.rb

@@ -1,3 +1,3 @@
 module Charwidth
-  VERSION = "0.1.3"
+  VERSION = "0.1.4"
 end

data/spec/charwidth_spec.rb

@@ -2,43 +2,87 @@
 require File.expand_path(File.dirname(__FILE__) + '/spec_helper')
 
 describe "Charwidth" do
-  it "should convert full-width alphabet to half-width" do
-    Charwidth.normalize("ＡＢＣａｂｃ").should == "ABCabc"
-  end
-  it "should convert full-width number to half-width" do
-    Charwidth.normalize("１２３").should == "123"
-  end
-  it "should convert full-width ASCII symbol before numbers to half-width" do
-    Charwidth.normalize("！＂＃").should == "!\"#"
-  end
-  it "should convert full-width ASCII symbol between numbers and upper-case to half-width" do
-    Charwidth.normalize("：；").should == ":;"
-  end
-  it "should convert full-width ASCII symbol between upper-case and lower-case to half-width" do
-    Charwidth.normalize("［＼］").should == "[\\]"
-  end
-  it "should convert full-width ASCII symbol after lower-case to half-width" do
-    Charwidth.normalize("｛｜｝").should == "{|}"
-  end
+  describe ".normalize" do
+    it "should convert full-width alphabet to half-width" do
+      expect(Charwidth.normalize("ＡＢＣａｂｃ")).to eq("ABCabc")
+    end
+    it "should convert full-width number to half-width" do
+      expect(Charwidth.normalize("１２３")).to eq("123")
+    end
+    it "should convert full-width ASCII symbol before numbers to half-width" do
+      expect(Charwidth.normalize("！＂＃")).to eq("!\"#")
+    end
+    it "should convert full-width ASCII symbol between numbers and upper-case to half-width" do
+      expect(Charwidth.normalize("：；")).to eq(":;")
+    end
+    it "should convert full-width ASCII symbol between upper-case and lower-case to half-width" do
+      expect(Charwidth.normalize("［＼］")).to eq("[\\]")
+    end
+    it "should convert full-width ASCII symbol after lower-case to half-width" do
+      expect(Charwidth.normalize("｛｜｝")).to eq("{|}")
+    end
 
-  it "should convert half-width CJK punctuation to full-width" do
-    Charwidth.normalize("､｡｢｣").should == "、。「」"
-  end
+    it "should convert half-width CJK punctuation to full-width" do
+      expect(Charwidth.normalize("､｡｢｣")).to eq("、。「」")
+    end
 
-  it "should convert half-width katakana to full-width" do
-    Charwidth.normalize("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･").should == "アカサタナハマヤラワヲンァャッー・"
-    Charwidth.normalize("ﾀﾁﾂﾃﾄﾅﾆﾇﾈﾉ").should == "タチツテトナニヌネノ"
-  end
+    it "should convert half-width katakana to full-width" do
+      expect(Charwidth.normalize("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･")).to eq("アカサタナハマヤラワヲンァャッー・")
+      expect(Charwidth.normalize("ﾀﾁﾂﾃﾄﾅﾆﾇﾈﾉ")).to eq("タチツテトナニヌネノ")
+    end
 
-  it "should unify half-width (semi) voiced katakana with dakuon to full-width" do
-    Charwidth.normalize("ｶﾞｻﾞﾀﾞﾊﾞﾊﾟ").should == "ガザダバパ"
-  end
+    it "should unify half-width (semi) voiced katakana with dakuon to full-width" do
+      expect(Charwidth.normalize("ｶﾞｻﾞﾀﾞﾊﾞﾊﾟ")).to eq("ガザダバパ")
+    end
+
+    it "should convert half-width hangul to full-width" do
+      expect(Charwidth.normalize("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･")).to eq("アカサタナハマヤラワヲンァャッー・")
+    end
 
-  it "should convert half-width hangul to full-width" do
-    Charwidth.normalize("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･").should == "アカサタナハマヤラワヲンァャッー・"
+    it "should convert IDIOGRAPHIC-SPACE to SPACE" do
+      expect(Charwidth.normalize("\u3000")).to eq(" ")
+    end
   end
 
-  it "should convert IDIOGRAPHIC-SPACE to SPACE" do
-    Charwidth.normalize("\u3000").should == " "
+  describe ".to_full_width" do
+    it "should convert half-width alphabet to full-width" do
+      expect(Charwidth.to_full_width("ABCabc")).to eq("ＡＢＣａｂｃ")
+    end
+    it "should convert half-width number to full-width" do
+      expect(Charwidth.to_full_width("123")).to eq("１２３")
+    end
+    it "should convert half-width ASCII symbol before numbers to full-width" do
+      expect(Charwidth.to_full_width("!\"#")).to eq("！＂＃")
+    end
+    it "should convert half-width ASCII symbol between numbers and upper-case to full-width" do
+      expect(Charwidth.to_full_width(":;")).to eq("：；")
+    end
+    it "should convert half-width ASCII symbol between upper-case and lower-case to full-width" do
+      expect(Charwidth.to_full_width("[\\]")).to eq("［＼］")
+    end
+    it "should convert half-width ASCII symbol after lower-case to full-width" do
+      expect(Charwidth.to_full_width("{|}")).to eq("｛｜｝")
+    end
+
+    it "should convert half-width CJK punctuation to full-width" do
+      expect(Charwidth.to_full_width("､｡｢｣")).to eq("、。「」")
+    end
+
+    it "should convert half-width katakana to full-width" do
+      expect(Charwidth.to_full_width("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･")).to eq("アカサタナハマヤラワヲンァャッー・")
+      expect(Charwidth.to_full_width("ﾀﾁﾂﾃﾄﾅﾆﾇﾈﾉ")).to eq("タチツテトナニヌネノ")
+    end
+
+    it "should unify half-width (semi) voiced katakana with dakuon to full-width" do
+      expect(Charwidth.to_full_width("ｶﾞｻﾞﾀﾞﾊﾞﾊﾟ")).to eq("ガザダバパ")
+    end
+
+    it "should convert half-width hangul to full-width" do
+      expect(Charwidth.to_full_width("ｱｶｻﾀﾅﾊﾏﾔﾗﾜｦﾝｧｬｯｰ･")).to eq("アカサタナハマヤラワヲンァャッー・")
+    end
+
+    it "should convert SPACE to IDIOGRAPHIC-SPACE" do
+      expect(Charwidth.to_full_width(" ")).to eq("\u3000")
+    end
   end
 end

data/spec/spec_helper.rb

@@ -1,6 +1,7 @@
 $LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
 $LOAD_PATH.unshift(File.dirname(__FILE__))
 require 'rspec'
+require 'rspec/its'
 require 'charwidth'
 
 # Requires supporting files with custom matchers and macros, etc,

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: charwidth
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
 platform: ruby
 authors:
 - labocho
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-04-04 00:00:00.000000000 Z
+date: 2017-05-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -52,6 +52,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: rspec-its
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: simplecov
   requirement: !ruby/object:Gem::Requirement
@@ -143,7 +157,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.2.0
+rubygems_version: 2.6.10
 signing_key: 
 specification_version: 4
 summary: Normalize Japanese / Korean fullwidth (zenkaku) and halfwidth (hankaku) characters