unicode-display_width 1.0.3 → 1.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: df3414ad82ceddc67108a78ec815e72fd9ee15e8
-  data.tar.gz: 03ecf18632a9e2aa160f65e4a488832849a50059
+  metadata.gz: af67bc03745452e81cf9a29a48ed797fd82cef01
+  data.tar.gz: 7f85e858df6f46406060198108be6b8b14a7b6eb
 SHA512:
-  metadata.gz: 97769f68b776f09abbca2d172e823024d4d8b7c409c023b7906aebdad0a8f86d105411f5986eb788d98d9137b6b651e9113eace075e25bde24b1d42918ab91c5
-  data.tar.gz: 27a6242bdd2fb86677238d33a78de8df42a8113d4867a9f9107c3a6e6d686090138bca0949a53a57b454aac9e1abdc59ba4ed42b8cfd5bf596d3d4cecac39c03
+  metadata.gz: b3fae9148367d5b39cfb04cc52cb800c53b5ed4e8bfcb5b93a8e1831d86af6a5179a00c144b993345b84020b3e4c1c56816f8d8859d283657bfc9f1521bb748e
+  data.tar.gz: b228d550c5d55822a1b4bf70dc2130a09bb18482dbcb96ff30c7fb4bb694ee9a1933e6b9931e7a962028d8296676b16a343c4286e355e8029f5006175c594abf

data/CHANGELOG.txt

@@ -1,5 +1,11 @@
 # CHANGELOG
 
+## 1.0.4
+
+* New index format (much smaller) and internal API changes
+* Move index generation to a builder plugin for the unicoder gem
+* No public API changes
+
 ## 1.0.3
 
 * Avoid circular dependency warning

data/lib/unicode/display_width.rb

@@ -6,18 +6,16 @@ module Unicode
       require_relative 'display_width/index' unless defined? ::Unicode::DisplayWidth::INDEX
 
       res = string.unpack('U*').inject(0){ |total_width, codepoint|
-        total_width + (
-          overwrite[codepoint] || case width = INDEX[codepoint]
-                                  when Integer
-                                    width
-                                  when :F, :W
-                                    2
-                                  when :A
-                                    ambiguous
-                                  else # including :N, :Na, :H
-                                    1
-                                  end
-        )
+        index_or_value = INDEX
+        codepoint_depth_offset = codepoint
+        [0x10000, 0x1000, 0x100, 0x10].each{ |depth|
+          index_or_value         = index_or_value[codepoint_depth_offset / depth]
+          codepoint_depth_offset = codepoint_depth_offset % depth
+          break unless index_or_value.is_a? Array
+        }
+        width = index_or_value.is_a?(Array) ? index_or_value[codepoint_depth_offset] : index_or_value
+        width = ambiguous if width == :A
+        total_width + (overwrite[codepoint] || width || 1)
       }
 
       res < 0 ? 0 : res

data/lib/unicode/display_width/constants.rb

@@ -1,7 +1,8 @@
 module Unicode
   module DisplayWidth
-    VERSION = '1.0.3'.freeze
-    DATA_DIRECTORY = File.join(File.dirname(__FILE__), '../../../data/').freeze
-    INDEX_FILENAME = (DATA_DIRECTORY + 'unicode-width.index').freeze
+    VERSION = '1.0.4'
+    UNICODE_VERSION = "8.0.0".freeze
+    DATA_DIRECTORY = File.expand_path(File.dirname(__FILE__) + '/../../../data/').freeze
+    INDEX_FILENAME = (DATA_DIRECTORY + '/display_width.marshal.gz').freeze
   end
 end

data/lib/unicode/display_width/index.rb

@@ -2,6 +2,6 @@ require_relative 'constants'
 
 module Unicode
   module DisplayWidth
-    INDEX = Marshal.load(File.binread(INDEX_FILENAME))
+    INDEX = Marshal.load(Gem.gunzip(File.binread(INDEX_FILENAME)))
   end
 end

data/unicode-display_width.gemspec

@@ -12,7 +12,7 @@ Gem::Specification.new do |s|
   s.files = Dir.glob(%w[{lib,spec}/**/*.rb [A-Z]*.{txt,rdoc} data/unicode-width.index]) + %w{Rakefile unicode-display_width.gemspec}
   s.extra_rdoc_files = ["README.md", "MIT-LICENSE.txt", "CHANGELOG.txt"]
   s.license = 'MIT'
-  s.required_ruby_version = ['>= 1.9.3', '< 3.0.0']
+  s.required_ruby_version = '>= 1.9.3'
   s.add_development_dependency 'rspec', '~> 3.4'
   s.add_development_dependency 'rake', '~> 10.4'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: unicode-display_width
 version: !ruby/object:Gem::Version
-  version: 1.0.3
+  version: 1.0.4
 platform: ruby
 authors:
 - Jan Lelis
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-03-29 00:00:00.000000000 Z
+date: 2016-04-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
@@ -52,11 +52,9 @@ files:
 - MIT-LICENSE.txt
 - README.md
 - Rakefile
-- data/unicode-width.index
 - lib/unicode/display_width.rb
 - lib/unicode/display_width/constants.rb
 - lib/unicode/display_width/index.rb
-- lib/unicode/display_width/index_builder.rb
 - lib/unicode/display_width/no_string_ext.rb
 - lib/unicode/display_width/string_ext.rb
 - spec/display_width_spec.rb
@@ -74,9 +72,6 @@ required_ruby_version: !ruby/object:Gem::Requirement
   - - ">="
     - !ruby/object:Gem::Version
       version: 1.9.3
-  - - "<"
-    - !ruby/object:Gem::Version
-      version: 3.0.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
@@ -84,7 +79,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.5.1
+rubygems_version: 2.6.3
 signing_key: 
 specification_version: 4
 summary: Determines the monospace display width of a string in Ruby.

data/lib/unicode/display_width/index_builder.rb

@@ -1,68 +0,0 @@
-require_relative 'constants'
-
-module Unicode
-  module DisplayWidth
-    module IndexBuilder
-      EAST_ASIAN_WIDTH_DATA_URL      = "http://www.unicode.org/Public/UNIDATA/EastAsianWidth.txt".freeze
-      EAST_ASIAN_WIDTH_DATA_FILENAME = (DATA_DIRECTORY + 'EastAsianWidth.txt').freeze
-      IGNORE_CATEGORIES     = %w[Cs Co Cn].freeze
-      ZERO_WIDTH_CATEGORIES = %w[Mn Me Cf].freeze
-      ZERO_WIDTH_CODEPOINTS = [*0x1160..0x11FF].freeze
-      SPECIAL_WIDTHS = {
-        0x0    =>  0, # \0 NULL
-        0x5    =>  0, #    ENQUIRY
-        0x7    =>  0, # \a BELL
-        0x8    => -1, # \b BACKSPACE
-        0xA    =>  0, # \n LINE FEED
-        0xB    =>  0, # \v LINE TABULATION
-        0xC    =>  0, # \f FORM FEED
-        0xD    =>  0, # \r CARRIAGE RETURN
-        0xE    =>  0, #    SHIFT OUT
-        0xF    =>  0, #    SHIFT IN
-        0x00AD =>  1, #    SOFT HYPHEN
-        0x2E3A =>  2, #    TWO-EM DASH
-        0x2E3B =>  3, #    THREE-EM DASH
-      }.freeze
-
-      def self.fetch!
-        require 'open-uri'
-        open(EAST_ASIAN_WIDTH_DATA_URL) { |f|
-          File.write(EAST_ASIAN_WIDTH_DATA_FILENAME, f.read)
-        }
-      end
-
-      def self.build!
-        data = File.open(EAST_ASIAN_WIDTH_DATA_FILENAME)
-        data.rewind
-        Dir.mkdir(DATA_DIRECTORY) unless Dir.exists?(DATA_DIRECTORY)
-        index = {}
-
-        data.each_line{ |line|
-          line =~ /^(\S+?);(\S+)\s+#\s(\S+).*$/
-          if $1 && $2
-            cps, width, category = $1, $2, $3
-            next if IGNORE_CATEGORIES.include?(category)
-            if cps['..']
-              codepoints = Range.new(*cps.split('..').map{ |cp| cp.to_i(16) })
-            else
-              codepoints = [cps.to_i(16)]
-            end
-
-            codepoints.each{ |cp|
-              index[cp] = is_zero_width?(category, cp) ? 0 : width.to_sym
-            }
-          end
-        }
-
-        index.merge! SPECIAL_WIDTHS
-        File.open(INDEX_FILENAME, 'wb') { |f| Marshal.dump(index, f) }
-      end
-
-      def self.is_zero_width?(category, cp)
-        ( ZERO_WIDTH_CATEGORIES.include?(category) &&
-            [cp].pack('U') !~ /\p{Cf}(?<=\p{Arabic})/ ) ||
-          ZERO_WIDTH_CODEPOINTS.include?(cp)
-      end
-    end
-  end
-end