java-properties 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 3a3e467287d2328e09893d5975725e9d0bfde497
-  data.tar.gz: a775715d414e13dd7afa7bb36cc603013c334e84
+  metadata.gz: d4f07c15c1f58cb3478fab7debf52316d9649cc3
+  data.tar.gz: 7b340371a2ea3b0420fb2e0ee532e926fd8bc006
 SHA512:
-  metadata.gz: 33dba5e873cb1a2bf16e7d74561a043e5770c313346e958cc0c6b7c4947b1e7f70a3079899a7b4e7902236316ee0c96ae62aa1168c817663ad151fe75cf53c58
-  data.tar.gz: da4f1c8dcdc65b8c730777be9777e7ea9d999f68762b6dc7e847d2341563cfd49c2cea0587822b5dabd3a3e4ed8b63e39271e2a51e64942ec06f4529ad2c483d
+  metadata.gz: c9c763b44183f8e3175326d1d699af24386cdc6abe95b6de3df9fa5514ed6ade61470be36a84403e032238403ac8eca5bf1dd036d5fb59e4db5ed6b7b8581491
+  data.tar.gz: 08865b814f74ab2fb98381843a0550b25b107a825ecc985813db0580c2f24d3aefab9f1914b50b787c05819d1a3c242bb595d5bb1e787e5d7d412b75cb487492

data/README.md

@@ -61,6 +61,7 @@ As Java properties files normally hold UTF-8 chars in their escaped representati
 
 ```
 "ה" <=> "\u05d4"
+"𪀯"  <=> "\ud868\udc2f"
 ```
 
 The tool also escaped every '=', ' ' and ':' in the name part of a property line:

data/java-properties.gemspec

@@ -20,4 +20,5 @@ Gem::Specification.new do |spec|
   spec.test_files = Dir.glob("spec/fixtures/**/*.properties")
 
   spec.required_rubygems_version = '>= 1.3.5'
-end
+  spec.required_ruby_version = '~> 2.0'
+end

data/lib/java-properties/encoding/unicode.rb

@@ -1,24 +1,63 @@
 module JavaProperties
   module Encoding
     # Module to encode and decode unicode chars
-    # @see JavaProperties::Encoding
+    # This code is highly influced by Florian Frank's JSON gem
+    # @see https://github.com/flori/json/
     module Unicode
 
-      # Marker for encoded unicode chars
-      # @return [Regexp]
-      UNICODE_MARKER  = /\\[uU]([0-9a-fA-F]{4,5}|10[0-9a-fA-F]{4})/
+      # @private
+      MAP = {
+        "\x0" => '\u0000',
+        "\x1" => '\u0001',
+        "\x2" => '\u0002',
+        "\x3" => '\u0003',
+        "\x4" => '\u0004',
+        "\x5" => '\u0005',
+        "\x6" => '\u0006',
+        "\x7" => '\u0007',
+        "\xb" => '\u000b',
+        "\xe" => '\u000e',
+        "\xf" => '\u000f',
+        "\x10" => '\u0010',
+        "\x11" => '\u0011',
+        "\x12" => '\u0012',
+        "\x13" => '\u0013',
+        "\x14" => '\u0014',
+        "\x15" => '\u0015',
+        "\x16" => '\u0016',
+        "\x17" => '\u0017',
+        "\x18" => '\u0018',
+        "\x19" => '\u0019',
+        "\x1a" => '\u001a',
+        "\x1b" => '\u001b',
+        "\x1c" => '\u001c',
+        "\x1d" => '\u001d',
+        "\x1e" => '\u001e',
+        "\x1f" => '\u001f',
+      }
 
-      # Escape char for unicode chars
-      # @return [String]
-      UNICODE_ESCAPE = "\\u"
+      # @private
+      EMPTY_8BIT_STRING = ''
+      EMPTY_8BIT_STRING.force_encoding(::Encoding::ASCII_8BIT)
 
       # Decodes all unicode chars from escape sequences in place
       # @param text [String]
       # @return [String] The encoded text for chaining
       def self.decode!(text)
-        text.gsub!(UNICODE_MARKER) do
-          unicode($1.hex)
+        string = text.dup
+        string = string.gsub(%r((?:\\[uU](?:[A-Fa-f\d]{4}))+)) do |c|
+          c.downcase!
+          bytes = EMPTY_8BIT_STRING.dup
+          i = 0
+          while c[6 * i] == ?\\ && c[6 * i + 1] == ?u
+            bytes << c[6 * i + 2, 2].to_i(16) << c[6 * i + 4, 2].to_i(16)
+            i += 1
+          end
+          bytes.encode("utf-8", "utf-16be")
         end
+        string.force_encoding(::Encoding::UTF_8)
+
+        text.replace string
         text
       end
 
@@ -26,16 +65,25 @@ module JavaProperties
       # @param text [String]
       # @return [String] The decoded text for chaining
       def self.encode!(text)
-        buffer = StringIO.new
-        text.each_char do |char|
-          if char.ascii_only?
-            buffer << char
-          else
-            buffer << UNICODE_ESCAPE
-            buffer << hex(char.codepoints.first)
-          end
-        end
-        text.replace buffer.string
+        string = text.dup
+        string.force_encoding(::Encoding::ASCII_8BIT)
+        string.gsub!(/["\\\x0-\x1f]/n) { |c| MAP[c] || c }
+        string.gsub!(/(
+          (?:
+           [\xc2-\xdf][\x80-\xbf]    |
+           [\xe0-\xef][\x80-\xbf]{2} |
+           [\xf0-\xf4][\x80-\xbf]{3}
+          )+ |
+          [\x80-\xc1\xf5-\xff]       # invalid
+        )/nx) { |c|
+          c.size == 1 and raise "Invalid utf8 byte: '#{c}'"
+          s = c.encode("utf-16be", "utf-8").unpack('H*')[0]
+          s.force_encoding(::Encoding::ASCII_8BIT)
+          s.gsub!(/.{4}/n, '\\\\u\&')
+          s.force_encoding(::Encoding::UTF_8)
+        }
+        string.force_encoding(::Encoding::UTF_8)
+        text.replace string
         text
       end
 

data/lib/java-properties/version.rb

@@ -2,6 +2,6 @@ module JavaProperties
 
   # Current version
   # @return [String]
-  VERSION = "0.1.1".freeze
+  VERSION = "0.2.0".freeze
 
 end

data/spec/fixtures/test.properties

@@ -1,16 +1,16 @@
 # Comment 1
 ! Comment 2
 item0
-item1 = item1 
-item2 : item2 
+item1 = item1
+item2 : item2
 item3   item3
 
  #Comment 3
  ! Comment 4
- 
+
  it\ em4=item4
- it\=em5:item5 
- item6 item6 
+ it\=em5:item5
+ item6 item6
 
 !Comment 4
 # Comment 5
@@ -31,3 +31,4 @@ item10=test\n\ttest\u0050 \
        test\n\ttest \
        test\n\ttest = test
 
+item11=a\u00e4b \ud868\udc2f

data/spec/fixtures/test_normalized.properties

@@ -8,4 +8,5 @@ item6=item6
 item7=line 1 line 2 line 3
 item8=line 1 line 2 line 3
 item9=line 1 line 2 line 3
-item10=test\n\ttest\u0050 test\n\ttest test\n\ttest = test
+item10=test\n\ttest\u0050 test\n\ttest test\n\ttest = test
+item11=a\u00e4b \ud868\udc2f

data/spec/fixtures/test_out.properties

@@ -8,4 +8,5 @@ it\:em6=item6
 item7=line 1 line 2 line 3
 item8=line 1 line 2 line 3
 item9=line 1 line 2 line 3
-item10=test\n\ttest\u05d4 test\n\ttest test\n\ttest\u00fc = test
+item10=test\n\ttest\u05d4 test\n\ttest test\n\ttest\u00fc = test
+item11=a\u00e4b \ud868\udc2f

data/spec/fixtures/test_out_skip_separators.properties

@@ -8,4 +8,5 @@ it:em6=item6
 item7=line 1 line 2 line 3
 item8=line 1 line 2 line 3
 item9=line 1 line 2 line 3
-item10=test\n\ttest\u05d4 test\n\ttest test\n\ttest\u00fc = test
+item10=test\n\ttest\u05d4 test\n\ttest test\n\ttest\u00fc = test
+item11=a\u00e4b \ud868\udc2f

data/spec/fixtures/test_out_skip_special_chars.properties

@@ -11,4 +11,5 @@ item9=line 1 line 2 line 3
 item10=test
 	test\u05d4 test
 	test test
-	test\u00fc = test
+	test\u00fc = test
+item11=a\u00e4b \ud868\udc2f

data/spec/fixtures/test_out_skip_unicode.properties

@@ -8,4 +8,5 @@ it\:em6=item6
 item7=line 1 line 2 line 3
 item8=line 1 line 2 line 3
 item9=line 1 line 2 line 3
-item10=test\n\ttestה test\n\ttest test\n\ttestü = test
+item10=test\n\ttestה test\n\ttest test\n\ttestü = test
+item11=aäb 𪀯

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: java-properties
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - Jonas Thiel
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-10-21 00:00:00.000000000 Z
+date: 2016-09-28 00:00:00.000000000 Z
 dependencies: []
 description: Tool for loading and writing Java properties files
 email:
@@ -50,17 +50,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - "~>"
     - !ruby/object:Gem::Version
-      version: '0'
+      version: '2.0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: 1.3.5
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.6
+rubygems_version: 2.5.1
 signing_key: 
 specification_version: 4
 summary: Loader and writer for *.properties files