taxamatch_rb 0.9.0 → 0.9.2

data/CHANGELOG

@@ -0,0 +1,4 @@
+0.9.2 - Taxamatch::Normalizer.normalize always returns only ASCII 
+characters, all utf-8 characters unknown to normalizer are becoming '?'
+
+0.9.1 - updated gems

data/Gemfile

@@ -2,7 +2,7 @@ source "http://rubygems.org"
 require 'yaml'
 # YAML::ENGINE.yamler= 'syck'
 
-gem "biodiversity19",">= 1.0.10"
+gem "biodiversity19",">= 1.0.14"
 gem "damerau-levenshtein", ">= 0.5.4"
 
 
@@ -10,9 +10,8 @@ group :development do
   gem "rake-compiler"
   gem "rspec", "~> 2.3.0"
   gem "cucumber", ">= 0"
-  gem "bundler", "~> 1.0.0"
+  gem "bundler", "~> 1.1.3"
   gem "jeweler", "~> 1.6.0"
-  gem "rcov", ">= 0"
   gem "ruby-debug19"
   gem "ruby-prof"
   gem "shoulda"

data/Gemfile.lock

@@ -1,39 +1,44 @@
 GEM
   remote: http://rubygems.org/
   specs:
+    activesupport (3.2.6)
+      i18n (~> 0.6)
+      multi_json (~> 1.0)
     archive-tar-minitar (0.5.2)
-    biodiversity19 (1.0.10)
+    biodiversity19 (1.0.14)
       parallel
       parallel
       treetop
       treetop
     builder (3.0.0)
-    columnize (0.3.3)
-    cucumber (1.0.0)
+    columnize (0.3.6)
+    cucumber (1.2.1)
       builder (>= 2.1.2)
-      diff-lcs (>= 1.1.2)
-      gherkin (~> 2.4.1)
+      diff-lcs (>= 1.1.3)
+      gherkin (~> 2.11.0)
       json (>= 1.4.6)
-      term-ansicolor (>= 1.0.5)
     damerau-levenshtein (0.5.4)
-    diff-lcs (1.1.2)
-    gherkin (2.4.1)
+    diff-lcs (1.1.3)
+    gherkin (2.11.1)
       json (>= 1.4.6)
     git (1.2.5)
-    jeweler (1.6.2)
+    i18n (0.6.0)
+    jeweler (1.6.4)
       bundler (~> 1.0)
       git (>= 1.2.5)
       rake
-    json (1.5.3)
+    json (1.7.3)
     linecache19 (0.5.12)
       ruby_core_source (>= 0.1.4)
-    mocha (0.9.12)
-    parallel (0.5.12)
+    metaclass (0.0.1)
+    mocha (0.12.0)
+      metaclass (~> 0.0.1)
+    multi_json (1.3.6)
+    parallel (0.5.17)
     polyglot (0.3.3)
-    rake (0.9.2)
-    rake-compiler (0.7.9)
+    rake (0.9.2.2)
+    rake-compiler (0.8.1)
       rake
-    rcov (0.9.9)
     rspec (2.3.0)
       rspec-core (~> 2.3.0)
       rspec-expectations (~> 2.3.0)
@@ -50,11 +55,15 @@ GEM
       columnize (>= 0.3.1)
       linecache19 (>= 0.5.11)
       ruby-debug-base19 (>= 0.11.19)
-    ruby-prof (0.10.7)
+    ruby-prof (0.11.2)
     ruby_core_source (0.1.5)
       archive-tar-minitar (>= 0.5.2)
-    shoulda (2.11.3)
-    term-ansicolor (1.0.5)
+    shoulda (3.1.1)
+      shoulda-context (~> 1.0)
+      shoulda-matchers (~> 1.2)
+    shoulda-context (1.0.0)
+    shoulda-matchers (1.2.0)
+      activesupport (>= 3.0.0)
     treetop (1.4.10)
       polyglot
       polyglot (>= 0.3.1)
@@ -63,14 +72,13 @@ PLATFORMS
   ruby
 
 DEPENDENCIES
-  biodiversity19 (>= 1.0.10)
-  bundler (~> 1.0.0)
+  biodiversity19 (>= 1.0.14)
+  bundler (~> 1.1.3)
   cucumber
   damerau-levenshtein (>= 0.5.4)
   jeweler (~> 1.6.0)
   mocha
   rake-compiler
-  rcov
   rspec (~> 2.3.0)
   ruby-debug19
   ruby-prof

data/VERSION

@@ -1 +1 @@
-0.9.0
+0.9.2

data/lib/taxamatch_rb/normalizer.rb

@@ -4,7 +4,7 @@ module Taxamatch
   
   module Normalizer
     def self.normalize(string)
-      utf8_to_ascii(string.strip).upcase
+      utf8_to_ascii(string.strip).upcase.gsub(/[^\x00-\x7F]/,'?')
     end
   
     def self.normalize_word(word)
@@ -25,7 +25,7 @@ module Taxamatch
   private
     def self.utf8_to_ascii(string)
       string = string.gsub(/\s{2,}/, ' ')
-      string = string.gsub(/[ÀÂÅÃÄÁẤẠ]/, "A")
+      string = string.gsub(/[ÀÂÅÃÄÁẤẠÁ]/, "A")
       string = string.gsub(/[ÉÈÊË]/, "E")
       string = string.gsub(/[ÍÌÎÏ]/, "I")
       string = string.gsub(/[ÓÒÔØÕÖỚỔ]/, "O")
@@ -40,11 +40,11 @@ module Taxamatch
       string = string.gsub(/Œ/, "OE")
       string = string.gsub(/ß/, "B")
       string = string.gsub(/Ķ/, "K")
-      string = string.gsub(/[áàâåãäăãắảạậầằ]/, "a")
+      string = string.gsub(/[áàâåãäăãắảạậầằá]/, "a")
       string = string.gsub(/[éèêëĕěếệểễềẻ]/, "e")
       string = string.gsub(/[íìîïǐĭīĩỉï]/, "i")
-      string = string.gsub(/[óòôøõöŏỏỗộơọỡốơồờớổ]/, "o")
-      string = string.gsub(/[úùûüůưừựủứụ]/, "u")
+      string = string.gsub(/[óòôøõöŏỏỗộơọỡốơồờớổő]/, "o")
+      string = string.gsub(/[úùûüůưừựủứụű]/, "u")
       string = string.gsub(/[žź]/, "z")
       string = string.gsub(/[ýÿỹ]/, "y")
       string = string.gsub(/[đð]/, "d")

data/spec/taxamatch_rb_spec.rb

@@ -32,6 +32,8 @@ describe 'Taxamatch::Normalizer' do
     Taxamatch::Normalizer.normalize('Leœptura').should == 'LEOEPTURA'
     Taxamatch::Normalizer.normalize('Ærenea').should == 'AERENEA'
     Taxamatch::Normalizer.normalize('Fallén').should == 'FALLEN'
+    Taxamatch::Normalizer.normalize('Fallé€n').should == 'FALLE?N'
+    Taxamatch::Normalizer.normalize('Fallén привет').should == 'FALLEN ??????'
     Taxamatch::Normalizer.normalize('Choriozopella trägårdhi').should == 'CHORIOZOPELLA TRAGARDHI'
   end
 

data/taxamatch_rb.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = "taxamatch_rb"
-  s.version = "0.9.0"
+  s.version = "0.9.2"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Dmitry Mozzherin"]
-  s.date = "2012-03-08"
+  s.date = "2012-07-18"
   s.description = "This gem implements algorithm for fuzzy matching scientific names developed by Tony Rees"
   s.email = "dmozzherin@eol.org"
   s.extra_rdoc_files = [
@@ -17,6 +17,7 @@ Gem::Specification.new do |s|
     "README.rdoc"
   ]
   s.files = [
+    "CHANGELOG",
     "Gemfile",
     "Gemfile.lock",
     "LICENSE",
@@ -37,48 +38,45 @@ Gem::Specification.new do |s|
   ]
   s.homepage = "http://github.com/GlobalNamesArchitecture/taxamatch_rb"
   s.require_paths = ["lib"]
-  s.rubygems_version = "1.8.10"
+  s.rubygems_version = "1.8.24"
   s.summary = "Implementation of Tony Rees Taxamatch algorithms"
 
   if s.respond_to? :specification_version then
     s.specification_version = 3
 
     if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
-      s.add_runtime_dependency(%q<biodiversity19>, [">= 1.0.10"])
+      s.add_runtime_dependency(%q<biodiversity19>, [">= 1.0.14"])
       s.add_runtime_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
       s.add_development_dependency(%q<rake-compiler>, [">= 0"])
       s.add_development_dependency(%q<rspec>, ["~> 2.3.0"])
       s.add_development_dependency(%q<cucumber>, [">= 0"])
-      s.add_development_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_development_dependency(%q<bundler>, ["~> 1.1.3"])
       s.add_development_dependency(%q<jeweler>, ["~> 1.6.0"])
-      s.add_development_dependency(%q<rcov>, [">= 0"])
       s.add_development_dependency(%q<ruby-debug19>, [">= 0"])
       s.add_development_dependency(%q<ruby-prof>, [">= 0"])
       s.add_development_dependency(%q<shoulda>, [">= 0"])
       s.add_development_dependency(%q<mocha>, [">= 0"])
     else
-      s.add_dependency(%q<biodiversity19>, [">= 1.0.10"])
+      s.add_dependency(%q<biodiversity19>, [">= 1.0.14"])
       s.add_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
       s.add_dependency(%q<rake-compiler>, [">= 0"])
       s.add_dependency(%q<rspec>, ["~> 2.3.0"])
       s.add_dependency(%q<cucumber>, [">= 0"])
-      s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_dependency(%q<bundler>, ["~> 1.1.3"])
       s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
-      s.add_dependency(%q<rcov>, [">= 0"])
       s.add_dependency(%q<ruby-debug19>, [">= 0"])
       s.add_dependency(%q<ruby-prof>, [">= 0"])
       s.add_dependency(%q<shoulda>, [">= 0"])
       s.add_dependency(%q<mocha>, [">= 0"])
     end
   else
-    s.add_dependency(%q<biodiversity19>, [">= 1.0.10"])
+    s.add_dependency(%q<biodiversity19>, [">= 1.0.14"])
     s.add_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
     s.add_dependency(%q<rake-compiler>, [">= 0"])
     s.add_dependency(%q<rspec>, ["~> 2.3.0"])
     s.add_dependency(%q<cucumber>, [">= 0"])
-    s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+    s.add_dependency(%q<bundler>, ["~> 1.1.3"])
     s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
-    s.add_dependency(%q<rcov>, [">= 0"])
     s.add_dependency(%q<ruby-debug19>, [">= 0"])
     s.add_dependency(%q<ruby-prof>, [">= 0"])
     s.add_dependency(%q<shoulda>, [">= 0"])

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: taxamatch_rb
 version: !ruby/object:Gem::Version
-  version: 0.9.0
+  version: 0.9.2
   prerelease: 
 platform: ruby
 authors:
@@ -9,22 +9,27 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-03-08 00:00:00.000000000Z
+date: 2012-07-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: biodiversity19
-  requirement: &70230050526820 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
-        version: 1.0.10
+        version: 1.0.14
   type: :runtime
   prerelease: false
-  version_requirements: *70230050526820
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 1.0.14
 - !ruby/object:Gem::Dependency
   name: damerau-levenshtein
-  requirement: &70230050526220 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -32,10 +37,15 @@ dependencies:
         version: 0.5.4
   type: :runtime
   prerelease: false
-  version_requirements: *70230050526220
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 0.5.4
 - !ruby/object:Gem::Dependency
   name: rake-compiler
-  requirement: &70230050513120 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -43,10 +53,15 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050513120
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &70230050512520 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -54,10 +69,15 @@ dependencies:
         version: 2.3.0
   type: :development
   prerelease: false
-  version_requirements: *70230050512520
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 2.3.0
 - !ruby/object:Gem::Dependency
   name: cucumber
-  requirement: &70230050511860 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -65,21 +85,31 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050511860
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: bundler
-  requirement: &70230050511260 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: 1.0.0
+        version: 1.1.3
   type: :development
   prerelease: false
-  version_requirements: *70230050511260
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.1.3
 - !ruby/object:Gem::Dependency
   name: jeweler
-  requirement: &70230050510740 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -87,10 +117,15 @@ dependencies:
         version: 1.6.0
   type: :development
   prerelease: false
-  version_requirements: *70230050510740
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.6.0
 - !ruby/object:Gem::Dependency
-  name: rcov
-  requirement: &70230050510140 !ruby/object:Gem::Requirement
+  name: ruby-debug19
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -98,21 +133,15 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050510140
-- !ruby/object:Gem::Dependency
-  name: ruby-debug19
-  requirement: &70230050509540 !ruby/object:Gem::Requirement
+  version_requirements: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: *70230050509540
 - !ruby/object:Gem::Dependency
   name: ruby-prof
-  requirement: &70230050508960 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -120,10 +149,15 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050508960
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: shoulda
-  requirement: &70230050508360 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -131,10 +165,15 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050508360
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: mocha
-  requirement: &70230050507760 !ruby/object:Gem::Requirement
+  requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -142,7 +181,12 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70230050507760
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 description: This gem implements algorithm for fuzzy matching scientific names developed
   by Tony Rees
 email: dmozzherin@eol.org
@@ -152,6 +196,7 @@ extra_rdoc_files:
 - LICENSE
 - README.rdoc
 files:
+- CHANGELOG
 - Gemfile
 - Gemfile.lock
 - LICENSE
@@ -183,7 +228,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 1938855165172895621
+      hash: 649911613402645259
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -192,7 +237,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 1.8.10
+rubygems_version: 1.8.24
 signing_key: 
 specification_version: 3
 summary: Implementation of Tony Rees Taxamatch algorithms