taxamatch_rb 0.9.2 → 0.9.4

data/CHANGELOG

@@ -1,3 +1,8 @@
+0.9.4 - updated parser (to 1.0.16), updated code to ruby 1.9.3
+
+0.9.3 - Taxamatch::Normalizer substitutes multiplication sign to 'x' 
+(lowcase) instead of '?'
+
 0.9.2 - Taxamatch::Normalizer.normalize always returns only ASCII 
 characters, all utf-8 characters unknown to normalizer are becoming '?'
 

data/Gemfile

@@ -2,17 +2,18 @@ source "http://rubygems.org"
 require 'yaml'
 # YAML::ENGINE.yamler= 'syck'
 
-gem "biodiversity19",">= 1.0.14"
+gem "biodiversity19",">= 1.0.16"
 gem "damerau-levenshtein", ">= 0.5.4"
 
 
 group :development do
+  gem "rake"
   gem "rake-compiler"
-  gem "rspec", "~> 2.3.0"
+  gem "rspec"
   gem "cucumber", ">= 0"
   gem "bundler", "~> 1.1.3"
   gem "jeweler", "~> 1.6.0"
-  gem "ruby-debug19"
+  gem "debugger"
   gem "ruby-prof"
   gem "shoulda"
   gem "mocha"

data/Gemfile.lock

@@ -4,8 +4,7 @@ GEM
     activesupport (3.2.6)
       i18n (~> 0.6)
       multi_json (~> 1.0)
-    archive-tar-minitar (0.5.2)
-    biodiversity19 (1.0.14)
+    biodiversity19 (1.0.16)
       parallel
       parallel
       treetop
@@ -18,6 +17,13 @@ GEM
       gherkin (~> 2.11.0)
       json (>= 1.4.6)
     damerau-levenshtein (0.5.4)
+    debugger (1.2.0)
+      columnize (>= 0.3.1)
+      debugger-linecache (~> 1.1.1)
+      debugger-ruby_core_source (~> 1.1.3)
+    debugger-linecache (1.1.2)
+      debugger-ruby_core_source (>= 1.1.1)
+    debugger-ruby_core_source (1.1.3)
     diff-lcs (1.1.3)
     gherkin (2.11.1)
       json (>= 1.4.6)
@@ -28,36 +34,24 @@ GEM
       git (>= 1.2.5)
       rake
     json (1.7.3)
-    linecache19 (0.5.12)
-      ruby_core_source (>= 0.1.4)
     metaclass (0.0.1)
     mocha (0.12.0)
       metaclass (~> 0.0.1)
     multi_json (1.3.6)
-    parallel (0.5.17)
+    parallel (0.5.18)
     polyglot (0.3.3)
     rake (0.9.2.2)
     rake-compiler (0.8.1)
       rake
-    rspec (2.3.0)
-      rspec-core (~> 2.3.0)
-      rspec-expectations (~> 2.3.0)
-      rspec-mocks (~> 2.3.0)
-    rspec-core (2.3.1)
-    rspec-expectations (2.3.0)
-      diff-lcs (~> 1.1.2)
-    rspec-mocks (2.3.0)
-    ruby-debug-base19 (0.11.25)
-      columnize (>= 0.3.1)
-      linecache19 (>= 0.5.11)
-      ruby_core_source (>= 0.1.4)
-    ruby-debug19 (0.11.6)
-      columnize (>= 0.3.1)
-      linecache19 (>= 0.5.11)
-      ruby-debug-base19 (>= 0.11.19)
+    rspec (2.11.0)
+      rspec-core (~> 2.11.0)
+      rspec-expectations (~> 2.11.0)
+      rspec-mocks (~> 2.11.0)
+    rspec-core (2.11.1)
+    rspec-expectations (2.11.2)
+      diff-lcs (~> 1.1.3)
+    rspec-mocks (2.11.2)
     ruby-prof (0.11.2)
-    ruby_core_source (0.1.5)
-      archive-tar-minitar (>= 0.5.2)
     shoulda (3.1.1)
       shoulda-context (~> 1.0)
       shoulda-matchers (~> 1.2)
@@ -72,14 +66,15 @@ PLATFORMS
   ruby
 
 DEPENDENCIES
-  biodiversity19 (>= 1.0.14)
+  biodiversity19 (>= 1.0.16)
   bundler (~> 1.1.3)
   cucumber
   damerau-levenshtein (>= 0.5.4)
+  debugger
   jeweler (~> 1.6.0)
   mocha
+  rake
   rake-compiler
-  rspec (~> 2.3.0)
-  ruby-debug19
+  rspec
   ruby-prof
   shoulda

data/VERSION

@@ -1 +1 @@
-0.9.2
+0.9.4

data/lib/taxamatch_rb/normalizer.rb

@@ -4,7 +4,7 @@ module Taxamatch
   
   module Normalizer
     def self.normalize(string)
-      utf8_to_ascii(string.strip).upcase.gsub(/[^\x00-\x7F]/,'?')
+      utf8_to_ascii(string.strip.upcase).gsub(/[^\x00-\x7F]/,'?')
     end
   
     def self.normalize_word(word)
@@ -25,37 +25,24 @@ module Taxamatch
   private
     def self.utf8_to_ascii(string)
       string = string.gsub(/\s{2,}/, ' ')
-      string = string.gsub(/[ÀÂÅÃÄÁẤẠÁ]/, "A")
-      string = string.gsub(/[ÉÈÊË]/, "E")
-      string = string.gsub(/[ÍÌÎÏ]/, "I")
-      string = string.gsub(/[ÓÒÔØÕÖỚỔ]/, "O")
-      string = string.gsub(/[ÚÙÛÜ]/, "U")
-      string = string.gsub(/[Ý]/, "Y")
-      string = string.gsub(/Æ/, "AE")
-      string = string.gsub(/[ČÇ]/, "C")
-      string = string.gsub(/[ŠŞ]/, "S")
-      string = string.gsub(/[Đ]/, "D")
-      string = string.gsub(/Ž/, "Z")
-      string = string.gsub(/Ñ/, "N")
-      string = string.gsub(/Œ/, "OE")
+      string = string.gsub("×", "x")
+      string = string.gsub(/[ÀÂÅÃÄÁẤẠÁáàâåãäăãắảạậầằá]/, "A")
+      string = string.gsub(/[ÉÈÊËéèêëĕěếệểễềẻ]/, "E")
+      string = string.gsub(/[ÍÌÎÏíìîïǐĭīĩỉï]/, "I")
+      string = string.gsub(/[ÓÒÔØÕÖỚỔóòôøõöŏỏỗộơọỡốơồờớổő]/, "O")
+      string = string.gsub(/[ÚÙÛÜúùûüůưừựủứụű]/, "U")
+      string = string.gsub(/[Ýýÿỹ]/, "Y")
+      string = string.gsub(/[Ææ]/, "AE")
+      string = string.gsub(/[ČÇčćç]/, "C")
+      string = string.gsub(/[ŠŞśšşſ]/, "S")
+      string = string.gsub(/[Đđð]/, "D")
+      string = string.gsub(/Žžź/, "Z")
+      string = string.gsub(/[Ññńň]/, "N")
+      string = string.gsub(/[Œœ]/, "OE")
       string = string.gsub(/ß/, "B")
       string = string.gsub(/Ķ/, "K")
-      string = string.gsub(/[áàâåãäăãắảạậầằá]/, "a")
-      string = string.gsub(/[éèêëĕěếệểễềẻ]/, "e")
-      string = string.gsub(/[íìîïǐĭīĩỉï]/, "i")
-      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")
-      string = string.gsub(/æ/, "ae")
-      string = string.gsub(/[čćç]/, "c")
-      string = string.gsub(/[ñńň]/, "n")
-      string = string.gsub(/œ/, "oe")
-      string = string.gsub(/[śšş]/, "s")
-      string = string.gsub(/ř/, "r")
-      string = string.gsub(/ğ/, "g")
-      string = string.gsub(/Ř/, "R")
+      string = string.gsub(/ğ/, "G")
+      string = string.gsub(/[Řř]/, "R")
     end
 
   end

data/spec/taxamatch_rb_spec.rb

@@ -35,6 +35,7 @@ describe 'Taxamatch::Normalizer' do
     Taxamatch::Normalizer.normalize('Fallé€n').should == 'FALLE?N'
     Taxamatch::Normalizer.normalize('Fallén привет').should == 'FALLEN ??????'
     Taxamatch::Normalizer.normalize('Choriozopella trägårdhi').should == 'CHORIOZOPELLA TRAGARDHI'
+    Taxamatch::Normalizer.normalize('×Zygomena').should == 'xZYGOMENA'
   end
 
   it 'should normalize words' do

data/taxamatch_rb.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = "taxamatch_rb"
-  s.version = "0.9.2"
+  s.version = "0.9.4"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Dmitry Mozzherin"]
-  s.date = "2012-07-18"
+  s.date = "2012-08-27"
   s.description = "This gem implements algorithm for fuzzy matching scientific names developed by Tony Rees"
   s.email = "dmozzherin@eol.org"
   s.extra_rdoc_files = [
@@ -45,39 +45,42 @@ Gem::Specification.new do |s|
     s.specification_version = 3
 
     if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
-      s.add_runtime_dependency(%q<biodiversity19>, [">= 1.0.14"])
+      s.add_runtime_dependency(%q<biodiversity19>, [">= 1.0.16"])
       s.add_runtime_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
+      s.add_development_dependency(%q<rake>, [">= 0"])
       s.add_development_dependency(%q<rake-compiler>, [">= 0"])
-      s.add_development_dependency(%q<rspec>, ["~> 2.3.0"])
+      s.add_development_dependency(%q<rspec>, [">= 0"])
       s.add_development_dependency(%q<cucumber>, [">= 0"])
       s.add_development_dependency(%q<bundler>, ["~> 1.1.3"])
       s.add_development_dependency(%q<jeweler>, ["~> 1.6.0"])
-      s.add_development_dependency(%q<ruby-debug19>, [">= 0"])
+      s.add_development_dependency(%q<debugger>, [">= 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>, [">= 1.0.14"])
+      s.add_dependency(%q<biodiversity19>, [">= 1.0.16"])
       s.add_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
+      s.add_dependency(%q<rake>, [">= 0"])
       s.add_dependency(%q<rake-compiler>, [">= 0"])
-      s.add_dependency(%q<rspec>, ["~> 2.3.0"])
+      s.add_dependency(%q<rspec>, [">= 0"])
       s.add_dependency(%q<cucumber>, [">= 0"])
       s.add_dependency(%q<bundler>, ["~> 1.1.3"])
       s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
-      s.add_dependency(%q<ruby-debug19>, [">= 0"])
+      s.add_dependency(%q<debugger>, [">= 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>, [">= 1.0.14"])
+    s.add_dependency(%q<biodiversity19>, [">= 1.0.16"])
     s.add_dependency(%q<damerau-levenshtein>, [">= 0.5.4"])
+    s.add_dependency(%q<rake>, [">= 0"])
     s.add_dependency(%q<rake-compiler>, [">= 0"])
-    s.add_dependency(%q<rspec>, ["~> 2.3.0"])
+    s.add_dependency(%q<rspec>, [">= 0"])
     s.add_dependency(%q<cucumber>, [">= 0"])
     s.add_dependency(%q<bundler>, ["~> 1.1.3"])
     s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
-    s.add_dependency(%q<ruby-debug19>, [">= 0"])
+    s.add_dependency(%q<debugger>, [">= 0"])
     s.add_dependency(%q<ruby-prof>, [">= 0"])
     s.add_dependency(%q<shoulda>, [">= 0"])
     s.add_dependency(%q<mocha>, [">= 0"])

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: taxamatch_rb
 version: !ruby/object:Gem::Version
-  version: 0.9.2
+  version: 0.9.4
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-18 00:00:00.000000000 Z
+date: 2012-08-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: biodiversity19
@@ -18,7 +18,7 @@ dependencies:
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
-        version: 1.0.14
+        version: 1.0.16
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -26,7 +26,7 @@ dependencies:
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
-        version: 1.0.14
+        version: 1.0.16
 - !ruby/object:Gem::Dependency
   name: damerau-levenshtein
   requirement: !ruby/object:Gem::Requirement
@@ -43,6 +43,22 @@ dependencies:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: 0.5.4
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rake-compiler
   requirement: !ruby/object:Gem::Requirement
@@ -64,17 +80,17 @@ dependencies:
   requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ~>
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: 2.3.0
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     none: false
     requirements:
-    - - ~>
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: 2.3.0
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: cucumber
   requirement: !ruby/object:Gem::Requirement
@@ -124,7 +140,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: 1.6.0
 - !ruby/object:Gem::Dependency
-  name: ruby-debug19
+  name: debugger
   requirement: !ruby/object:Gem::Requirement
     none: false
     requirements:
@@ -228,7 +244,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 649911613402645259
+      hash: 2280518613036556094
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: