name-tamer 0.4.0 → 0.5.1

data/lib/{string_extras.rb → name_tamer/string.rb}

@@ -1,5 +1,11 @@
 # encoding: utf-8
 class String
+  unless respond_to? :presence
+    def presence
+      self unless empty?
+    end
+  end
+
   # Strip illegal characters out completely
   def strip_unwanted!(filter)
     substitute!(filter, '')
@@ -70,11 +76,11 @@ class String
   end
 
   def upcase_first_letter!
-    gsub!(/\b\w/) { |first| first.upcase } || self
+    gsub!(/\b\w/, &:upcase) || self
   end
 
   def downcase_after_apostrophe!
-    gsub!(/\'\w\b/) { |c| c.downcase } || self # Lowercase 's
+    gsub!(/\'\w\b/, &:downcase) || self # Lowercase 's
   end
 
   # Our list of terminal characters that indicate a non-celtic name used
@@ -170,19 +176,19 @@ class String
     gsub!(pattern, replacement) || self
   end
 
-  NONBREAKING_SPACE = "\u00a0"
-  ASCII_SPACE = "\u0020"
+  NONBREAKING_SPACE = "\u00a0".freeze
+  ASCII_SPACE = ' '.freeze
 
   COMPOUND_NAMES = [
     'Lane Fox', 'Bonham Carter', 'Pitt Rivers', 'Lloyd Webber', 'Sebag Montefiore', 'Holmes à Court', 'Holmes a Court',
     'Baron Cohen', 'Strang Steel',
     'Service Company', 'Corporation Company', 'Corporation System', 'Incorporations Limited'
-  ]
+  ].freeze
 
   NAME_MODIFIERS = [
     'Al', 'Ap', 'Ben', 'Dell[ae]', 'D[aeiou]', 'De[lrn]', 'D[ao]s', 'El', 'La', 'L[eo]', 'V[ao]n', 'Of', 'San',
     'St[\.]?', 'Zur'
-  ]
+  ].freeze
 
   # Transliterations (like the i18n defaults)
   # see https://github.com/svenfuchs/i18n/blob/master/lib/i18n/backend/transliterator.rb
@@ -215,7 +221,7 @@ class String
     'ů' => 'u', 'Ű' => 'U', 'ű' => 'u', 'Ų' => 'U', 'ų' => 'u', 'Ŵ' => 'W', 'ŵ' => 'w',
     'Ŷ' => 'Y', 'ŷ' => 'y', 'Ÿ' => 'Y', 'Ź' => 'Z', 'ź' => 'z', 'Ż' => 'Z', 'ż' => 'z',
     'Ž' => 'Z', 'ž' => 'z'
-  }
+  }.freeze
 
   # When strings are mistakenly encoded as single-byte character sets, instead
   # of UTF-8, there are some distinctive character combinations that we can spot
@@ -251,7 +257,7 @@ class String
     'ø' => 'ø', 'ù' => 'ù', 'ú' => 'ú', 'û' => 'û', 'ü' => 'ü',
     'ý' => 'ý', 'þ' => 'þ', 'ÿ' => 'ÿ',
     "\x00" => '' # Manually added to avoid Bad Argument exception
-  }
+  }.freeze
 
   BAD_ENCODING_PATTERNS = /(#{BAD_ENCODING.keys.join('|')})/
 

data/lib/name_tamer/text.rb

@@ -0,0 +1,53 @@
+module NameTamer
+  class Text
+    # All the potential slugs from the string
+    # e.g. 'lorem ipsum dolor' -> ['lorem', 'ipsum' ,'dolor', 'lorem-ipsum', 'ipsum-dolor', 'lorem-ipsum-dolor']
+    def slugs
+      @slugs ||= segments.flat_map { |s| self.class.new(s).neighbours }.uniq
+    end
+
+    # Split the string into segments (e.g. sentences)
+    def segments
+      string.split(%r{(?:[\.\?,:;!]|[[:space:]][/-])[[:space:]]})
+    end
+
+    # The string as a slug
+    def parameterize
+      @parameterize ||= (
+        string
+        .dup
+        .whitespace_to!(separator)
+        .invalid_chars_to!(separator)
+        .strip_unwanted!(filter)
+        .fix_separators!(separator)
+        .approximate_latin_chars!
+        .presence || '_'
+      ).downcase
+    end
+
+    def neighbours
+      @neighbours ||= NameTamer[string].array.neighbours.map { |a| a.join('-') }
+    end
+
+    private
+
+    attr_reader :string, :args
+
+    def initialize(string, args = {})
+      @string = string
+      @args = args
+    end
+
+    def separator
+      @seperator ||= args[:sep] || SLUG_DELIMITER
+    end
+
+    def rfc3987
+      @rfc3987 ||= args[:rfc3987] || false
+    end
+
+    def filter
+      @filter ||= args[:filter] || (rfc3987 ? FILTER_RFC3987 : FILTER_COMPAT)
+    end
+  end
+end

data/lib/name_tamer/version.rb

@@ -0,0 +1,3 @@
+module NameTamer
+  VERSION = '0.5.1'.freeze
+end

data/name-tamer.gemspec

@@ -1,6 +1,6 @@
 lib = File.expand_path('../lib', __FILE__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
-require 'name-tamer/version'
+require 'name_tamer/version'
 
 Gem::Specification.new do |spec|
   spec.name = 'name-tamer'
@@ -17,13 +17,13 @@ Gem::Specification.new do |spec|
   spec.test_files = spec.files.grep(%r{^(test|spec|features|coverage)\/})
   spec.require_paths = ['lib']
 
-  spec.add_development_dependency 'rake', '~> 10'
-  spec.add_development_dependency 'rspec', '~> 3.3'
-  spec.add_development_dependency 'gem-release', '~> 0.7'
-  spec.add_development_dependency 'simplecov', '~> 0.10'
-  spec.add_development_dependency 'coveralls', '~> 0.8'
-  spec.add_development_dependency 'rubocop', '~> 0.32'
-  spec.add_development_dependency 'guard', '~> 2.12'
-  spec.add_development_dependency 'guard-rspec', '~> 4.5'
-  spec.add_development_dependency 'guard-rubocop', '~> 1.2'
+  spec.add_development_dependency 'rake', '~> 11'
+  spec.add_development_dependency 'rspec', '~> 3'
+  spec.add_development_dependency 'gem-release', '~> 0'
+  spec.add_development_dependency 'simplecov', '~> 0'
+  spec.add_development_dependency 'coveralls', '~> 0'
+  spec.add_development_dependency 'rubocop', '~> 0'
+  spec.add_development_dependency 'guard', '~> 2'
+  spec.add_development_dependency 'guard-rspec', '~> 4'
+  spec.add_development_dependency 'guard-rubocop', '~> 1'
 end

data/spec/{name_tamer_spec.rb → name_tamer/name_spec.rb}

@@ -2,7 +2,7 @@
 require 'spec_helper'
 require 'name-tamer'
 
-describe NameTamer do
+describe NameTamer::Name do
   let(:names) do
     [
       { n: 'John Smith', t: :person, nn: 'John Smith', sn: 'John Smith', s: 'john-smith' },
@@ -207,7 +207,7 @@ describe NameTamer do
       { n: 'Dibble & Grub LLP', t: :organization,
         nn: 'Dibble & Grub',
         sn: 'Dibble and Grub',
-        s: 'dibble-and-grub' },
+        s: 'dibble-and-grub' }
     ]
   end
 

data/spec/name_tamer/text_spec.rb

@@ -0,0 +1,42 @@
+# encoding: utf-8
+require 'spec_helper'
+require 'name-tamer'
+
+describe NameTamer::Text do
+  context '#segments' do
+    it 'splits a string into segments at appropriate boundaries' do
+      string = 'Lorem ipsum dolor sit amet, consectetur adipiscing elit. '\
+               'Nullam venenatis? Risus eu: auctor feugiat; libero nisl congue '\
+               'arcu - eget molestie metus / erat eu diam'
+
+      text = NameTamer::Text.new string
+
+      expect(text.segments).to include(
+        'Lorem ipsum dolor sit amet',
+        'consectetur adipiscing elit',
+        'Nullam venenatis',
+        'Risus eu',
+        'auctor feugiat',
+        'libero nisl congue arcu',
+        'eget molestie metus',
+        'erat eu diam'
+      )
+    end
+  end
+
+  context '#slugs' do
+    it 'compiles all the potential slugs into an array' do
+      string = 'Lorem Ipsum Limited, lorem ipsum dolor. Dolor Mr Sit Amet.'
+      text = NameTamer::Text.new string
+      slugs = text.slugs
+
+      expect(slugs).to include(
+        'lorem', 'lorem-ipsum', 'ipsum', 'lorem-ipsum-dolor', 'ipsum-dolor',
+        'dolor', 'dolor-mr', 'dolor-mr-sit', 'dolor-mr-sit-amet', 'mr',
+        'mr-sit', 'mr-sit-amet', 'sit', 'sit-amet', 'amet'
+      )
+
+      expect(slugs.length).to eq 15
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: name-tamer
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.1
 platform: ruby
 authors:
 - Xenapto
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-03-08 00:00:00.000000000 Z
+date: 2016-05-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -16,126 +16,126 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10'
+        version: '11'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10'
+        version: '11'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.3'
+        version: '3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.3'
+        version: '3'
 - !ruby/object:Gem::Dependency
   name: gem-release
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.7'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.7'
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: simplecov
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.10'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.10'
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: coveralls
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.8'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.8'
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rubocop
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.32'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.32'
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: guard
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.12'
+        version: '2'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.12'
+        version: '2'
 - !ruby/object:Gem::Dependency
   name: guard-rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.5'
+        version: '4'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.5'
+        version: '4'
 - !ruby/object:Gem::Dependency
   name: guard-rubocop
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
+        version: '1'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
+        version: '1'
 description: Useful methods for taming names
 email:
 - developers@xenapto.com
@@ -158,10 +158,16 @@ files:
 - doc/prefixes.csv
 - doc/suffixes.csv
 - lib/name-tamer.rb
-- lib/name-tamer/version.rb
-- lib/string_extras.rb
+- lib/name_tamer.rb
+- lib/name_tamer/array.rb
+- lib/name_tamer/constants.rb
+- lib/name_tamer/name.rb
+- lib/name_tamer/string.rb
+- lib/name_tamer/text.rb
+- lib/name_tamer/version.rb
 - name-tamer.gemspec
-- spec/name_tamer_spec.rb
+- spec/name_tamer/name_spec.rb
+- spec/name_tamer/text_spec.rb
 - spec/spec_helper.rb
 homepage: https://github.com/Xenapto/name-tamer
 licenses:
@@ -189,5 +195,6 @@ specification_version: 4
 summary: 'Example: NameTamer[''Mr. John Q. Smith III, MD''].simple_name # => John
   Smith'
 test_files:
-- spec/name_tamer_spec.rb
+- spec/name_tamer/name_spec.rb
+- spec/name_tamer/text_spec.rb
 - spec/spec_helper.rb

data/lib/name-tamer/version.rb

@@ -1,3 +0,0 @@
-class NameTamer
-  VERSION = '0.4.0'
-end