pascoale 0.0.1 → 0.1.0

data/lib/pascoale/constants.rb

@@ -0,0 +1,8 @@
+module Pascoale
+  module Constants
+    VOWELS = 'aeiouáéíóúâêôãõày'
+    SEMIVOWELS = 'iu'
+    CONSONANTS = 'bcdfghjklmnpqrstvwxzç'
+    LETTERS = VOWELS + CONSONANTS
+  end
+end

data/lib/pascoale/edits.rb

@@ -1,5 +1,5 @@
 class Pascoale::Edits
-  LETTERS = ' abcdefghijklmnopqrstuvwxyzáéíóúâêôãõç'.scan(/./)
+  LETTERS = [' '] + Pascoale::Constants::LETTERS.scan(/./)
 
   def initialize(word)
     @splits = (0..(word.size)).map do |i|

data/lib/pascoale/syllable_separator.rb

@@ -0,0 +1,44 @@
+module Pascoale
+  class SyllableSeparator
+    include Constants
+
+    ONSET = "(?:ch|lh|nh|gu|qu|[pbtdcgfv][lr]|[#{CONSONANTS}])"
+
+    # Still in doubt if we should add suffixes to the "i" semivowel...
+    # it slightly improves the the matches, but some of them causes more
+    # noise than fix things =\
+    #NUCLEUS = "(?:ãe|ão|õe|[#{VOWELS}](?:u|i(?!nh|r$|m$|dora?$|ção$|dade$))?)"
+    NUCLEUS = "(?:ãe|ão|õe|[#{VOWELS}](?:u|i(?!nh|r$|m$|ção$|dora?$))?)"
+
+    CODA = "[#{CONSONANTS}]"
+
+    # The concept of "rhyme" does not help in this algorithm. It seems the
+    # concept makes no sense for syllable separation in portuguese
+    KERNEL = "#{ONSET}?#{NUCLEUS}"
+
+    def initialize(word)
+      @word = word
+    end
+
+    def separated
+      rest = @word
+      result = []
+      while rest && rest.size > 0
+        if rest =~ /^(#{KERNEL})(?:(#{KERNEL})|(#{CODA})(#{KERNEL})|(#{CODA}#{CODA})(#{KERNEL})|(#{CODA}#{CODA})|(#{CODA}))?(.*)$/
+          result << $1 + $3.to_s + $5.to_s + $7.to_s + $8.to_s
+          rest = $2.to_s + $4.to_s + $6.to_s + $9.to_s
+          # Special case! Hate them :(
+          # Pneu, Gnomo, Mnemônica, Pseudônimo
+        elsif result.size == 0
+          if rest =~ /^([#{CONSONANTS}]#{KERNEL})(?:(#{KERNEL})|(#{CODA})(#{KERNEL})|(#{CODA}#{CODA})(#{KERNEL})|(#{CODA}#{CODA})|(#{CODA}))?(.*)$/
+            result << $1 + $3.to_s + $5.to_s + $7.to_s + $8.to_s
+            rest = $2.to_s + $4.to_s + $6.to_s + $9.to_s
+          end
+        else
+          raise %(Cannot separate "#{@word}". No rule match next syllable at "#{result.join('')}|>#{rest}")
+        end
+      end
+      result
+    end
+  end
+end

data/lib/pascoale/syllable_separator_benchmark.rb

@@ -0,0 +1,29 @@
+require 'pascoale'
+
+correct_counter = 0
+wrong_counter = 0
+word = nil
+
+begin
+  open("#{Pascoale.root}/data/everything.txt") do |file|
+    file.each_line do |line|
+      begin
+        word, _, separation = eval(line)
+        next if word =~ /\-/
+        s = Pascoale::SyllableSeparator.new(word).separated
+        if s == separation
+          correct_counter += 1
+        else
+          wrong_counter += 1
+          puts "#{word} - #{s} - #{separation}"
+        end
+      rescue => e
+        puts e.message
+      end
+    end
+  end
+ensure
+  puts "   Last: #{word}"
+  puts "Correct: #{correct_counter}"
+  puts "  Wrong: #{wrong_counter}"
+end

data/lib/pascoale/version.rb

@@ -1,3 +1,3 @@
 module Pascoale
-  VERSION = "0.0.1"
+  VERSION = "0.1.0"
 end

data/lib/pascoale.rb

@@ -1,5 +1,10 @@
-require 'pascoale/version'
-require 'pascoale/edits'
-
 module Pascoale
+  def self.root
+    File.expand_path("#{File.dirname(__FILE__)}/..")
+  end
 end
+
+require 'pascoale/version'
+require 'pascoale/constants'
+require 'pascoale/edits'
+require 'pascoale/syllable_separator'

data/pascoale.gemspec

@@ -19,5 +19,6 @@ Gem::Specification.new do |spec|
 
   spec.add_development_dependency 'bundler', ['~>1.5', '>=1.5.1']
   spec.add_development_dependency 'rake'
+  spec.add_development_dependency 'rspec', '~> 3.0.beta'
   spec.add_development_dependency 'guard-rspec'
 end

data/spec/lib/pascoale/syllable_separator_spec.rb

@@ -0,0 +1,150 @@
+require 'spec_helper'
+
+RSpec::Matchers.define :separate_as do |expected|
+  result = nil
+  match do |actual|
+    result = Pascoale::SyllableSeparator.new(actual).separated
+    result == expected
+  end
+  failure_message do |actual|
+    %( expected "#{actual}" to separate as "#{expected}", but was "#{result}")
+  end
+end
+
+describe Pascoale::SyllableSeparator do
+  it 'separates simple words' do
+    expect('bola').to separate_as %w(bo la)
+    expect('batata').to separate_as %w(ba ta ta)
+  end
+
+  it 'keeps some "dígrafos" together' do
+    expect('chocalho').to separate_as %w(cho ca lho)
+    expect('batuque').to separate_as %w(ba tu que)
+    expect('guelha').to separate_as %w(gue lha)
+  end
+
+  it 'separates other "dígrafos"' do
+    expect('bossa').to separate_as %w(bos sa)
+    expect('bosta').to separate_as %w(bos ta)
+    expect('cassado').to separate_as %w(cas sa do)
+    expect('carrasco').to separate_as %w(car ras co)
+  end
+
+  it 'handles double consonants border cases' do
+    expect('destravar').to separate_as %w(des tra var)
+    expect('desencontro').to separate_as %w(de sen con tro)
+    expect('exaltar').to separate_as %w(e xal tar)
+    expect('excomungar').to separate_as %w(ex co mun gar)
+  end
+
+  it 'keeps "oclusivas" and "fricativas" together with "r" or "l"' do
+    expect('brasa').to separate_as %w(bra sa)
+    expect('agrupado').to separate_as %w(a gru pa do)
+    expect('fragrância').to separate_as %w(fra grân ci a)
+    expect('protocriogênico').to separate_as %w(pro to cri o gê ni co)
+  end
+
+  it 'keeps long codas' do
+    expect('transpiração').to separate_as %w(trans pi ra ção)
+    expect('transatlântico').to separate_as %w(tran sa tlân ti co)
+    expect('mirins').to separate_as %w(mi rins)
+    expect('transerrano').to separate_as %w(tran ser ra no)
+    expect('solstício').to separate_as %w(sols tí ci o)
+    expect('perspectiva').to separate_as %w(pers pec ti va)
+    expect('substância').to separate_as %w(subs tân ci a)
+    # Yes! It's a real word o_O
+    expect('falansterialismo').to separate_as %w(fa lans te ri a lis mo)
+  end
+
+  it 'separates single vowels at beggining' do
+    expect('abacaxi').to separate_as %w(a ba ca xi)
+    expect('exceto').to separate_as %w(ex ce to)
+    expect('arrocho').to separate_as %w(ar ro cho)
+  end
+
+  it 'keeps "ditongos" together' do
+    expect('maizena').to separate_as %w(mai ze na)
+    expect('mausoléu').to separate_as %w(mau so léu)
+    expect('ação').to separate_as %w(a ção)
+    expect('põe').to separate_as %w(põe)
+    expect('exceção').to separate_as %w(ex ce ção)
+  end
+
+  it 'handle "tritongos"' do
+    expect('ideia').to separate_as %w(i dei a)
+    expect('tireoide').to separate_as %w(ti re oi de)
+    expect('praia').to separate_as %w(prai a)
+    expect('feio').to separate_as %w(fei o)
+    expect('vaia').to separate_as %w(vai a)
+  end
+
+  it 'separates "hiatos"' do
+    expect('moeda').to separate_as %w(mo e da)
+    expect('leal').to separate_as %w(le al)
+    expect('aéreo').to separate_as %w(a é re o)
+    expect('pior').to separate_as %w(pi or)
+    expect('raíz').to separate_as %w(ra íz)
+    expect('ruído').to separate_as %w(ru í do)
+  end
+
+  it 'keeps the first consonant of the word together' do
+    expect('pneu').to separate_as %w(pneu)
+    expect('apneia').to separate_as %w(ap nei a)
+    expect('pneumático').to separate_as %w(pneu má ti co)
+    expect('piropneumático').to separate_as %w(pi rop neu má ti co)
+    expect('mnemônica').to separate_as %w(mne mô ni ca)
+    expect('pseudônimo').to separate_as %w(pseu dô ni mo)
+    expect('gnomo').to separate_as %w(gno mo)
+  end
+
+  it 'keeps "sineréses" together' do
+    expect('saudade').to separate_as %w(sau da de)
+    expect('vaidade').to separate_as %w(vai da de)
+    expect('suave').to separate_as %w(su a ve)
+
+    # Not sure how to deal with these
+    #expect('traidor').to separate_as %w(trai dor)
+  end
+
+  it 'separates "dieréses"' do
+    expect('conluiador').to separate_as %w(con lui a dor)
+    expect('conluio').to separate_as %w(con lui o)
+    expect('aleluia').to separate_as %w(a le lui a)
+    expect('toluico').to separate_as %w(to lui co)
+
+    expect('alauita').to separate_as %w(a lau i ta)
+
+    expect('rainha').to separate_as %w(ra i nha)
+    expect('tainheira').to separate_as %w(ta i nhei ra)
+
+    expect('construir').to separate_as %w(cons tru ir)
+    expect('destruir').to separate_as %w(des tru ir)
+    expect('destruição').to separate_as %w(des tru i ção)
+
+    #expect('acuidade').to separate_as %w(a cu i da de)
+    #expect('ajuizar').to separate_as %w(a ju i zar)
+  end
+
+  it 'separates random words' do
+    expect('acidentariamente').to separate_as %w(a ci den ta ri a men te)
+    expect('cooperar').to separate_as %w(co o pe rar)
+    expect('abstraído').to separate_as %w(abs tra í do)
+    expect('abstenção').to separate_as %w(abs ten ção)
+    expect('colapso').to separate_as %w(co lap so)
+    expect('piauí').to separate_as %w(pi au í)
+    expect('aguei').to separate_as %w(a guei)
+    expect('compreender').to separate_as %w(com pre en der)
+    expect('caatinga').to separate_as %w(ca a tin ga)
+    expect('atmosfera').to separate_as %w(at mos fe ra)
+
+    expect('tuiuiú').to separate_as %w(tui ui ú)
+
+    # I really don't buy the whole "comes from latin" thing.
+    # Our separation if phonetic based, so, keep it that way!
+    expect('abrupto').to separate_as %w(a brup to)
+    expect('abruptamente').to separate_as %w(a brup ta men te)
+    # For example, the word bellow is correctly separated
+    # (as the dictionary says). =\
+    expect('abrupção').to separate_as %w(a brup ção)
+  end
+end

metadata

@@ -1,61 +1,75 @@
 --- !ruby/object:Gem::Specification
 name: pascoale
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.1.0
 platform: ruby
 authors:
 - Ronie Uliana
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-21 00:00:00.000000000 Z
+date: 2014-04-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
         version: '1.5'
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.5.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
         version: '1.5'
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.5.1
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.0.beta
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.0.beta
 - !ruby/object:Gem::Dependency
   name: guard-rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
 description: 
@@ -65,17 +79,26 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
-- .gitignore
+- ".gitignore"
+- ".ruby-gemset"
+- ".ruby-version"
 - Gemfile
 - Guardfile
 - LICENSE.txt
 - README.md
 - Rakefile
+- data/errors.txt
+- data/everything.txt
+- data/unique_errors.txt
 - lib/pascoale.rb
+- lib/pascoale/constants.rb
 - lib/pascoale/edits.rb
+- lib/pascoale/syllable_separator.rb
+- lib/pascoale/syllable_separator_benchmark.rb
 - lib/pascoale/version.rb
 - pascoale.gemspec
 - spec/lib/pascoale/edits_spec.rb
+- spec/lib/pascoale/syllable_separator_spec.rb
 - spec/spec_helper.rb
 homepage: http://github.com/ruliana/pascoale
 licenses:
@@ -87,20 +110,21 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.1.11
+rubygems_version: 2.2.0
 signing_key: 
 specification_version: 4
 summary: Text processing utilities for Brazilian Portuguese
 test_files:
 - spec/lib/pascoale/edits_spec.rb
+- spec/lib/pascoale/syllable_separator_spec.rb
 - spec/spec_helper.rb