morphy 0.0.3 → 0.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ffbef5e95829de72124f62a477969795869948e5
-  data.tar.gz: 462188fe9f72702fe28d8bb9549118fb959fa26c
+  metadata.gz: fcbdc7a59f41e79c1cf9ac95ecec4887176c5024
+  data.tar.gz: 4277e263c6d63ba16540b196c0207d7ec98fadcd
 SHA512:
-  metadata.gz: ae7c186490bd5f217953f27727879d9afaccc3c21a2f54ab53aa117d6d64822695d187d21974d9f484c41531c5a1db01a5ee477c0c035e5c61f78bfea063948d
-  data.tar.gz: 160c9cfa219860ecebc06955ac165b93f9db067fd6c7b3c67d33160113b8ce17a5638fb29c653100991bfaca8f57f49ca1e9419aca3f3d750df6ca39f899bf6d
+  metadata.gz: c0fddd3b4910808a4160474e9214ad09138a745fd0e95e61fe774017bfa782c09c8ac6f4cf411357e76b514338f2bf78154d4ce0f4fe02cd30ef0f33c19a1a78
+  data.tar.gz: c58a7f67a75eebd7223458927d79f678d0f25e360f9745287fbe5901f1e8ab9c27e3a2f49742e1a29806f3ab00ffa17dd7fdaece354e9d2337d0b4cb0cdc7960

data/README.md

@@ -15,19 +15,20 @@ Or install it yourself as:
 
     $ gem install morphy
 
-## Usage
-    # Words must be added in alphabetical order
+## Usage    
     require "morphy"
-    
+
     morphy = Morphy.new
-    
-    word = morphy.find_similar("облако").first
+
+    word = morphy.query("облако").first
 
     datv = word.inflect(["datv"])
     datv.to_s
     => облаку
     datv.inflect(["nomn"]).to_s
     => облако
+    datv.inflect(["nomn", "plur"]).to_s
+    => облака
     datv.lexemme.map(&:to_s)
     => ["облако", "облака", "облаку", "облако", "облаком", "облаке", "облака", "облаков", "облакам", "облака", "облаками", "облаках"]
     datv.normal_form
@@ -35,6 +36,8 @@ Or install it yourself as:
 
 
 
+
+
 ## Contributing
 
 1. Fork it
@@ -42,4 +45,3 @@ Or install it yourself as:
 3. Commit your changes (`git commit -am 'Add some feature'`)
 4. Push to the branch (`git push origin my-new-feature`)
 5. Create new Pull Request
-

data/lib/morphy.rb

@@ -1,98 +1,53 @@
 require "dawg"
+require_relative "word"
 
-class Morphy
-  class Word
-    attr_accessor :para_id
-    
-    def initialize(word,para_id,index)
-      @word = word
-      @para_id = para_id.to_i
-      @index = index.to_i
-      @prefix_id = Morphy.paradigms[@para_id][@index*3]
-      @suffix_id = Morphy.paradigms[@para_id][@index*3+1]
-      @grammeme_id = Morphy.paradigms[@para_id][@index*3+2]
-      
-    end
-    
-    def to_s
-      @word      
-    end
-    
-    def normal_form
-      prefix = Morphy.prefixes[Morphy.paradigms[@para_id][0]]
-      suffix = Morphy.suffixes[Morphy.paradigms[@para_id][1]]      
-      
-      "#{prefix}#{stem}#{suffix}"
-    end
-    def grammemes
-      Morphy.grammemes[@grammeme_id]      
-    end
-    def stem
-      prefix = Morphy.prefixes[Morphy.paradigms[@para_id][0]]
-      suffix = Morphy.suffixes[Morphy.paradigms[@para_id][1]]
-      grammeme = Morphy.grammemes[Morphy.paradigms[@para_id][2]]
-      word = @word.dup
-      word.sub!(Morphy.prefixes[@prefix_id],"")
-      word = word.reverse.sub(Morphy.suffixes[@suffix_id],"").reverse
-      word
-    end
+module Morphy
 
-    def same_paradigm?(other)
-      @para_id == other.para_id      
-    end
+  class Morphy
 
-    def tag
-      Morphy.grammemes[@grammeme_id].join(",") 
+    def initialize
+      @dawg = Dawg.load("#{::Morphy.path}/dawg.bin")
     end
 
-    def lexemme
-      (0..(Morphy.paradigms[@para_id].length / 3)-1).map do |index|        
-        prefix = Morphy.prefixes[Morphy.paradigms[@para_id][index*3]]
-        suffix = Morphy.suffixes[Morphy.paradigms[@para_id][index*3+1]]
-        grammeme = Morphy.grammemes[Morphy.paradigms[@para_id][index*3+2]]
-        Word.new(prefix+stem+suffix,@para_id,index)
-      end      
+    def query(word)
+      results = @dawg.query(word)
+      results = results.map do |result|
+        result = result.to_s
+        word, para_id, index = result.split(' ')
+        Word.new(word, para_id, index)
+      end
+      results
     end
 
-    def inflect(grammemes)
-      words = lexemme
-      words.each do |word|
-        return word if word.grammemes.last(grammemes.length) == grammemes
-      end
-      nil
-    end      
+    def to_s
+      "Morphy"
+    end
   end
-  def initialize    
-    
-    path = File.dirname(__FILE__)+"/dictionary/"
 
-    @dawg = Dawg.load("#{path}/dawg.dat") # why it's eating so much memory?
-    @@suffixes ||= File.open("#{path}/suffixes.txt", 'r').read.split("\n")
-    @@prefixes ||= File.open("#{path}/prefixes.txt", 'r').read.split("\n")
-    @@grammemes ||= File.open("#{path}/grammemes.txt", 'r').read.split("\n").map{|g| g.split(",")}    
-    @@paradigms ||= Marshal.load(File.read("#{path}/paradigms.dat"))
-  end
-  def self.paradigms
-    @@paradigms    
+  extend self
+
+  def new
+    Morphy.new
   end
-  def self.prefixes
-    @@prefixes    
+
+  def path
+    File.dirname(__FILE__)+"/dictionary"
   end
-  def self.suffixes
-    @@suffixes
+
+  def paradigms
+    @@paradigms ||= Marshal.load(File.read("#{path}/paradigms.dat"))
   end
-  def self.grammemes
-    @@grammemes    
+
+  def prefixes
+    @@prefixes ||= File.open("#{path}/prefixes.txt", 'r').read.split("\n")
   end
-  def find_similar(word)
-    results = @dawg.find_similar(word)
-    results = results.map do |result|
-      word,para_id,index = result.split(" ")
-      Word.new(word,para_id,index)
-    end
-    results
+
+  def suffixes
+    @@suffixes ||= File.open("#{path}/suffixes.txt", 'r').read.split("\n")
   end
-  def to_s
-    "Morphy"    
+
+  def grammemes
+    @@grammemes ||= File.open("#{path}/grammemes.txt", 'r').read.split("\n").map{|g| g.split(",")}
   end
-end
+
+end

data/lib/word.rb

@@ -0,0 +1,61 @@
+module Morphy
+  class Word
+    attr_accessor :para_id
+
+    def initialize(word,para_id,index)
+      @word = word
+      @para_id = para_id.to_i
+      @index = index.to_i
+      @prefix_id = ::Morphy.paradigms[@para_id][@index * 3]
+      @suffix_id = ::Morphy.paradigms[@para_id][@index * 3 + 1]
+      @grammeme_id = ::Morphy.paradigms[@para_id][@index * 3 + 2]
+    end
+
+    def to_s
+      @word
+    end
+
+    def normal_form
+      self.inflect(["nomn"])
+    end
+
+    def grammemes
+      ::Morphy.grammemes[@grammeme_id]
+    end
+
+    def stem
+      prefix = ::Morphy.prefixes[::Morphy.paradigms[@para_id][0]]
+      suffix = ::Morphy.suffixes[::Morphy.paradigms[@para_id][1]]
+      grammeme = ::Morphy.grammemes[::Morphy.paradigms[@para_id][2]]
+      word = @word.dup
+      word.sub!(::Morphy.prefixes[@prefix_id], '')
+      word = word.reverse.sub(::Morphy.suffixes[@suffix_id].reverse, '').reverse
+      word
+    end
+
+    def same_paradigm?(other)
+      @para_id == other.para_id
+    end
+
+    def tag
+      ::Morphy.grammemes[@grammeme_id].join(',')
+    end
+
+    def lexemme
+      (0..(::Morphy.paradigms[@para_id].length / 3) - 1).map do |index|
+        prefix = ::Morphy.prefixes[::Morphy.paradigms[@para_id][index * 3]]
+        suffix = ::Morphy.suffixes[::Morphy.paradigms[@para_id][index * 3 + 1]]
+        grammeme = ::Morphy.grammemes[::Morphy.paradigms[@para_id][index * 3 + 2]]
+        Word.new(prefix + stem + suffix, @para_id, index)
+      end
+    end
+
+    def inflect(grammemes)
+      words = lexemme
+      words.each do |word|
+        return word if (word.grammemes & grammemes).length == grammemes.length
+      end
+      nil
+    end
+  end
+end

data/morphy.gemspec

@@ -2,8 +2,8 @@
 
 Gem::Specification.new do |spec|
   spec.name          = "morphy"
-  spec.version       = "0.0.3"
-  spec.date          = '2015-05-29'
+  spec.version       = "0.0.4"
+  spec.date          = '2017-03-19'
   spec.authors       = ["Maksatbek Mansurov"]
   spec.email         = ["maksat.mansurov@gmail.com"]
   spec.description   = %q{Morphological analyzer (POS tagger + inflection engine) for Russian language in ruby. Inspired by pymorphy2}
@@ -14,6 +14,6 @@ Gem::Specification.new do |spec|
   spec.files         = `git ls-files`.split($/)
   spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
-  spec.require_paths = ["lib"]    
-  spec.add_development_dependency 'dawg', '~> 0.0', '>= 0.0.2'
+  spec.require_paths = ["lib"]
+  spec.add_runtime_dependency 'dawg', '~> 0.0', '>= 0.0.4'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: morphy
 version: !ruby/object:Gem::Version
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors:
 - Maksatbek Mansurov
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-05-29 00:00:00.000000000 Z
+date: 2017-03-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: dawg
@@ -19,8 +19,8 @@ dependencies:
         version: '0.0'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.0.2
-  type: :development
+        version: 0.0.4
+  type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
@@ -29,7 +29,7 @@ dependencies:
         version: '0.0'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.0.2
+        version: 0.0.4
 description: Morphological analyzer (POS tagger + inflection engine) for Russian language
   in ruby. Inspired by pymorphy2
 email:
@@ -42,12 +42,13 @@ files:
 - Gemfile
 - LICENSE
 - README.md
-- lib/dictionary/dawg.dat
+- lib/dictionary/dawg.bin
 - lib/dictionary/grammemes.txt
 - lib/dictionary/paradigms.dat
 - lib/dictionary/prefixes.txt
 - lib/dictionary/suffixes.txt
 - lib/morphy.rb
+- lib/word.rb
 - morphy.gemspec
 homepage: https://github.com/baltavay/morphy
 licenses:
@@ -69,7 +70,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.2.2
+rubygems_version: 2.4.5.1
 signing_key: 
 specification_version: 4
 summary: Morphological analyzer for Russian language in ruby. Inspired by pymorphy2