yandex_mystem 0.0.1

data/.gitignore

@@ -0,0 +1,18 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp
+.idea/

data/.rspec

@@ -0,0 +1,2 @@
+--color
+--format progress

data/.rvmrc

@@ -0,0 +1 @@
+rvm use 1.9.3 --create

data/Gemfile

@@ -0,0 +1,4 @@
+source 'http://rubygems.org'
+
+# Specify your gem's dependencies in yandex_mystem.gemspec
+gemspec

data/README.md

@@ -0,0 +1,26 @@
+# Yandex Mystem
+
+## Introduction
+
+Mystem is a software that provided by the Yandex only for non-commercial project. With use of it you can detect base forms of the words in a text, make a simple morphological analysis of russian words.
+
+## License
+
+First of all, read license on http://company.yandex.ru/technologies/mystem/
+
+`Mystem` available only for non-commercial usage.
+
+## OS
+
+This gem contains executables for there platforms:
+
+* Windows
+* Linux 2.6 32-bit
+* Linux 2.6 64-bit
+* Mac OS X 10.5
+
+...of six, FreeBSD not in the gem. If you need it, add pull request or issue.
+
+## Usage
+
+    YandexMystem::Base.stem 'О предложении в котором много слов.'

data/Rakefile

@@ -0,0 +1,6 @@
+#!/usr/bin/env rake
+require "bundler/gem_tasks"
+require 'rspec/core/rake_task'
+
+RSpec::Core::RakeTask.new(:spec)
+task default: :spec

data/lib/yandex_mystem.rb

@@ -0,0 +1,51 @@
+require 'open3'
+require 'yandex_mystem/version'
+
+module YandexMystem
+  class Base
+    # TODO add -i
+    def self.stem(text)
+      exec = Array(command).tap do |c|
+        c << '-e utf-8 -n'
+      end.join(' ')
+
+      data = Open3.popen3(exec) do |stdin, stdout, stderr|
+        stdin.write text
+        stdin.close
+        #stderr.read
+        stdout.read
+      end
+
+      data = data.scan(/^([^\{]+)\{(.+)\}$/).map do |(word, words)|
+        words = words.split('|').select do |w|
+          !(w =~ /.+\?\?$/)
+        end
+
+        [word, words]
+      end.flatten(1)
+
+      Hash[*data]
+    end
+
+    private
+
+    def self.command
+      postfix = if RUBY_PLATFORM =~ /(win|w)32$/
+                  'win.exe'
+                elsif RUBY_PLATFORM =~ /32.+linux$/
+                  'linux-32'
+                elsif RUBY_PLATFORM =~ /64.+linux$/
+                  'linux-64'
+                elsif RUBY_PLATFORM =~ /darwin/
+                  'mac'
+                elsif RUBY_PLATFORM =~ /freebsd/
+                  raise 'Create an issue or add pull request on a github.'
+                else
+                  raise 'Unknown OS'
+                end
+
+      path = Pathname.new(__FILE__) + '../../app/'
+      path + "mystem-#{postfix}"
+    end
+  end
+end

data/lib/yandex_mystem/version.rb

@@ -0,0 +1,3 @@
+module YandexMystem
+  VERSION = "0.0.1"
+end

data/spec/spec_helper.rb

@@ -0,0 +1,7 @@
+require 'yandex_mystem'
+
+RSpec.configure do |config|
+  config.treat_symbols_as_metadata_keys_with_true_values = true
+  config.run_all_when_everything_filtered = true
+  config.filter_run :focus
+end

data/spec/yandex_mystem_spec.rb

@@ -0,0 +1,15 @@
+# encoding: utf-8
+require 'spec_helper'
+
+describe YandexMystem do
+  it "should stem words" do
+    data = YandexMystem::Base.stem('мальчики мальчиков девочки девочек компьютеров компьютере сов пошли elements')
+    data['мальчики'].should eq ['мальчик']
+    data['мальчиков'].should eq ['мальчик', "мальчиков", "мальчиковый"]
+    data['девочки'].should eq ['девочка']
+    data['девочек'].should eq ['девочка']
+    data['сов'].should eq ['сова']
+    data['пошли'].should eq %w(пойти посылать)
+    data['elements'].should eq []
+  end
+end

data/yandex_mystem.gemspec

@@ -0,0 +1,19 @@
+# -*- encoding: utf-8 -*-
+require File.expand_path('../lib/yandex_mystem/version', __FILE__)
+
+Gem::Specification.new do |gem|
+  gem.authors       = ["Dmitry Polushkin"]
+  gem.email         = ["dmitry.polushkin@gmail.com"]
+  gem.description   = %q{Mystem is a software that provided by the Yandex only for non-commercial project. With use of it you can detect base forms of the words in a text, make a simple morphological analysis of russian words.}
+  gem.summary       = %q{Yandex Mystem makes morphological analysis of a russian text}
+  gem.homepage      = ""
+
+  gem.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  gem.files         = `git ls-files`.split("\n")
+  gem.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
+  gem.name          = "yandex_mystem"
+  gem.require_paths = ["lib"]
+  gem.version       = YandexMystem::VERSION
+
+  gem.add_development_dependency "rspec", '~> 2.8'
+end

metadata

@@ -0,0 +1,73 @@
+--- !ruby/object:Gem::Specification
+name: yandex_mystem
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+  prerelease: 
+platform: ruby
+authors:
+- Dmitry Polushkin
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2012-01-09 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: &18846940 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '2.8'
+  type: :development
+  prerelease: false
+  version_requirements: *18846940
+description: Mystem is a software that provided by the Yandex only for non-commercial
+  project. With use of it you can detect base forms of the words in a text, make a
+  simple morphological analysis of russian words.
+email:
+- dmitry.polushkin@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- .rspec
+- .rvmrc
+- Gemfile
+- README.md
+- Rakefile
+- app/mystem-linux-32
+- app/mystem-linux-64
+- app/mystem-mac
+- app/mystem-win.exe
+- lib/yandex_mystem.rb
+- lib/yandex_mystem/version.rb
+- spec/spec_helper.rb
+- spec/yandex_mystem_spec.rb
+- yandex_mystem.gemspec
+homepage: ''
+licenses: []
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.10
+signing_key: 
+specification_version: 3
+summary: Yandex Mystem makes morphological analysis of a russian text
+test_files: []