nebrija 0.0.0

data/Rakefile

@@ -0,0 +1,9 @@
+require 'rake/testtask'
+
+Rake::TestTask.new do |t|
+	t.libs << "test"
+	t.test_files = FileList['test/test*.rb']
+end
+
+desc "Run tests"
+task :default => :test

data/bin/nebrija

@@ -0,0 +1,9 @@
+#!/usr/bin/env ruby
+
+require 'json'
+require 'nebrija'
+
+puts 'Oh, hai!'
+puts 'In future versions this will print something more human readable.'
+puts
+puts JSON.pretty_generate(HTTPRae.new.search(ARGV[0]))

data/lib/nebrija/parser.rb

@@ -0,0 +1,75 @@
+require 'nokogiri'
+
+class Parser
+  
+  META_REGEX = /^([a-zA-Z]{1,4}+\.[ ]{1,2})+/
+
+  def initialize(rae_data, word)
+    @doc = Nokogiri::HTML(rae_data
+                      .gsub(/[\n]+/, '')
+                      .gsub(/[ ]{2,}+/, ' '))
+    @word = word
+  end
+
+  def parse
+
+    return {:error => 'Word does not exist. Sorry.'} if !valid? 
+
+    if single?
+      parse_single
+    else
+      parse_multiple
+    end
+  end
+
+  def single?
+    @doc.css('body > ul').length.zero?
+  end
+
+  private 
+  def parse_single
+    data = []
+    result = {:id => @doc.css('body > div > a').first['name'].to_i, :data => data}
+    state = :entry # TODO. Improve FSM syntax.
+    index = -1 # HACK(javierhonduco)
+
+    @doc.css('body > div > p').each do |entry|
+      if entry['class'] == 'p' and state == :entry
+        word = entry.css('span').inner_text
+        word = '=>' if word == ''
+        data << {
+          :word => word.gsub(/~/, @word).strip.capitalize, 
+          :meanings => []
+        }
+        index+=1
+      else
+        text = entry.inner_text.strip.gsub(/[0-9]+\.[ ]/, '')
+        next if text[0] == '(' # Del latín, Nil.    
+        unparsed_meta = text.scan META_REGEX
+        text = text.gsub(META_REGEX, '')
+        data[index][:meanings] << {
+          :word => text, 
+          :meta => (unparsed_meta.join.strip if unparsed_meta.join.strip != ''),
+        } if !text.nil? and text != ''
+        state = :definitions
+      end
+      state = :entry
+    end 
+    result
+  end
+
+  def parse_multiple
+    multiple_result = []
+    @doc.css('body > ul > li > a').each do |word|
+      multiple_result << {
+        :word => word.css('span').first.inner_text,
+        :href => word['href'].gsub(/search\?id=/, '')
+      }
+    end 
+    multiple_result
+  end
+
+  def valid?
+    (@doc.css('title').inner_text =~/error/).nil?
+  end
+end

data/lib/nebrija.rb

@@ -0,0 +1,64 @@
+require 'nebrija/parser'
+require 'typhoeus'
+
+
+class Rae
+
+  def search(word)
+    Parser.new(query(word), word).parse
+  end
+
+  private
+  def query(word)
+    raise 'NotImplementedError'
+  end
+end
+
+
+class FileRae < Rae
+
+  private
+  def query(file)
+    IO.read(file)
+  end
+end
+
+
+class HTTPRae < Rae
+  USER_AGENT = 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2049.0 Safari/537.36'
+  SEARCH_URL = 'http://lema.rae.es/drae/srv/search?'
+  REQUEST_TIMEOUT = 
+  ID_REGEX = /[0-9]/
+
+  private
+  def query(word)
+    @word = word
+
+    params = 'id=' 
+    params = 'val=' if val?
+
+    response = Typhoeus::Request.post(
+      "http://lema.rae.es/drae/srv/search?#{params}#{word}",
+      body: build_headers
+    )
+    response.body
+  end
+
+  def val?
+    (@word =~ ID_REGEX).nil?
+  end
+
+  def build_headers
+    {
+      'TS014dfc77_id' => 3,
+      'TS014dfc77_cr' => '42612abd48551544c72ae36bc40f440a%3Akkmj%3AQG60Q2v4%3A1477350835',
+      'TS014dfc77_76' => 0,
+      'TS014dfc77_md' => 1,
+      'TS014dfc77_rf' => 0,
+      'TS014dfc77_ct' => 0,
+      'TS014dfc77_pd' => 0
+    }.map {|key, value|
+      "#{key}=#{value}"
+    }.join('&')
+  end
+end

data/test/test_basic.rb

@@ -0,0 +1,63 @@
+require 'test/unit'
+require 'nebrija' 
+
+MOCKS_DIR = "#{Dir.pwd}/test/mocks"
+
+class TestMockedParserBasic < Test::Unit::TestCase
+
+  def test_error_basic
+    assert_not_nil FileRae.new.search("#{MOCKS_DIR}/error.html")[:error] 
+  end
+  
+  def test_single_basic
+    assert_not_nil FileRae.new.search("#{MOCKS_DIR}/single.html")[:data]
+  end
+
+  def test_multiple_basic
+    assert FileRae.new.search("#{MOCKS_DIR}/multiple.html").length == 2
+  end
+end
+
+class TestMockedParserContent < Test::Unit::TestCase
+  
+  def test_single_basic
+    assert FileRae.new.search("#{MOCKS_DIR}/single.html")[:data].length > 20
+  end
+
+  def test_multiple_basic
+    assert FileRae.new.search("#{MOCKS_DIR}/multiple.html")[0][:word] == 'bancar' 
+    assert FileRae.new.search("#{MOCKS_DIR}/multiple.html")[1][:word] == 'banco'  
+  end
+end
+
+
+class TestMockedParserBasic < Test::Unit::TestCase
+
+  def test_single_basic_id
+    assert_not_nil HTTPRae.new.search('MHpGWYJ6YDXX2bw9Ghwm')[:data]
+  end
+
+  def test_error_basic
+    assert_not_nil HTTPRae.new.search('jddhfgsd')[:error] 
+  end
+  
+  def test_single_basic
+    assert_not_nil HTTPRae.new.search('a')[:data]
+  end
+
+  def test_multiple_basic
+    assert HTTPRae.new.search('banco').length == 2
+  end
+end
+
+class TestParserContent < Test::Unit::TestCase
+  
+  def test_single_basic
+    assert HTTPRae.new.search('a')[:data].length > 4
+  end
+
+  def test_multiple_basic
+    assert HTTPRae.new.search('banco')[0][:word] == 'bancar' 
+    assert HTTPRae.new.search('banco')[1][:word] == 'banco'  
+  end
+end

metadata

@@ -0,0 +1,100 @@
+--- !ruby/object:Gem::Specification
+name: nebrija
+version: !ruby/object:Gem::Version
+  version: 0.0.0
+  prerelease: 
+platform: ruby
+authors:
+- ! '@javierhonduco'
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-07-11 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: typhoeus
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !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'
+description: A gem to access the rae dictionary
+email: a@a.a
+executables:
+- nebrija
+extensions: []
+extra_rdoc_files: []
+files:
+- Rakefile
+- lib/nebrija.rb
+- lib/nebrija/parser.rb
+- bin/nebrija
+- test/test_basic.rb
+homepage: http://rubygems.org/gems/nebrija
+licenses:
+- MIT
+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: 1.8.23
+signing_key: 
+specification_version: 3
+summary: dictionary gem and stuff
+test_files:
+- test/test_basic.rb