yanser 0.0.1 → 0.0.2

data/test/test_yanser.rb

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 require 'test/unit'
-require 'yanser'
+require 'yanser/yanser'
+require 'yanser/version'
 
 class TestYanser < Test::Unit::TestCase
   def setup
@@ -10,7 +11,7 @@ class TestYanser < Test::Unit::TestCase
   end
   
   def test_public_methods
-    yanser = Yanser.new(
+    yanser = Yanser::Yanser.new(
                         :query_params => {
                           :appid => 'YahooDemo',
                           :query => 'Haus AND grün',
@@ -24,7 +25,10 @@ class TestYanser < Test::Unit::TestCase
     
     assert_respond_to(yanser, :start)
   end
-  
+
+  def test_constants
+    assert(Yanser::VERSION.instance_of?(String))
+  end
   def test_mandatory_options
   end
   

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: yanser
 version: !ruby/object:Gem::Version 
-  hash: 29
-  prerelease: false
+  hash: 27
+  prerelease: 
   segments: 
   - 0
   - 0
-  - 1
-  version: 0.0.1
+  - 2
+  version: 0.0.2
 platform: ruby
 authors: 
 - Andrei Beliankou
@@ -15,8 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-04-06 00:00:00 +02:00
-default_executable: 
+date: 2011-07-08 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: yanapi
@@ -41,28 +40,20 @@ executables:
 extensions: []
 
 extra_rdoc_files: 
-- README.rdoc
+- README
 - LICENSE
 files: 
 - lib/tester.rb
-- lib/yanser.rb
-- lib/option_parser.rb
-- lib/version.rb
+- lib/yanser/error.rb
+- lib/yanser/yanser.rb
+- lib/yanser/opt_parser.rb
+- lib/yanser/version.rb
 - bin/yanser
-- README.rdoc
 - LICENSE
 - Rakefile
 - README
-- test/test_option_parser.rb
-- test/yanapi/test_query.rb
-- test/yanapi/test_term_query.rb
+- test/test_opt_parser.rb
 - test/test_yanser.rb
-- test/data/bad_xml.txt
-- test/data/empty_result.txt
-- test/data/error_code.txt
-- test/data/response_with_error.txt
-- test/data/successfull_response.txt
-has_rdoc: true
 homepage: http://www.uni-trier.de/index.php?id=34451
 licenses: []
 
@@ -95,10 +86,10 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: yanser
-rubygems_version: 1.3.7
+rubygems_version: 1.7.2
 signing_key: 
 specification_version: 3
 summary: Yanser is a convinient search tool providing access to the Yahoo! Answers Q/A collection.
 test_files: 
-- test/test_option_parser.rb
+- test/test_opt_parser.rb
 - test/test_yanser.rb

data/README.rdoc

@@ -1,21 +0,0 @@
-= YANSER
-
-* {RubyGems}[http://rubygems.org/gems/yanser]
-* Developers {Homepage}[http://www.uni-trier.de/index.php?id=24140]
-* {YANAPI Project Page}[http://yanser.rubyforge.org/]
-
-== DESCRIPTION
-
-YANSER (Yahoo! ANSwers harvestER) is a convinient search tool providing access to the Yahoo! Answers Q&A collection. Based on YANAPI it provides a simple CLI and helps to search for Questions and Answers which contain a set of key words, belong to a specific semantic domain or are posted by a certain user. Yanser is a research tool in the field of Computational Linguistics.
-
-== SYNOPSIS
-  $ yanser --help
-
-
-
-== LICENSE
-
-YANSER is a copyrighted software by Andrei Beliankou, 2011.
-You may use, redistribute and change it under the terms
-provided in the LICENSE file.
-

data/lib/option_parser.rb

@@ -1,263 +0,0 @@
-# -*- coding: utf-8 -*-
-require 'optparse'
-require 'fileutils'
-
-require 'version'
-
-class OptionParser
-  # OP expects cmd_args to be an array like ARGV
-  # dummy output for temporary usage
-  def self.parse(cmd_args)
-    options = {}
-    options[:query_params] = {}
-    
-    parser = OptionParser.new do |opts|
-      opts.banner = 'Usage: yanser OPTIONS'
-
-      opts.separator ''
-      opts.separator 'Program specific options:'
-      
-      opts.on('--appid APPID',
-              'Provide an ApplicationID given by Yahoo,',
-              'to test Yanser you can use <YahooDemo> as the APPID,',
-              'think in this case on limitations placed by Yahoo.',
-              'This option is required!'
-              ) do |appid|
-        options[:query_params][:appid] = appid
-      end
-
-
-      opts.separator ''
-      opts.separator '  Mandatory search arguments:'
-      
-      opts.on('-k', '--key-word KEYWORD',
-              'Provide a single keyword or a boolean expression.'
-              ) do |keyword|
-        options[:query_params][:query] = keyword
-        # not a solution!!!
-        options[:query_params][:search_in] = 'question'
-
-      end
-
-      opts.separator ''
-
-      opts.on('-c', '--category CATEGORY',
-              'Provide a category name or ID.'
-              ) do |category|
-
-        if category =~ /^[[:digit:]]+$/
-          options[:query_params][:category_id] = category
-        else
-          options[:query_params][:category_name] = category
-        end
-
-      end
-      
-      opts.separator ''
-
-      opts.on('--user-id ID',
-              'Provide an user ID of questions you search for.',
-              'This way you can get question by a specific user.'
-              ) do |user_id|
-        options[:query_params][:user_id] = user_id
-      end
-      
-      opts.separator ''
-
-      opts.on('--question-id ID',
-              'Provide a question ID of the question you search for.',
-              'It returns a unique question.'
-              ) do |question_id|
-        options[:query_params][:question_id] = question_id
-      end
-
-      opts.separator ''
-      opts.separator '  Optional search arguments:'
-      
-      opts.on('-r', '--region REGION',
-              'Provide a geographic region to search in for terms.',
-              'Possible values are: de, us, uk, ca, au, in, es, br,',
-              '                     ar, mx, e1, it, fr, sg.',
-              'This defaults to en.'
-              ) do |region|
-        # todo
-        prove_region(region)
-        options[:query_params][:region] = region
-      end
-
-      opts.separator ''
-      
-      opts.on('-o', '--output-dir DIR',
-              'Provide an output folder.',
-              'This directory will be created if it does not exist yet.'
-              ) do |output_dir|
-        options[:output_dir] = provide_dir(output_dir)
-      end
-      
-      opts.separator ''
-      
-      opts.on('-l', '--limit NUMBER', Integer,
-              'Provide a number of answers you want to get from Yahoo.',
-              'This argument is not mandatory, if you want to get',
-              'all answers simply ommit this argument.'
-              ) do |limit|
-        options[:limit] = limit
-      end
-
-      opts.separator ''
-      opts.on('-f', '--output-format FORMAT',
-              'Provide an output format: xml, json, rss, php.',
-              'It defaults to xml, and you can simply ommit this option.'
-              ) do |f|
-        raise NotImplementedError, 'Only default output format is implemented'
-      end
-      
-      opts.separator ''
-      opts.on('--prefix PREFIX',
-              'Provide a prefix for the output files. By default',
-              'the filename begins with the index of the retrieved',
-              'question. You can alter this by providing a prefix.',
-              'It can be useful if you want to put many query results',
-              'in the same output folder.'
-              ) do |pref|
-        raise NotImplementedError, 'No prefixes implemeted.'
-      end
-
-      opts.separator ""
-      opts.separator "Common options:"
-
-      opts.on_tail('-h', '--help', 'Show the help message.') do
-        puts opts
-        exit
-      end
-
-      opts.on_tail('-v', '--version', 'Show the program version.') do
-        puts YANSER::VERSION
-        exit
-      end
-    end
-    
-    # if no options provided print the help
-    if cmd_args.empty?
-      $stderr.printf "You have to provide some options.\n\n"
-      puts parser
-      exit 1
-    end
-
-    # Parse ARGV and provide the options hash.
-    # Check if everything is correct and handle exceptions
-    begin
-      parser.parse!(cmd_args)
-    # rescue all exceptions from OptionParser  
-    rescue => e
-      $stderr.printf "#{e.message.capitalize}\n\n"
-      puts parser
-      exit 1
-    end
-
-    # Check to see if we got the required arguments needed.
-    check_required_options(options)
-
-    # Set the search method.
-    options[:query_type] = set_query_type(options[:query_params])
-
-    return options
-  end # parse
-
-  private
-
-  # Check if the value of given region is correct.
-  # Now 14 regions are supported by Yahoo! Answers.
-  def self.prove_region(region)
-    regions = ['de', 'us', 'uk', 'ca', 'au', 'in', 'es',
-	       'br', 'ar', 'mx', 'e1', 'it', 'fr', 'sg']
-    
-    unless regions.include?(region)
-      $stderr.puts "The provided search region #{region} is currently not supported by Yahoo!"
-      exit 1
-    end
-  end
-  
-  # define one of the following query types: TermQuery|CategoryQuery|
-  # QuestionQuery|UserQuery
-  def self.set_query_type(params)
-    case
-    when (params[:category_id] || params[:category_name]) && ! params[:query]
-      return 'CategoryQuery'
-    when params[:query]
-      return 'TermQuery'
-    when params[:user_id]
-      return 'UserQuery'
-    when params[:question_id]
-      return 'QuestionQuery'
-    end
-  end
-
-  def self.check_required_options(options)
-    required_opts = [:appid]
-    
-    required_opts.each do |opt|
-      if options[:query_params].has_key?(opt)
-        next
-      else
-        $stderr.puts "A required option --#{opt} is missing."
-        exit 1
-      end
-    end
-  end
-  
-  def self.provide_dir(dir)
-    dir = File.expand_path(dir)
-    #check for existens
-    if File.directory?(dir)
-      if File.writable?(dir)
-	return dir
-      else
-	$stderr.puts 'The directory you have provided is not writable!'
-	exit 1
-      end
-    else
-      FileUtils.mkdir_p(dir)
-      return dir
-    end
-  end # provide_dir
-  
-end # OptionParser
-
-__END__
-
--c, --category-id # Term & CategorySearch
--q, --question-id # QuestionSearch
-
--t, --time-interval
--f, --output-format
--p, --prefix #prefix for output files
-
-
-
-
-instance interface
-
-{
-  :query_params => {
-    :appid => 'YahooDemo' | 'SomeStringWithYourID',
-    :callback => '',
-    :category_id => '',
-    :category_name => '',
-    :date_range => '',
-    :filter => '',
-    :output => '',
-    :query => '',
-    :question_id => '',
-    :region => 'de'|'us'|'uk'|'ca'|'au'|'in'|'es'|'br'|'ar'|'mx'|'e1'|'it'|'fr'|'sg',  # default 'en'
-    :results => Integer, # 0..50
-    :search_in => "all" | "question" | "best_answer", # default 'all'
-    :sort => 'relevance' | 'date_desc'| 'date_asc', # default 'relevance'
-    :start => Integer, # <= 1000
-    :type => "all" | "resolved" | "open" | "undecided", # default 'all'
-    :user_id => ''
-  },
-  :query_type => 'TermQuery'|'QuestionQuery'|'UserQuery'|'CategoryQuery',
-  :output_dir => 'some path',
-  :prefix => 'some prefix' # prefix for output files 
-}

data/lib/version.rb

@@ -1,3 +0,0 @@
-module YANSER
-  VERSION = '0.0.1'
-end

data/lib/yanser.rb

@@ -1,121 +0,0 @@
-require 'yanapi'
-
-# :title: YANSER, Yahoo! ANSwers harvestER
-# :main: Yanser
-# Main processing class.
-# Yanser encapsulates the main routine and instantiates
-# all other classes.
-#--
-# Yanser takes the users input and validates it.
-# It decides which search method to choose.
-# Then it collects all parameters and useful default values,
-# creates an XyzQuery with the starting point of 0 and gets the first results.
-# If more results were requested, Yanser creates a similar XyzQuery and gets
-# the next result set until the result limitation set by the user is met.
-# 
-class Yanser
-
-  # Yahoo! Answers returns starting not more than at the 1000st question.
-  # It makes no sense to step over.
-  START_LIMIT = 1000
-  
-  # We query the web service every two seconds.
-  QUERY_INTERVAL = 2
-
-  # Yahoo! Answers returns maximum 50 results.
-  MAX_RESULTS = 50
-  
-  def initialize(opts)
-
-    # the minimal output of an OptionParser
-    # {:query_type=>u|q|c|w, :query_params=>{appid, start, results, query}}
-    # opts come from the OptionParser
-    # they are supposed to be correct, no validation here
-    @options = opts
-
-    @options[:limit] = @options[:limit] || START_LIMIT + MAX_RESULTS
-
-    @query_params = @options[:query_params]
-
-    @output_type = @query_params[:output] || 'xml'
-
-    @query_params[:start] = @query_params[:start] || 0
-  end
-
-  # TODO: implement some logging
-  def start
-
-    if @options[:query_type] == 'QuestionQuery'
-      query(@query_params)
-    elsif @options[:limit] < MAX_RESULTS
-      @query_params[:results] = @options[:limit]
-      query(@query_params)
-    else
-      @query_params[:results] = MAX_RESULTS
-
-      while query(@query_params) do
-        # we get the next start point here
-        @query_params[:start] += MAX_RESULTS
-
-        results_left = @options[:limit] - @query_params[:start]
-        if results_left == 0
-          break
-        elsif results_left < MAX_RESULTS
-          @query_params[:results] = results_left
-          break
-        end
-      end # while
-      
-    end # if
-    
-  end # start
-
-  private
-
-  def query(params)
-    q = create_query(params)
-    tries = 0
-    begin
-      tries += 1
-      result = q.get
-      output(result)
-      sleep(2)
-    rescue YANAPI::EmptyResponse => e
-      $sderr.puts e
-      return false # do not iterate futher
-    rescue => e # some errors to retry
-      if (tries < 4)
-        sleep(QUERY_INTERVAL**tries)
-        retry
-      else
-        $stderr.puts e
-        return false # do not iterate futher
-      end
-    end
-    return true # we may iterate futher
-  end
-
-  def create_query(params)
-    eval("YANAPI::#{@options[:query_type]}.new(params)")
-  end
-  
-  def output(result)
-    if @options[:output_dir]
-      save(result)
-    else
-      puts result
-    end
-  end
-
-  # save results to a dir
-  # this dir exists since has been proved by OptionParser
-  # not a good implementation
-  # interface (filename, data)
-  def save(result)
-    filename = File.join(@options[:output_dir], "#{@query_params[:start]}.#{@output_type}")
-    file = File.new(filename, 'w')
-    file.puts result
-    file.close
-  end
-
-end # Yanser