picolena 0.1.1 → 0.1.2

data/History.txt

@@ -1,3 +1,11 @@
+== 0.1.2  2008-04-20
+
+* major enhancement:
+  * complete Indexer & Index rewrite
+  * new DSL syntax
+  * multi-threaded Indexer
+
+
 == 0.1.1  2008-04-12
 
 * major enhancement:

data/Manifest.txt

@@ -9,6 +9,7 @@ config/hoe.rb
 config/requirements.rb
 lib/picolena/USAGE
 lib/picolena/config/basic.rb
+lib/picolena/config/icons_and_filetypes.yml
 lib/picolena/config/indexed_directories.yml
 lib/picolena/config/title_and_names_and_links.yml
 lib/picolena/config/white_list_ip.yml
@@ -21,6 +22,11 @@ lib/picolena/templates/app/helpers/application_helper.rb
 lib/picolena/templates/app/helpers/documents_helper.rb
 lib/picolena/templates/app/models/document.rb
 lib/picolena/templates/app/models/finder.rb
+lib/picolena/templates/app/models/index_reader.rb
+lib/picolena/templates/app/models/index_writer.rb
+lib/picolena/templates/app/models/indexer.rb
+lib/picolena/templates/app/models/plain_text_extractor.rb
+lib/picolena/templates/app/models/query.rb
 lib/picolena/templates/app/views/documents/_document.html.haml
 lib/picolena/templates/app/views/documents/cached.html.haml
 lib/picolena/templates/app/views/documents/content.html.haml
@@ -35,27 +41,26 @@ lib/picolena/templates/config/environments/test.rb
 lib/picolena/templates/config/initializers/001_load_custom_config.rb
 lib/picolena/templates/config/initializers/002_load_indexed_dirs.rb
 lib/picolena/templates/config/initializers/003_load_white_list_IPs.rb
-lib/picolena/templates/config/initializers/004_load_filters.rb
+lib/picolena/templates/config/initializers/004_load_plain_text_extractors.rb
 lib/picolena/templates/config/initializers/005_load_custom_title_and_names_and_links.rb
+lib/picolena/templates/config/initializers/006_load_icons.rb
 lib/picolena/templates/config/routes.rb
 lib/picolena/templates/lang/ui/de.yml
 lib/picolena/templates/lang/ui/en.yml
 lib/picolena/templates/lang/ui/es.yml
 lib/picolena/templates/lang/ui/fr.yml
 lib/picolena/templates/lib/core_exts.rb
-lib/picolena/templates/lib/ff.rb
-lib/picolena/templates/lib/filter.rb
-lib/picolena/templates/lib/filter_DSL.rb
-lib/picolena/templates/lib/filters/adobe.pdf.rb
-lib/picolena/templates/lib/filters/html.rb
-lib/picolena/templates/lib/filters/ms.excel.rb
-lib/picolena/templates/lib/filters/ms.powerpoint.rb
-lib/picolena/templates/lib/filters/ms.rtf.rb
-lib/picolena/templates/lib/filters/ms.word.rb
-lib/picolena/templates/lib/filters/opendocument.presentation.rb
-lib/picolena/templates/lib/filters/opendocument.spreadsheet.rb
-lib/picolena/templates/lib/filters/opendocument.text.rb
-lib/picolena/templates/lib/filters/plain_text.rb
+lib/picolena/templates/lib/plain_text_extractor_DSL.rb
+lib/picolena/templates/lib/plain_text_extractors/adobe.pdf.rb
+lib/picolena/templates/lib/plain_text_extractors/html.rb
+lib/picolena/templates/lib/plain_text_extractors/ms.excel.rb
+lib/picolena/templates/lib/plain_text_extractors/ms.powerpoint.rb
+lib/picolena/templates/lib/plain_text_extractors/ms.rtf.rb
+lib/picolena/templates/lib/plain_text_extractors/ms.word.rb
+lib/picolena/templates/lib/plain_text_extractors/opendocument.presentation.rb
+lib/picolena/templates/lib/plain_text_extractors/opendocument.spreadsheet.rb
+lib/picolena/templates/lib/plain_text_extractors/opendocument.text.rb
+lib/picolena/templates/lib/plain_text_extractors/plain_text.rb
 lib/picolena/templates/lib/tasks/annotations.rake
 lib/picolena/templates/lib/tasks/index.rake
 lib/picolena/templates/lib/tasks/install_dependencies.rake
@@ -112,10 +117,14 @@ lib/picolena/templates/spec/helpers/application_helper_spec.rb
 lib/picolena/templates/spec/helpers/documents_helper_spec.rb
 lib/picolena/templates/spec/models/basic_finder_spec.rb
 lib/picolena/templates/spec/models/document_spec.rb
-lib/picolena/templates/spec/models/filters_spec.rb
 lib/picolena/templates/spec/models/finder_spec.rb
 lib/picolena/templates/spec/models/host_indexing_system_spec.rb
 lib/picolena/templates/spec/models/index_directories_spec.rb
+lib/picolena/templates/spec/models/index_reader_spec.rb
+lib/picolena/templates/spec/models/index_writer_spec.rb
+lib/picolena/templates/spec/models/indexer_spec.rb
+lib/picolena/templates/spec/models/plain_text_extractor_spec.rb
+lib/picolena/templates/spec/models/query_spec.rb
 lib/picolena/templates/spec/rcov.opts
 lib/picolena/templates/spec/spec.opts
 lib/picolena/templates/spec/spec_helper.rb
@@ -140,6 +149,10 @@ lib/picolena/templates/spec/test_dirs/indexed/different_encodings/iso-8859-1.txt
 lib/picolena/templates/spec/test_dirs/indexed/different_encodings/iso-8859-15.txt
 lib/picolena/templates/spec/test_dirs/indexed/different_encodings/utf-8.txt
 lib/picolena/templates/spec/test_dirs/indexed/just_one_doc/for_test.txt
+lib/picolena/templates/spec/test_dirs/indexed/lang/goethe
+lib/picolena/templates/spec/test_dirs/indexed/lang/hugo
+lib/picolena/templates/spec/test_dirs/indexed/lang/lorca
+lib/picolena/templates/spec/test_dirs/indexed/lang/shakespeare
 lib/picolena/templates/spec/test_dirs/indexed/literature/Simulation of district heating systems for evaluation of real-time control strategies.pdf
 lib/picolena/templates/spec/test_dirs/indexed/literature/Types of malfunction in DH substations.doc
 lib/picolena/templates/spec/test_dirs/indexed/others/'weird'filename.txt

data/README.txt

@@ -14,7 +14,7 @@ Picolena is a lightweight ferret-powered documents search engine written in Ruby
 
 Picolena has many advantages:
 
-   * it can index .pdf, .doc, .docx, .odt, .xls, .ods, .ppt, .pptx, .odp, .rtf, .html and plain text files will full text search, and offers a very easy way to add new filters to index other filetype.
+   * it can index .pdf, .doc, .docx, .odt, .xls, .ods, .ppt, .pptx, .odp, .rtf, .html and plain text files will full text search, and offers a very easy way to add new extractors to index other filetype.
    * it is free as in free beer and as in free speech
    * thanks to Ferret, it is very fast
    * it keeps your data private. By default, only the computer on which it is installed can get access to the search engine. Other IP addresses can then be added to a white list.

data/config/files_to_clean

@@ -4,9 +4,10 @@ lib/picolena/templates/config/custom/picolena.rb
 lib/picolena/templates/config/custom/indexed_directories.yml
 lib/picolena/templates/config/custom/white_list_ip.yml
 lib/picolena/templates/config/custom/title_and_names_and_links.yml
+lib/picolena/templates/config/custom/icons_and_filetypes.yml
 lib/picolena/templates/log
 lib/picolena/templates/spec/test_dirs/indexed/others/bäñüßé.txt
 lib/picolena/templates/tmp
 lib/picolena/templates/vendor
 lib/picolena/templates/coverage
-lib/picolena/templates/doc
+lib/picolena/templates/doc

data/config/hoe.rb

@@ -61,7 +61,7 @@ hoe = Hoe.new(GEM_NAME, VERS) do |p|
   # == Optional
   p.changes = p.paragraphs_of("History.txt", 0..1).join("\n\n")
   p.extra_deps = [ ['rails', '>= 2.0.2'],
-                   # Ferret 0.11.6 is not yet available for win32
+                   # Ferret 0.11.6 is not yet available for win32 (and will never be, it is a fix release for *nix)
                    # FIXME: How to require 0.11.6 for *nix and 0.11.5 for win32?
                    ['ferret', '>= 0.11.5'],
                    ['haml', '>= 1.8.2'],

data/lib/picolena/config/basic.rb

@@ -1,35 +1,46 @@
-# Specify indexes path.
-# Storage should be sufficient in order to store all indexed data.
-IndexesSavePath=File.join(RAILS_ROOT, 'tmp/ferret_indexes/')
-
-
-# Which language should be used?
-# English (:en), German (:de), French (:fr) and Spanish (:es) are currently supported
-# English is chosen by default.
-# If you'd like to use another language, you can find templates in #{RAILS_ROOT}/lang/ui,
-# then add your own language in this directory, and modify this line:
-Globalite.language = :en
-
-
-# Specify which locale should be used by Ferret
-Ferret.locale = "en_US.UTF-8"
-
-
-# Results per page
-ResultsPerPage = 10
-
-
-# Length of "probably unique id" 's
-# Those id's are used to characterize every document, thus allowing tiny URLs in Controllers
-#  HashLength = 10
-#  Document.new("whatever.pdf").probably_unique_id => "bbuxhynait"
-#  HashLength = 20
-#  Document.new("whatever.pdf").probably_unique_id => "jfzjkyfkfkbbuxhynait"
-# The more documents you have, the bigger HashLength should be in order to avoid collisions.
-# It would not be wise (and specs won't pass) to specify HashLength smaller than 10.
-HashLength = 10
-
-
-# Specify the default Levenshtein distance when using FuzzyQuery
-# see http://ferret.davebalmain.com/api/classes/Ferret/QueryParser.html for more information.
-Ferret::Search::FuzzyQuery.default_min_similarity=0.6
+module Picolena
+  # Specify indexes path.
+  # Storage should be sufficient in order to store all indexed data.
+  IndexesSavePath=File.join(RAILS_ROOT, 'tmp/ferret_indexes/')
+  
+  
+  # Which language should be used?
+  # English (:en), German (:de), French (:fr) and Spanish (:es) are currently supported
+  # English is chosen by default.
+  # If you'd like to use another language, you can find templates in #{RAILS_ROOT}/lang/ui,
+  # then add your own language in this directory, and modify this line:
+  Globalite.language = :en
+  
+  
+  # Is more than one language used in indexed documents?
+  # Picolena can try to recognise the language used, and save it in the index.
+  # It is then possible to look for documents according to their language.
+  #
+  # If every document is written in the same language, turning UseLanguageRecognition to false
+  # will speed up the indexing process
+  UseLanguageRecognition = true
+  
+  # Specify which locale should be used by Ferret
+  Ferret.locale = "en_US.UTF-8"
+  
+  
+  # Results per page
+  ResultsPerPage = 10
+  
+  
+  # Length of "probably unique id" 's
+  # Those id's are used to characterize every document, thus allowing tiny URLs in Controllers
+  #  HashLength = 10
+  #  Document.new("whatever.pdf").probably_unique_id => "bbuxhynait"
+  #  HashLength = 20
+  #  Document.new("whatever.pdf").probably_unique_id => "jfzjkyfkfkbbuxhynait"
+  # The more documents you have, the bigger HashLength should be in order to avoid collisions.
+  # It would not be wise (and specs won't pass) to specify HashLength smaller than 10.
+  HashLength = 10
+  
+  
+  # Specify the default Levenshtein distance when using FuzzyQuery
+  # see http://ferret.davebalmain.com/api/classes/Ferret/QueryParser.html for more information.
+  Ferret::Search::FuzzyQuery.default_min_similarity=0.6
+  Analyzer=Ferret::Analysis::StandardAnalyzer.new
+end

data/lib/picolena/config/icons_and_filetypes.yml

@@ -0,0 +1,69 @@
+video:
+  avi
+  wmv
+  mpg
+  mpeg
+ogg:
+  mp3
+  ogg
+  wma
+  wav
+  wmv
+  tee
+txt:
+  txt
+  text
+  tex
+  bib
+  log
+  ini
+  no_extension
+doc:
+  doc
+  odt
+  rtf
+  dot
+  docx
+  dotx
+insel:
+  ins
+  vee
+ppt:
+  ppt
+  pps
+  pptx
+  odp
+pdf:
+  pdf
+package:
+  gz
+  rar
+  zip
+  bak
+code:
+  for
+  cpp
+  c
+  rb
+  java
+html:
+  html
+  htm
+xls:
+  xls
+  xlsx
+  ods
+picture:
+  psd
+  jpg
+  png
+  gif
+  eps
+  bmp
+  ico
+cad:
+  dwg
+  dxf
+exe:
+  exe
+  dll

data/lib/picolena/config/indexed_directories.yml

@@ -5,6 +5,6 @@ development:
   #alias path could be any smb, http, ftp or local directory that is available to the end-user.
   <%= directories_to_index %>
 test:
-  "spec/test_dirs/indexed": "http://picolena.devjavu.com/browser/trunk/spec/test_dirs/indexed"
+  "spec/test_dirs/indexed": "http://picolena.devjavu.com/browser/trunk/lib/picolena/templates/spec/test_dirs/indexed"
 production:
   <%= directories_to_index %>

data/lib/picolena/picolena_generator.rb

@@ -62,6 +62,7 @@ class PicolenaGenerator < RubiGen::Base #:nodoc:
       m.file '../config/basic.rb', 'config/custom/picolena.rb'
       m.template '../config/indexed_directories.yml', 'config/custom/indexed_directories.yml', :assigns => {:directories_to_index => @directories_to_index}
       m.template '../config/title_and_names_and_links.yml', 'config/custom/title_and_names_and_links.yml', :assigns => {:version => Picolena::VERSION::STRING}
+      m.file '../config/icons_and_filetypes.yml', 'config/custom/icons_and_filetypes.yml'
 
       # README, License & Rakefile
       m.file 'MIT-LICENSE', 'LICENSE'
@@ -135,7 +136,7 @@ EOS
     doc
     lang/ui
     lib
-    lib/filters
+    lib/plain_text_extractors
     lib/tasks
     log
     public
@@ -155,6 +156,7 @@ EOS
     spec/test_dirs/indexed/basic
     spec/test_dirs/indexed/different_encodings
     spec/test_dirs/indexed/just_one_doc
+    spec/test_dirs/indexed/lang
     spec/test_dirs/indexed/literature
     spec/test_dirs/indexed/others
     spec/test_dirs/indexed/others/nested

data/lib/picolena/templates/app/controllers/application.rb

@@ -23,9 +23,9 @@ class ApplicationController < ActionController::Base
   # Tries to match remote IP address with the white list defined in config/custom/white_list_ip.yml
   # Redirects to :access_denied if the remote IP is not white listed.
   def should_only_be_available_for_white_list_IPs
-    unless request.remote_ip =~ WhiteListIPs
+    unless request.remote_ip =~ Picolena::WhiteListIPs
       redirect_to :controller => 'application', :action=>'access_denied'
       return false
     end
   end
-end
+end

data/lib/picolena/templates/app/controllers/documents_controller.rb

@@ -24,7 +24,7 @@ class DocumentsController < ApplicationController
       page=params[:page]||1
       finder=Finder.new(@query,page)
       finder.execute!
-      pager=::Paginator.new(finder.total_hits, ResultsPerPage) do
+      pager=::Paginator.new(finder.total_hits, Picolena::ResultsPerPage) do
         finder.matching_documents
       end
       @matching_documents=pager.page(page)

data/lib/picolena/templates/app/helpers/documents_helper.rb

@@ -3,13 +3,13 @@ module DocumentsHelper
   def nothing_found?
     @matching_documents.nil? or @matching_documents.entries.empty?
   end
-
+  
   # Very basic pagination.
   # Provides liks to Next, Prev and FirstPage when needed.
   def should_paginate(page,query)
-      [(link_to("←←", :action => :show, :id => query, :page => 1) if page.number>2),
-      (link_to("←", :action => :show, :id => query, :page => page.prev.number) if page.prev?),
-      (link_to("→", :action => :show, :id => query, :page => page.next.number) if page.next?)].compact.join(" | ")
+    [(link_to("←←", :action => :show, :id => query, :page => 1) if page.number>2),
+     (link_to("←", :action => :show, :id => query, :page => page.prev.number) if page.prev?),
+     (link_to("→", :action => :show, :id => query, :page => page.next.number) if page.next?)].compact.join(" | ")
   end
   
   # Returns a localized sentence like "Results 1-10 of 12 for Zimbabwe (0.472s)" or
@@ -30,7 +30,7 @@ module DocumentsHelper
     content_tag(:small,'('<<number_with_precision(dt,3)<<'s)')
   end
   
-  # When possible, highlights content of the document that match the query.
+  # When possible, highlights content of the document that matches the query.
   def highlight_matching_content(document)
     content_tag(:ul,document.matching_content.collect{|sentence|
       content_tag(:li,h(sentence).gsub(/&lt;&lt;(.*?)&gt;&gt;/,'<strong>\1</strong>').gsub(/\v|\f/,''))
@@ -43,28 +43,9 @@ module DocumentsHelper
   end
   
   # Returns the location (if avaible) of the filetype icon.
-  # TODO: Move this hash to a .yml config file.
   def icon_for(filetype)    
-    pic_for_exts={
-      :xls=>%w{xls xlsx ods},
-      :doc=>%w{doc odt rtf dot docx dotx},
-      :pdf=>%w{pdf},
-      :txt=>%w{txt text tex bib log ini no_extension},
-      :ogg=>%w{mp3 ogg wma wav wmv tee},
-      :html=>%w{html htm},
-      :ppt=>%w{ppt pps pptx odp},
-      :package=>%w{gz rar zip bak},
-      :picture=>%w{psd jpg png gif eps bmp ico},
-      :cad=>%w{dwg dxf},
-      :exe=>%w{exe dll},
-      :video=>%w{avi wmv mpg mpeg},
-      :code=>%w{for cpp c rb java},
-      :insel=>%w{ins vee}
-    }
-    pic=pic_for_exts.find{|pic, extensions|
-      extensions.any? { |ext| filetype.sub(/\./,'').downcase==ext}
-    }
-    image_tag("icons/#{pic.first}.png") if pic
+    icon_symbol=FiletypeToIconSymbol[filetype.downcase.sub(/^\./,'')]
+    image_tag("icons/#{icon_symbol}.png") if icon_symbol
   end
   
   # Returns a link to a backup search engine that could maybe find more results for the same query.

data/lib/picolena/templates/app/models/document.rb

@@ -1,7 +1,8 @@
 # Document class retrieves information from filesystem and the index for any given document.
 class Document
   attr_reader :complete_path
-  attr_accessor :user, :score, :matching_content, :index_id
+  attr_writer :index_id
+  attr_accessor :user, :score, :matching_content
   
   def initialize(path)
     #To ensure @complete_path is an absolute direction.
@@ -10,8 +11,6 @@ class Document
     validate_in_indexed_directory
   end
   
-  alias_method :to_param, :id
-  
   #Delegating properties to File::method_name(complete_path)
   [:dirname, :basename, :extname, :size?, :file?, :read, :ext_as_sym].each{|method_name|
     define_method(method_name){File.send(method_name,complete_path)}
@@ -38,7 +37,7 @@ class Document
   #   "http://www.mycompany.com/wiki/organigram.odp"
   def alias_path
     original_dir=indexed_directory
-    alias_dir=IndexedDirectories[original_dir]
+    alias_dir=Picolena::IndexedDirectories[original_dir]
     dirname.sub(original_dir,alias_dir)
   end
   
@@ -50,48 +49,67 @@ class Document
     @probably_unique_id||=complete_path.base26_hash
   end
   
-  # Returns true iff some Filter has been defined to convert it to plain text.
+  # Returns true iff some PlainTextExtractor has been defined to convert it to plain text.
   #  Document.new("presentation.pdf").supported? => true
   #  Document.new("presentation.some_weird_extension").supported? => false
   def supported?
-    PlainText.supported_extensions.include?(self.ext_as_sym)
+    PlainTextExtractor.supported_extensions.include?(self.ext_as_sym)
   end
   
   # Retrieves content as it is *now*.
   def content
-    PlainText.extract_content_from(complete_path)
+    PlainTextExtractor.extract_content_from(complete_path)
   end
   
   # Cache à la Google.
   # Returns content as it was at the time it was indexed.
   def cached
-    get_index_id! unless index_id
-    Finder.index[index_id][:content]
+    from_index[:content]
   end
   
+  # FIXME: Not just date anymore.
   # Returns the last modification date before the document got indexed.
   # Useful to know how old a document is, and to which version the cache corresponds.
   def date
-    get_index_id! unless index_id
-    Finder.index[index_id][:date].sub(/(\d{4})(\d{2})(\d{2})/,'\1-\2-\3')
+    from_index[:date].sub(/(\d{4})(\d{2})(\d{2})(\d{2})(\d{2})(\d{2})/,'\1-\2-\3 \4:\5:\6')
+  end
+  
+  def mtime
+    from_index[:date].to_i
+  end
+  
+  # Returns language.
+  def lang
+    from_index[:lang]
+  end
+  
+  # Returns the id with which the document is indexed.
+  def index_id
+    @index_id ||= Document.find_by_complete_path(complete_path).index_id
   end
   
   private
   
-  def get_index_id!
-    @index_id = Document.find_by_unique_id(probably_unique_id).index_id
+  # Retrieves the document from the index.
+  # Useful to get meta-info about it.
+  def from_index
+    IndexReader.new[index_id]
   end
   
   def self.find_by_unique_id(some_id)
     Finder.new("probably_unique_id:"<<some_id).matching_document
   end
   
+  def self.find_by_complete_path(complete_path)
+    Finder.new('complete_path:"'<<complete_path<<'"').matching_document
+  end
+  
   def in_indexed_directory?
     !indexed_directory.nil?
   end
   
   def indexed_directory
-    IndexedDirectories.keys.find{|indexed_dir|
+    Picolena::IndexedDirectories.keys.find{|indexed_dir|
       dirname.starts_with?(indexed_dir)
     }    
   end

data/lib/picolena/templates/app/models/finder.rb

@@ -1,42 +1,36 @@
-require 'ff'
-
-class Finder
-  #FIXME: Should not use all those class methods to access index.
-  
+class Finder  
   attr_reader :query
   
-  def self.index
+  def index
     # caching index @@index ||=  
     # causes ferret-0.11.6/lib/ferret/index.rb:768: [BUG] Segmentation fault
-    Ferret::Index::Index.new(:path => IndexSavePath, :analyzer=>Analyzer)  
+    IndexReader.new
   end
   
-  def initialize(raw_query,page=1,results_per_page=ResultsPerPage)
-    query_parser = Ferret::QueryParser.new(:fields => [:content, :file, :basename, :filetype, :date], :or_default => false, :analyzer=>Analyzer)
-    @query = query_parser.parse(convert_to_english(raw_query))
+  def initialize(raw_query,page=1,results_per_page=Picolena::ResultsPerPage)
+    @query = Query.extract_from(raw_query)
     @raw_query= raw_query
-    Finder.ensure_that_index_exists_on_disk
+    IndexReader.ensure_existence
     @per_page=results_per_page
     @offset=(page.to_i-1)*results_per_page
-    validate_that_index_has_documents
+    index.should_have_documents
   end
   
   def execute!
     @matching_documents=[]
     start=Time.now
-    begin
-      top_docs=Finder.index.search(query, :limit => @per_page, :offset=>@offset)
-      top_docs.hits.each{|hit|
-        index_id,score=hit.doc,hit.score
-        begin
-          found_doc=Document.new(Finder.index[index_id][:complete_path])
-          found_doc.matching_content=Finder.index.highlight(query, index_id,
-                                                     :field => :content, :excerpt_length => 80,
-                                                     :pre_tag => "<<", :post_tag => ">>"
-          ) unless @raw_query=~/^\*+\.\w*$/
-          found_doc.score=score
-          found_doc.index_id=index_id
-          @matching_documents<<found_doc
+    top_docs=index.search(query, :limit => @per_page, :offset=>@offset)
+    top_docs.hits.each{|hit|
+      index_id,score=hit.doc,hit.score
+      begin
+        found_doc=Document.new(index[index_id][:complete_path])
+        found_doc.matching_content=index.highlight(query, index_id,
+                                                   :field => :content, :excerpt_length => 80,
+                                                   :pre_tag => "<<", :post_tag => ">>"
+        ) unless @raw_query=~/^\*+\.\w*$/
+        found_doc.score=score
+        found_doc.index_id=index_id
+        @matching_documents<<found_doc
         rescue Errno::ENOENT
           #"File has been moved/deleted!"
         end
@@ -44,9 +38,6 @@ class Finder
       @executed=true
       @time_needed=Time.now-start
       @total_hits=top_docs.total_hits
-    ensure
-      #index.close
-    end
   end
   
   # Returns true if it has been executed.
@@ -66,17 +57,7 @@ class Finder
     }
   }
    
-   # Returns true if index is existing.
-   def self.has_index?
-     index_filename and File.exists?(index_filename)
-   end
-   
-   # Returns true if there's at least one document indexed.
-   def has_documents?
-     Finder.index.size>0
-   end
-
-   # Returns matching document for any given query, if only
+   # Returns matching document for any given query only if
    # exactly one document is found.
    # Raises otherwise.
    def matching_document
@@ -89,42 +70,4 @@ class Finder
        raise IndexError, "More than one document found"
      end
    end
-   
-   private
-   
-   # Convert query keywords to english so they can be parsed by Ferret.
-   def convert_to_english(query)
-     to_en={
-       /\b#{:AND.l}\b/=>'AND',
-       /\b#{:OR.l}\b/=>'OR',
-       /\b#{:NOT.l}\b/=>'NOT',
-       /(#{:filetype.l}):/=>'filetype:',
-       /#{:content.l}:/ => 'content:',
-       /#{:date.l}:/ => 'date:',
-       /\b#{:LIKE.l}\s+(\S+)/=>'\1~'
-     }
-     to_en.inject(query){|mem,non_english_to_english_keyword|
-       mem.gsub(*non_english_to_english_keyword)
-     }
-   end
-   
-   def self.index_filename
-     Dir.glob(File.join(IndexSavePath,'*.cfs')).first
-   end
-
-   def self.ensure_that_index_exists_on_disk
-     force_index_creation unless has_index? or RAILS_ENV=="production"
-   end
-   
-   def self.force_index_creation
-     create_index(IndexedDirectories.keys)
-   end
-   
-   def self.delete_index
-     FileUtils.rm(Dir.glob(File.join(IndexSavePath,'*.cfs'))) if has_index?
-   end
-   
-   def validate_that_index_has_documents
-     raise IndexError, "no document found" unless has_documents?
-   end 
-end
+end