diff --git a/lib/rir.rb b/lib/rir.rb
index 0f336e0..0b27852 100644
--- a/lib/rir.rb
+++ b/lib/rir.rb
@@ -3,3 +3,5 @@
 require 'rir/document'
 require 'rir/string'
 require 'rir/query'
+require 'rir/corpus'
+require 'rir/regexp'
diff --git a/lib/rir/corpus.rb b/lib/rir/corpus.rb
index 44d2f3f..f443ec4 100644
--- a/lib/rir/corpus.rb
+++ b/lib/rir/corpus.rb
@@ -24,11 +24,17 @@ module RIR
     attr_accessor :path
 
     def initialize(path)
-      @path = path
+      @path = path.chomp "/"
     end
 
+    # Recursively outputs all files in +self.path+.
+    # WARNING ! This function may take a lot of time if many
+    # files are in subdirectories.
+    #
+    #   c = Corpus.new "my/path"
+    #   c.files                  # => ["README.txt", "lib/code.rb"]
     def files
-      Dir.glob("**/*.*")
+      Dir["#{@path}/**/*.*"]
     end
   end
 
diff --git a/lib/rir/document.rb b/lib/rir/document.rb
index 87a5c28..9bd05ae 100644
--- a/lib/rir/document.rb
+++ b/lib/rir/document.rb
@@ -117,5 +117,37 @@ module RIR
 
   # A WikipediaPage is a WebDocument.
   class WikipediaPage < WebDocument
+    require 'rexml/document'
+    require 'net/http'
+    require 'kconv'
+
+
+    def self.search_wikipedia_titles(name)
+      res = REXML::Document.new(Net::HTTP.get(URI.parse("http://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=#{URI.escape name}&format=xml")).toutf8).elements['api/query/search']
+
+      res.collect { |e| e.attributes['title'] } unless res.nil?
+    end
+
+    def self.get_url(name)
+      atts = REXML::Document.new(Net::HTTP.get(URI.parse("http://en.wikipedia.org/w/api.php?action=query&titles=#{URI.escape name}&inprop=url&prop=info&format=xml")).toutf8).elements['api/query/pages/page'].attributes
+
+      atts['fullurl'] if atts['missing'].nil?
+    end
+
+    def self.search_homepage(name)
+      title = WikipediaPage.search_wikipedia_titles name
+
+      begin
+        WikipediaPage.new(WikipediaPage.get_url title[0]) unless title.nil? || title.empty?
+      rescue
+        puts title[0]
+      end
+    end
+
+#    def initialize(name)
+#      title = WikipediaPage.search_wikipedia_titles name
+#      raise ArgumentError, "No page found" if title.empty? 
+#      super WikipediaPage.get_url title[0]
+#    end
   end
 end
diff --git a/lib/rir/query.rb b/lib/rir/query.rb
index 63ca4ca..581901e 100644
--- a/lib/rir/query.rb
+++ b/lib/rir/query.rb
@@ -61,7 +61,6 @@ module RIR
       attr_accessor :id, :query, :params, :rule
 
       def initialize(id,query,params)
-#        @params = Parameters === params ? params : Parameters.new(corpus)
         @params = params
         # Here we set the default retrieval model as Language Modeling
         # with a Dirichlet smoothing at 2500.
diff --git a/lib/rir/regexp.rb b/lib/rir/regexp.rb
new file mode 100644
index 0000000..dc718b9
--- /dev/null
+++ b/lib/rir/regexp.rb
@@ -0,0 +1,26 @@
+#!/usr/bin/env ruby
+
+# This file is a part of an Information Retrieval oriented Ruby library
+#
+# Copyright (C) 2010-2011 Romain Deveaud <romain.deveaud@gmail.com>
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+class Regexp
+
+  def negated
+    /^((?!#{self}).)*$/
+  end
+
+end
diff --git a/lib/rir/string.rb b/lib/rir/string.rb
index cbf4c23..7a95f5c 100644
--- a/lib/rir/string.rb
+++ b/lib/rir/string.rb
@@ -138,6 +138,23 @@ class String
     dup.strip_stylesheets!
   end
 
+  # Removes punctuation from +self+.
+  #
+  #   s = "hello, world. how are you?!"
+  #   s.strip_punctuation!
+  #   s                                 # => "hello world how are you"
+  def strip_punctuation!
+    replace strip_with_pattern /[^a-zA-Z0-9\-\s]/
+  end
+
+  # Removes punctuation from +self+.
+  #
+  #   s = "hello, world. how are you?!"
+  #   s.strip_punctuation               # => "hello world how are you"
+  def strip_punctuation
+    dup.strip_punctuation!
+  end
+
   # Returns the text values inside all occurences of a XML tag in +self+
   #
   #   s = "four-piece in <a href='#'>Indianapolis</a>, <a href='#'>Indiana</a> at the Murat Theatre"
diff --git a/main.rb b/main.rb
index 0ea2e47..f79f17c 100644
--- a/main.rb
+++ b/main.rb
@@ -6,6 +6,8 @@ w = RIR::WikipediaPage.new("http://en.wikipedia.org/wiki/The_Dillinger_Escape_Pl
 p w.entropy("guitar")
 
 params = RIR::Indri::Parameters.new("path_vers_mon_index")
-p params.rule
 q = RIR::Indri::IndriQuery.new("pouet", "bla", params)
 puts q
+
+c = RIR::Corpus.new "/home/romain/INEX/BookTrack/corpus/"
+puts c.files.size
diff --git a/test/string_test.rb b/test/string_test.rb
index 55bbaa4..b01b9be 100644
--- a/test/string_test.rb
+++ b/test/string_test.rb
@@ -21,4 +21,8 @@ class TestString < Test::Unit::TestCase
   def test_strip_xml
     assert_equal("testme", "<test>testme</test>".strip_xml_tags)
   end
+
+  def test_strip_punctuation
+    assert_equal("test test test test   test test", "test, test. .test, ;test !! ? test ...test./".strip_punctuation)
+  end
 end