ferret 0.1.0 → 0.1.1

data/README

@@ -87,10 +87,17 @@ only run on linux.
 
 == Contact
 
-Bug reports, patches, queries, discussion etc should be addressed to
-the mailing list. More information on the list can be found at:
+For bug reports and patches I have set up Trac here;
 
-http://ferret.davebalmain.com/
+  http://ferret.davebalmain.com/trac
+
+Queries, discussion etc should be addressed to the forum or mailing lists hosted
+at;
+
+  http://rubyforge.org/projects/ferret/
+
+Alternatively you could create a new page for discussion on the wiki at my Trac
+page above. Or, if you're shy, please feel free to email me directly at dbalmain@gmail.com
 
 Of course, since Ferret is almost a straight port of Java Lucene,
 everything said about Lucene at http://jakarta.apache.org/lucene/ should

data/Rakefile

@@ -115,6 +115,7 @@ PKG_FILES = FileList[
   'ext/**/*', 
   'lib/**/*.rb', 
   'test/**/*.rb',
+  'test/**/wordfile',
   'rake_utils/**/*.rb',
   'Rakefile'
 ]
@@ -170,7 +171,7 @@ else
 
     s.author = "David Balmain"
     s.email = "dbalmain@gmail.com"
-    s.homepage = "http://ferret.davebalmain.com"
+    s.homepage = "http://ferret.davebalmain.com/trac"
     s.rubyforge_project = "ferret"
 #     if ENV['CERT_DIR']
 #       s.signing_key = File.join(ENV['CERT_DIR'], 'gem-private_key.pem')

data/TUTORIAL

@@ -74,21 +74,21 @@ Index#search_each. The first method returns a Ferret::Index::TopDocs object.
 The second we'll show here. Lets say we wanted to find all documents with the
 phrase "quick brown fox" in the content field. We'd write;
 
-  index.search('content:"quick brown fox"') do |doc, score|
+  index.search_each('content:"quick brown fox"') do |doc, score|
     puts "Document #{doc} found with a score of #{score}"
   end
 
 But "fast" has a pretty similar meaning to "quick" and we don't mind if the
 fox is a little red. So we could expand our search like this;
 
-  index.search('content:"quick|fast brown|red fox"') do |doc, score|
+  index.search_each('content:"quick|fast brown|red fox"') do |doc, score|
     puts "Document #{doc} found with a score of #{score}"
   end
 
 What if we want to find all documents entered on or after 5th of September,
 2005 with the words "ruby" or "rails" in it. We could type something like;
 
-  index.search('date:( >= 20050905) content:(ruby OR rails)') do |doc, score|
+  index.search_each('date:( >= 20050905) content:(ruby OR rails)') do |doc, score|
     puts "Document #{doc} found with a score of #{score}"
   end
 

data/lib/ferret.rb

@@ -22,7 +22,7 @@
 #++
 # :include: ../TUTORIAL
 module Ferret
-  VERSION = '0.1.0'
+  VERSION = '0.1.1'
 end
 
 require 'ferret/utils'

data/lib/ferret/index/index.rb

@@ -6,6 +6,7 @@ module Ferret::Index
 
     def initialize(options = {})
       if options[:path]
+        options[:create_if_missing] = true if options[:create_if_missing].nil? 
         @dir = FSDirectory.new(options[:path], true)
         options[:close_dir] = true
       elsif options[:dir]

data/lib/ferret/index/index_writer.rb

@@ -94,11 +94,12 @@ module Index
           else
             begin
               @segment_infos.read(@directory)
-            rescue IOError => ioe
+            rescue Exception => e
               if options[:create_if_missing]
                 @segment_infos.write(@directory)
               else
-                raise ioe
+                @write_lock.release() # obtain write lock
+                raise e
               end
             end
           end 

data/lib/ferret/query_parser.rb

@@ -106,12 +106,12 @@ module Ferret
   #
   # Some examples;
   #
-  #   '+sport:ski -sport:snowboard sport:taboggen'
+  #   '+sport:ski -sport:snowboard sport:toboggen'
   #   '+ingredient:chocolate +ingredient:strawberries -ingredient:wheat'
   #
   # You may also use the boolean operators "AND", "&&", "OR" and "||". eg;
   #
-  #   'sport:ski AND NOT sport:snowboard OR sport:taboggen'
+  #   'sport:ski AND NOT sport:snowboard OR sport:toboggen'
   #   'ingredient:chocolate AND ingredient:strawberries AND NOT ingredient:wheat'
   #
   # You can set the default operator when you create the query parse.

data/test/unit/analysis/data/wordfile

@@ -0,0 +1,6 @@
+and
+to
+it
+the
+there
+their

data/test/unit/index/tc_index.rb

@@ -130,11 +130,36 @@ class IndexTest < Test::Unit::TestCase
 
   def test_fs_index
     fs_path = File.join(File.dirname(__FILE__), '../../temp/fsdir')
-    index = Index.new(:path => fs_path, :create => true, :default_field => "def_field")
+    `rm -rf #{fs_path}`
+    assert_raise(Errno::ENOENT) {Index.new(:path => fs_path, :create_if_missing => false, :default_field => "def_field")}
+    index = Index.new(:path => fs_path, :default_field => "def_field")
     do_test_index_with_array(index)
+    `rm -rf #{fs_path}`
     index = Index.new(:path => fs_path, :create => true, :default_field => "def_field")
     do_test_index_with_hash(index)
-    index = Index.new(:create => true, :default_field => "def_field")
+    index = Index.new(:path => fs_path, :create => true, :default_field => "def_field")
     do_test_index_with_doc_array(index)
   end
+
+  def test_fs_index_is_persistant
+    fs_path = File.join(File.dirname(__FILE__), '../../temp/fsdir')
+    `rm -rf #{fs_path}`
+    data = [
+      {"def_field" => "one two", :id => "me"},
+      {"def_field" => "one", :field2 => "three"},
+      {"def_field" => "two"},
+      {"def_field" => "one", :field2 => "four"},
+      {"def_field" => "one two"},
+      {"def_field" => "two", :field2 => "three", "field3" => "four"},
+      {"def_field" => "one"},
+      {"def_field" => "two", :field2 => "three", "field3" => "five"}
+    ]
+    index = Index.new(:path => fs_path, :default_field => "def_field")
+    data.each {|doc| index << doc }
+    assert_equal(8, index.size)
+    index.close
+    index = Index.new(:path => fs_path, :default_field => "def_field")
+    assert_equal(8, index.size)
+    assert_equal("four", index[5]["field3"])
+  end
 end

metadata

@@ -3,13 +3,13 @@ rubygems_version: 0.8.11
 specification_version: 1
 name: ferret
 version: !ruby/object:Gem::Version 
-  version: 0.1.0
-date: 2005-10-19 00:00:00 +09:00
+  version: 0.1.1
+date: 2005-10-22 00:00:00 +09:00
 summary: Ruby indexing library.
 require_paths: 
   - lib
 email: dbalmain@gmail.com
-homepage: http://ferret.davebalmain.com
+homepage: http://ferret.davebalmain.com/trac
 rubyforge_project: ferret
 description: Ferret is a port of the Java Lucene project. It is a powerful indexing and search library.
 autorequire: 
@@ -229,6 +229,7 @@ files:
   - test/longrunning/tm_store.rb
   - test/benchmark/tb_rw_vint.rb
   - test/benchmark/tb_ram_store.rb
+  - test/unit/analysis/data/wordfile
   - rake_utils/code_statistics.rb
 test_files: []
 rdoc_options: