ferret 0.10.4 → 0.10.5

data/lib/ferret_version.rb

@@ -1,3 +1,3 @@
 module Ferret
-  VERSION = '0.10.4'
+  VERSION = '0.10.5'
 end

data/test/unit/analysis/tc_analyzer.rb

@@ -246,7 +246,7 @@ class AsciiStandardAnalyzerTest < Test::Unit::TestCase
     assert_equal(Token.new('tnt', 86, 91), t2.next)
     assert_equal(Token.new('123-1235-asd-1234', 93, 110), t2.next)
     assert(! t2.next())
-    a = AsciiStandardAnalyzer.new(false)
+    a = AsciiStandardAnalyzer.new(ENGLISH_STOP_WORDS, false)
     t = a.token_stream("fieldname", input)
     t2 = a.token_stream("fieldname", input)
     assert_equal(Token.new('DBalmain@gmail.com', 0, 18), t.next)

data/test/unit/analysis/tc_token_stream.rb

@@ -486,7 +486,6 @@ class CustomTokenizerTest < Test::Unit::TestCase
     assert(! t.next())
     t = AsciiLowerCaseFilter.new(MyCSVTokenizer.new(input))
     assert_equal(Token.new("first field", 0, 11), t.next)
-    return
     assert_equal(Token.new("2nd field", 12, 21), t.next)
     assert_equal(Token.new("  p a d d e d  f i e l d  ", 22, 48), t.next)
     assert(! t.next())

data/test/unit/index/tc_index.rb

@@ -753,14 +753,14 @@ class IndexTest < Test::Unit::TestCase
                                  :excerpt_length => 10,
                                  :num_excerpts => 1)
     assert_equal(1, highlights.size)
-    assert_equal("...<b>for are one and two</b>...", highlights[0])
+    assert_equal("<b>the words</b>...", highlights[0])
 
     highlights = index.highlight('"the words" "for are one and two" ' +
                                  'words one two', 0,
-                                 :excerpt_length => 10,
+                                 :excerpt_length => 20,
                                  :num_excerpts => 2)
     assert_equal(2, highlights.size)
-    assert_equal("<b>the words</b>...", highlights[0])
+    assert_equal("<b>the words</b> we are...", highlights[0])
     assert_equal("...<b>for are one and two</b>...", highlights[1])
 
 

data/test/unit/index/tc_index_reader.rb

@@ -198,6 +198,11 @@ module IndexReaderCommon
 
     assert_equal(4, doc.size)
     [:author, :body, :title, :year].each {|fn| assert(doc.keys.include?(fn))}
+    assert_equal([@ir[0].load, @ir[1].load, @ir[2].load], @ir[0, 3].collect {|d| d.load})
+    assert_equal([@ir[61].load, @ir[62].load, @ir[63].load], @ir[61, 100].collect {|d| d.load})
+    assert_equal([@ir[0].load, @ir[1].load, @ir[2].load], @ir[0..2].collect {|d| d.load})
+    assert_equal([@ir[61].load, @ir[62].load, @ir[63].load], @ir[61..100].collect {|d| d.load})
+    assert_equal(@ir[-60], @ir[4])
   end
 
   def test_ir_norms()

data/test/unit/search/tc_filter.rb

@@ -98,4 +98,19 @@ class FilterTest < Test::Unit::TestCase
     rf2 = RangeFilter.new(:int, :>= => "3")
     do_test_top_docs(searcher, query, [4,6], rf2)
   end
+
+  class CustomFilter
+    def bits(ir)
+      bv = Ferret::Utils::BitVector.new
+      bv[0] = bv[2] = bv[4] = true
+      bv
+    end
+  end
+
+  def test_custom_filter
+    searcher = Searcher.new(@dir)
+    q = MatchAllQuery.new
+    filt = CustomFilter.new
+    do_test_top_docs(searcher, q, [0, 2, 4], filt)
+  end
 end

data/test/unit/search/tm_searcher.rb

@@ -23,6 +23,8 @@ module SearcherTests
     assert_equal(10, top_docs.hits.size)
     top_docs = @searcher.search(tq, {:limit => 20})
     assert_equal(@searcher.max_doc, top_docs.hits.size)
+
+    assert_equal([Ferret::Term.new(:field, "word1")], tq.terms(@searcher))
   end
 
   def check_docs(query, options, expected=[])
@@ -79,6 +81,12 @@ module SearcherTests
     assert_equal("field:<brown|word1|word2|fast^50.0>^80.1", mtq.to_s())
     mtq << "word3"
     assert_equal("field:<brown|word1|word2|fast^50.0>^80.1", mtq.to_s())
+    
+    terms = mtq.terms(@searcher)
+    assert(terms.index(Ferret::Term.new(:field, "brown")))
+    assert(terms.index(Ferret::Term.new(:field, "word1")))
+    assert(terms.index(Ferret::Term.new(:field, "word2")))
+    assert(terms.index(Ferret::Term.new(:field, "fast")))
   end
 
   def test_boolean_query
@@ -199,6 +207,9 @@ module SearcherTests
 
     wq = WildcardQuery.new(:category, "cat1*/su??ub2")
     check_hits(wq, [4, 16])
+
+    wq = WildcardQuery.new(:category, "*/sub2*")
+    check_hits(wq, [3, 4, 13, 15])
   end
 
   def test_multi_phrase_query()
@@ -327,13 +338,13 @@ module SearcherTests
                                     :excerpt_length => 10,
                                     :num_excerpts => 1)
     assert_equal(1, highlights.size)
-    assert_equal("...<b>for are one and two</b>...", highlights[0])
+    assert_equal("<b>the words</b>...", highlights[0])
 
     highlights = searcher.highlight(q, 0, :field,
                                     :excerpt_length => 10,
                                     :num_excerpts => 2)
     assert_equal(2, highlights.size)
     assert_equal("<b>the words</b>...", highlights[0])
-    assert_equal("...<b>for are one and two</b>...", highlights[1])
+    assert_equal("...<b>one</b> <b>two</b>...", highlights[1])
   end
 end

metadata

@@ -3,8 +3,8 @@ rubygems_version: 0.8.11
 specification_version: 1
 name: ferret
 version: !ruby/object:Gem::Version 
-  version: 0.10.4
-date: 2006-09-09 00:00:00 +09:00
+  version: 0.10.5
+date: 2006-09-19 00:00:00 +09:00
 summary: Ruby indexing library.
 require_paths: 
 - lib