Changeset a249f609ee1784b37b8bce802dfa88406cc2c3fd

Timestamp:

06/22/08 15:15:45 (7 months ago)

Author:

David Balmain <dbalmain@…>

Parents:

97564968f9aed076ddedd60a97b504b972e3f7fc

Children:

ff6f080b9c3c3ca054c9bbce77571f813aee457a

git-committer:

David Balmain <dbalmain@gmail.com> / 2008-06-22T15:15:45Z+1000

Message:

Added term frequency field to TVTerm

* You can now found out the frequency of a term in a document without having to

store offsets or positions.

Location:

ruby

Files:

• ext/r_index.c (modified) (2 diffs)
• lib/ferret/index.rb (modified) (1 diff)
• test/unit/index/tc_index_reader.rb (modified) (6 diffs)

ruby/ext/r_index.c

@@ -1228 +1228 @@
         RARRAY(rpositions)->len = freq;
     }
-    return rb_struct_new(cTVTerm, rtext, rpositions, NULL);
+    return rb_struct_new(cTVTerm, rtext, INT2FIX(freq), rpositions, NULL);
 }
 
@@ -3073 +3073 @@
     cTVTerm = rb_define_class_under(cTermVector, "TVTerm", rb_cObject);
     */
-    cTVTerm = rb_struct_define(tv_term_class, "text", "positions", NULL);
+    cTVTerm = rb_struct_define(tv_term_class, "text", "freq", "positions", NULL);
     rb_set_class_path(cTVTerm, cTermVector, tv_term_class);
     rb_const_set(mIndex, rb_intern(tv_term_class), cTVTerm);

ruby/lib/ferret/index.rb

@@ -491 +491 @@
           term_doc_enum = @reader.term_docs_for(@id_field, id.to_s)
           if term_doc_enum.next?
-            id = @reader[term_doc_enum.doc]
+            id = term_doc_enum.doc
           else
             return nil

ruby/test/unit/index/tc_index_reader.rb

@@ -192 +192 @@
     expected_tv = TermVector.new(:body,
       [
-        TVTerm.new("word1", [2, 4, 7]),
-        TVTerm.new("word2", [3]),
-        TVTerm.new("word3", [0, 5, 8, 9]),
-        TVTerm.new("word4", [1, 6])
+        TVTerm.new("word1", 3, [2, 4, 7]),
+        TVTerm.new("word2", 1, [3]),
+        TVTerm.new("word3", 4, [0, 5, 8, 9]),
+        TVTerm.new("word4", 2, [1, 6])
       ],
       [*(0...10)].collect {|i| TVOffsets.new(i*6, (i+1)*6 - 1)})
@@ -210 +210 @@
     tv = tvs[:author]
     assert_equal(:author, tv.field)
-    assert_equal([TVTerm.new("Leo", [0]), TVTerm.new("Tolstoy", [1])], tv.terms)
+    assert_equal([TVTerm.new("Leo", 1, [0]), TVTerm.new("Tolstoy", 1, [1])], tv.terms)
     assert(tv.offsets.nil?)
 
@@ -216 +216 @@
     tv = tvs[:title]
     assert_equal(:title, tv.field)
-    assert_equal([TVTerm.new("War And Peace", nil)], tv.terms)
+    assert_equal([TVTerm.new("War And Peace", 1, nil)], tv.terms)
     assert_equal([TVOffsets.new(0, 13)], tv.offsets)
   end
@@ -609 +609 @@
     expected_tv = TermVector.new(:body,
       [
-        TVTerm.new("word1", [2, 4, 7]),
-        TVTerm.new("word2", [3]),
-        TVTerm.new("word3", [0, 5, 8, 9]),
-        TVTerm.new("word4", [1, 6])
+        TVTerm.new("word1", 3, [2, 4, 7]),
+        TVTerm.new("word2", 1, [3]),
+        TVTerm.new("word3", 4, [0, 5, 8, 9]),
+        TVTerm.new("word4", 2, [1, 6])
       ],
       [*(0...10)].collect {|i| TVOffsets.new(i*6, (i+1)*6 - 1)})
@@ -627 +627 @@
     tv = tvs[:author]
     assert_equal(:author, tv.field)
-    assert_equal([TVTerm.new("Leo", [0]), TVTerm.new("Tolstoy", [1])], tv.terms)
+    assert_equal([TVTerm.new("Leo", 1, [0]), TVTerm.new("Tolstoy", 1, [1])], tv.terms)
     assert(tv.offsets.nil?)
 
@@ -633 +633 @@
     tv = tvs[:title]
     assert_equal(:title, tv.field)
-    assert_equal([TVTerm.new("War And Peace", nil)], tv.terms)
+    assert_equal([TVTerm.new("War And Peace", 1, nil)], tv.terms)
     assert_equal([TVOffsets.new(0, 13)], tv.offsets)
   end