taxamatch_rb 0.6.3 → 0.6.4

data/lib/taxamatch_rb.rb

@@ -7,7 +7,6 @@ require 'taxamatch_rb/atomizer'
 require 'taxamatch_rb/normalizer'
 require 'taxamatch_rb/phonetizer'
 require 'taxamatch_rb/authmatch'
-require 'ruby-debug'
 
 $KCODE='u' if RUBY_VERSION.split('.')[1].to_i < 9  
 
@@ -66,27 +65,29 @@ module Taxamatch
     def match_genera(genus1, genus2)
       genus1_length = genus1[:normalized].size
       genus2_length = genus2[:normalized].size
+      min_length = [genus1_length, genus2_length].min
       match = false
       ed = @dlm.distance(genus1[:normalized], genus2[:normalized],1,3) #TODO put block = 2
-      return {'edit_distance' => ed, 'phonetic_match' => false, 'match' => false} if ed/[genus1_length, genus2_length].min > 0.2
+      return {'edit_distance' => ed, 'phonetic_match' => false, 'match' => false} if ed/min_length.to_f > 0.2
       return {'edit_distance' => ed, 'phonetic_match' => true, 'match' => true} if genus1[:phonetized] == genus2[:phonetized] 
     
-      match = true if ed <= 3 && ([genus1_length, genus2_length].min > ed * 2) && (ed < 2 || genus1[0] == genus2[0])
+      match = true if ed <= 3 && (min_length > ed * 2) && (ed < 2 || genus1[0] == genus2[0])
       {'edit_distance' => ed, 'match' => match, 'phonetic_match' => false} 
     end
 
     def match_species(sp1, sp2)
       sp1_length = sp1[:normalized].size
       sp2_length = sp2[:normalized].size
+      min_length = [sp1_length, sp2_length].min
       sp1[:phonetized] = Taxamatch::Phonetizer.normalize_ending sp1[:phonetized]
       sp2[:phonetized] = Taxamatch::Phonetizer.normalize_ending sp2[:phonetized]
       match = false
       ed = @dlm.distance(sp1[:normalized], sp2[:normalized], 1, 4) #TODO put block 4
-      return {'edit_distance' => ed, 'phonetic_match' => false, 'match' => false} if ed/[sp1_length, sp2_length].min > 0.3334
+      return {'edit_distance' => ed, 'phonetic_match' => false, 'match' => false} if ed/min_length.to_f > 0.3334
       #puts 's: %s, %s, %s' % [sp1[:normalized], sp2[:normalized], ed]
       return {'edit_distance' => ed, 'phonetic_match' => true, 'match' => true} if sp1[:phonetized] == sp2[:phonetized]
     
-      match = true if ed <= 4 && ([sp1_length, sp2_length].min >= ed * 2) && (ed < 2 || sp1[:normalized][0] == sp2[:normalized][0]) && (ed < 4 || sp1[:normalized][0...3] == sp2[:normalized][0...3])
+      match = true if ed <= 4 && (min_length >= ed * 2) && (ed < 2 || sp1[:normalized][0] == sp2[:normalized][0]) && (ed < 4 || sp1[:normalized][0...3] == sp2[:normalized][0...3])
       { 'edit_distance' => ed, 'match' => match, 'phonetic_match' => false}
     end
   

data/spec/taxamatch_rb_spec.rb

@@ -59,7 +59,7 @@ describe 'Taxamatch::Base' do
       if y
         y[2] = y[2] == 'true' ? true : false
         res = @tm.taxamatch(y[0], y[1], false)
-        #puts "%s, %s, %s, %s" % [y[0], y[1], y[2], y[3]] 
+        puts "%s, %s, %s, %s" % [y[0], y[1], y[2], y[3]] 
         res['match'].should == y[2]
         res['edit_distance'].should == y[3].to_i
       end
@@ -89,16 +89,16 @@ describe 'Taxamatch::Base' do
     g2 = make_taxamatch_hash 'Pantheri'
     @tm.match_genera(g1, g2).should == {'phonetic_match' => false, 'edit_distance' => 1, 'match' => true}
     #phonetic match tramps everything
-    g1 = make_taxamatch_hash 'Xantheriiiiiiiiiiiiiii'
-    g2 = make_taxamatch_hash 'Zanthery'
+    g1 = make_taxamatch_hash 'Xaaaaantheriiiiiiiiiiiiiii'
+    g2 = make_taxamatch_hash 'Zaaaaaaaaaaaantheryyyyyyyy'
     @tm.match_genera(g1, g2).should == {'phonetic_match' => true, 'edit_distance' => 4, 'match' => true}
     #same first letter and distance 2 should match
-    g1 = make_taxamatch_hash 'Xantherii'
-    g2 = make_taxamatch_hash 'Xantherrr'
+    g1 = make_taxamatch_hash 'Xaaaantherii'
+    g2 = make_taxamatch_hash 'Xaaaantherrr'
     @tm.match_genera(g1, g2).should == {'phonetic_match' => false, 'match' => true, 'edit_distance' => 2}
     #First letter is the same and distance is 3 should match, no phonetic match
-    g1 = make_taxamatch_hash 'Xantheriii'
-    g2 = make_taxamatch_hash 'Xantherrrr'
+    g1 = make_taxamatch_hash 'Xaaaaaaaaaaantheriii'
+    g2 = make_taxamatch_hash 'Xaaaaaaaaaaantherrrr'
     @tm.match_genera(g1, g2).should == {'phonetic_match' => false, 'match' => true, 'edit_distance' => 3}
     #Should not match if one of words is shorter than 2x edit distance and distance is 2 or 3
     g1 = make_taxamatch_hash 'Xant'
@@ -116,24 +116,24 @@ describe 'Taxamatch::Base' do
     s2 = make_taxamatch_hash 'major'
     @tm.match_species(s1, s2).should ==  {'phonetic_match' => true, 'match' => true, 'edit_distance' => 0}
     #Phonetic match always works
-    s1 = make_taxamatch_hash 'xanteriiiiiiii'
-    s2 = make_taxamatch_hash 'zantereeeeeeee'
-    @tm.match_species(s1, s2).should ==  {'phonetic_match' => true, 'match' => true, 'edit_distance' => 5}
+    s1 = make_taxamatch_hash 'xanteriiieeeeeeeeeeeee'
+    s2 = make_taxamatch_hash 'zantereeeeeeeeeeeeeeee'
+    @tm.match_species(s1, s2).should ==  {'phonetic_match' => true, 'match' => true, 'edit_distance' => 4}
     #Phonetic match works with different endings
     s1 = make_taxamatch_hash 'majorum'
     s2 = make_taxamatch_hash 'majoris'
     @tm.match_species(s1, s2).should ==  {'phonetic_match' => true, 'match' => true, 'edit_distance' => 2}
     #Distance 4 matches if first 3 chars are the same
-    s1 = make_taxamatch_hash 'majorrrrr'
-    s2 = make_taxamatch_hash 'majoraaaa'
+    s1 = make_taxamatch_hash 'majjjjorrrrr'
+    s2 = make_taxamatch_hash 'majjjjoraaaa'
     @tm.match_species(s1, s2).should == {'phonetic_match' => false, 'match' => true, 'edit_distance' => 4}
     #Should not match if Distance 4 matches and first 3 chars are not the same
     s1 = make_taxamatch_hash 'majorrrrr'
     s2 = make_taxamatch_hash 'marorraaa'
     @tm.match_species(s1, s2).should == {'phonetic_match' => false, 'match' => false, 'edit_distance' => 4}
     #Distance 2 or 3 matches if first 1 char is the same
-    s1 = make_taxamatch_hash 'morrrr'
-    s2 = make_taxamatch_hash 'moraaa'
+    s1 = make_taxamatch_hash 'moooorrrr'
+    s2 = make_taxamatch_hash 'mooooraaa'
     @tm.match_species(s1, s2).should == {'phonetic_match' => false, 'match' => true, 'edit_distance' => 3}
     #Should not match if Distance 2 or 3 and first 1 char is not the same
     s1 = make_taxamatch_hash 'morrrr'

data/spec/taxamatch_test.txt

@@ -9,6 +9,7 @@ Pomatomus|Pomatomas|true|1
 Pomatomus L.|Pomatomas Linn.|true|1
 Pomatomus Ber|Pomatomas Linn|false|1
 Pomatomus L. 1753|Pomatomus Linn. 1800|false|0
+Patella|Abbella|false|3
 
 ## additional authorship should match
 Puma concolor|Puma concolor L.|true|0
@@ -17,7 +18,7 @@ Puma concolor|Puma concolor L.|true|0
 Puma concolor|Puma cancolor|true|1
 #
 Pomatomus saltatrix|Pomatomus saltratix|true|2
-Pomatomus saltator|Pomatomus saltatrix|true|3
+Pomatomus saltator|Pomatomus saltatrix|false|3 #!!!
 #
 Loligo pealeii|Loligo plei|false|3
 #

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 6
-  - 3
-  version: 0.6.3
+  - 4
+  version: 0.6.4
 platform: ruby
 authors: 
 - Dmitry Mozzherin
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-03-30 00:00:00 -04:00
+date: 2010-04-08 00:00:00 -04:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency