name-spotter 0.1.8 → 0.1.9

data/VERSION

@@ -1 +1 @@
-0.1.8
+0.1.9

data/lib/name-spotter/taxon_finder_client.rb

@@ -1,11 +1,8 @@
+# encoding: utf-8
 class NameSpotter
   class TaxonFinderClient < NameSpotter::Client
     def initialize(opts = { host: "0.0.0.0", port: "1234" })
       super
-      # We keep track of the document to get accurate offsets.
-      # Other methods such as keeping track of the character number
-      #  didn't work so well due to the nature of TaxonFinder.
-      @document = ""
     end
 
     def find(str, from_web_form=false)
@@ -17,20 +14,21 @@ class NameSpotter
       @current_string = ''
       @current_string_state = ''
       @word_list_matches = 0
-      @empty_count = 0
+      @cursor = 5.times.inject([]) { |res| res << ['',-1] }
+      @current_index = nil
       words = str.split(/\s/)
       words.each do |word|
-        # Since we split on whitespace, this addition of a " " char
-        # allows us to keep the document accurate and is basically
-        # replacing all \s matches with " "
-        @document << word + " "
-        unless word.empty?
+        if word.empty?
+          @cursor[-1][0] << " "
+        else
+          cursor_entry = [word, 1 + @cursor[-1][0].size + @cursor[-1][-1]]
+          @cursor.shift
+          @cursor << cursor_entry
           taxon_find(word)
         end
       end
       socket.close
       @socket = nil
-      @document = ""
       @names
     end
     
@@ -64,6 +62,9 @@ class NameSpotter
       @current_string_state = current_string_state
       @word_list_matches = word_list_matches
       @return_score = return_score
+      if @current_string.size > 0 && !@current_index
+        @current_index = @cursor[-1][-1]
+      end
       if not return_string.blank? or not return_string_2.blank?
         OpenStruct.new( { :current_string       => current_string,
                        :current_string_state => current_string_state,
@@ -73,16 +74,22 @@ class NameSpotter
                        :return_string_2      => return_string_2,
                        :return_score_2       => return_score_2 })
       else
+        @current_index = nil if @current_string.empty? && @current_index
         false
       end
     end
 
     def process_response(str)
       str.force_encoding('utf-8')
-      verbatim_string = str.sub(/\[.*\]/, '.')
-      verbatim_regex = Regexp.new(verbatim_string.split(/\s/).join('\s+'), true)
-      start_position = @document.rindex(verbatim_regex)
-      verbatim_string = @document_verbatim[start_position..-1].match(verbatim_regex)[0] if start_position
+      start_position = verbatim_string = nil
+      if @current_index
+        start_position = @current_index
+        words, indices = @cursor.transpose
+        verbatim_string = words[indices.index(start_position)...-1].join(" ")
+        @current_index = nil
+      else
+        verbatim_string, start_position = @cursor[-1]
+      end
       scientific_string = str
       [verbatim_string, scientific_string, start_position]
     end

data/name-spotter.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = "name-spotter"
-  s.version = "0.1.8"
+  s.version = "0.1.9"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Anthony Goddard", "Chuck Ha", "Dmitry Mozzherin"]
-  s.date = "2012-06-06"
+  s.date = "2012-06-12"
   s.description = "The gem searches for scientific names in texts using socket servers running TaxonFinder (by Patrick Leary) and NetiNeti (by Lakshmi Manohar Akella)"
   s.email = "dmozzherin@gmail.com"
   s.extra_rdoc_files = [

data/spec/name-spotter_spec.rb

@@ -39,7 +39,7 @@ describe "NameSpotter" do
   end
 
   it "should be able to find scientific names in text" do
-    text = "Some text that has Betula\n alba and Mus musculus and \neven B. alba and even M. mus-\nculus and unicoded name Aranea röselii. Also it has name unknown before: Varanus bitatawa species"
+    text = "Some text   that has Betula\n alba and Mus musculus and \neven B. alba and even M. mus-\nculus and unicoded name Aranea röselii. Also it has name unknown before: Varanus bitatawa species"
     res = @neti.find(text)[:names].map { |n| n[:scientificName] } 
     res.should == ["Betula alba", "Mus musculus", "B. alba", "Aranea röselii", "Varanus bitatawa"]
     tf_res = @tf.find(text)
@@ -72,11 +72,11 @@ describe "NameSpotter" do
   end
 
   it "should be able to return offsets for all names found by taxonfinder" do
-    text = "We have to be sure that Betula\n alba and PSEUDOSCORPIONIDA and Aranea röselii and capitalized ARANEA RÖSELII and Pardosa\n moesta f. moesta Banks, 1892 all get their offsets"
+    text = "We have to be sure that Betula\n alba and PSEUDOSCORPIONIDA and ×Inkea which is not Passeriformes. We also have another hybrid Passiflora ×rosea and Aranea röselii and capitalized ARANEA RÖSELII and Pardosa\n moesta f. moesta Banks, 1892 all get their offsets"
     res = @neti.find(text)
-    res.should == {:names=>[{:verbatim=>"Betula\n alba", :scientificName=>"Betula alba", :offsetStart=>24, :offsetEnd=>35}, {:verbatim=>"Aranea röselii", :scientificName=>"Aranea röselii", :offsetStart=>63, :offsetEnd=>76}, {:verbatim=>"Pardosa\n moesta", :scientificName=>"Pardosa moesta", :offsetStart=>113, :offsetEnd=>127}]}
+    res.should == {:names=>[{:verbatim=>"Betula\n alba", :scientificName=>"Betula alba", :offsetStart=>24, :offsetEnd=>35}, {:verbatim=>"Passiflora ×rosea", :scientificName=>"Passiflora ×rosea", :offsetStart=>126, :offsetEnd=>142}, {:verbatim=>"Aranea röselii", :scientificName=>"Aranea röselii", :offsetStart=>148, :offsetEnd=>161}, {:verbatim=>"Pardosa\n moesta", :scientificName=>"Pardosa moesta", :offsetStart=>198, :offsetEnd=>212}]}
     tf_res = @tf.find(text)
-    tf_res.should == {:names=>[{:verbatim=>"Betula\n alba", :scientificName=>"Betula alba", :offsetStart=>24, :offsetEnd=>35}, {:verbatim=>"PSEUDOSCORPIONIDA", :scientificName=>"Pseudoscorpionida", :offsetStart=>41, :offsetEnd=>57}, {:verbatim=>"Aranea röselii", :scientificName=>"Aranea röselii", :offsetStart=>63, :offsetEnd=>76}, {:verbatim=>"ARANEA", :scientificName=>"Aranea", :offsetStart=>94, :offsetEnd=>99}, {:verbatim=>"Pardosa\n moesta f. moesta", :scientificName=>"Pardosa moesta f. moesta", :offsetStart=>113, :offsetEnd=>137}]}
+    tf_res.should == {:names=>[{:verbatim=>"Betula  alba", :scientificName=>"Betula alba", :offsetStart=>24, :offsetEnd=>35}, {:verbatim=>"PSEUDOSCORPIONIDA", :scientificName=>"Pseudoscorpionida", :offsetStart=>41, :offsetEnd=>57}, {:verbatim=>"Passeriformes.", :scientificName=>"Passeriformes", :offsetStart=>83, :offsetEnd=>96}, {:verbatim=>"Passiflora ×rosea", :scientificName=>"Passiflora rosea", :offsetStart=>126, :offsetEnd=>142}, {:verbatim=>"Aranea röselii", :scientificName=>"Aranea röselii", :offsetStart=>148, :offsetEnd=>161}, {:verbatim=>"ARANEA", :scientificName=>"Aranea", :offsetStart=>179, :offsetEnd=>184}, {:verbatim=>"Pardosa  moesta f. moesta", :scientificName=>"Pardosa moesta f. moesta", :offsetStart=>198, :offsetEnd=>222}]} 
   end
 
   it "should not make unsequential offsets on a page when using NetiNeti" do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: name-spotter
 version: !ruby/object:Gem::Version
-  version: 0.1.8
+  version: 0.1.9
   prerelease: 
 platform: ruby
 authors:
@@ -11,7 +11,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-06-06 00:00:00.000000000 Z
+date: 2012-06-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -269,7 +269,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 765055699781609627
+      hash: 3944184144538713044
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: