anystyle-parser 0.6.2 → 0.6.3

data/lib/anystyle/parser/version.rb

@@ -1,5 +1,5 @@
 module Anystyle
   module Parser
-    VERSION = '0.6.2'.freeze
+    VERSION = '0.6.3'.freeze
   end
 end

data/spec/anystyle/parser/normalizer_spec.rb

@@ -1,3 +1,5 @@
+# -*- encoding: utf-8 -*-
+
 module Anystyle
   module Parser
 
@@ -14,13 +16,13 @@ module Anystyle
           Normalizer.instance.normalize_names('A, B').should == 'A, B.'
         end
 
-        # it "tokenizes 'A, jr., B'" do
-        #   Normalizer.instance.normalize_names('A, jr., B').should == 'A, jr., B'
-        # end
-        #
-        # it "tokenizes 'A, B, jr.'" do
-        #   Normalizer.instance.normalize_names('A, B, jr.').should == 'A, B, jr.'
-        # end
+        it "tokenizes 'A, jr., Bbb'" do
+          Normalizer.instance.normalize_names('A, jr., B').should == 'A, jr., B.'
+        end
+
+        it "tokenizes 'A, B, jr.'" do
+          Normalizer.instance.normalize_names('A, B, jr.').should == 'A, jr., B.'
+        end
 
         it "tokenizes 'A, B, C, D'" do
           Normalizer.instance.normalize_names('A, B, C, D').should == 'A, B. and C, D.'
@@ -34,8 +36,13 @@ module Anystyle
           Normalizer.instance.normalize_names('Aa Bb, C.').should == 'Aa Bb, C.'
         end
 
+        it "tokenizes 'Plath, L.C., Asgaard, G., ... Botros, N.'" do
+          Normalizer.instance.normalize_names('Plath, L.C., Asgaard, G., ... Botros, N.').should == 'Plath, L.C. and Asgaard, G. and Botros, N.'
+          Normalizer.instance.normalize_names('Plath, L.C., Asgaard, G., … Botros, N.').should == 'Plath, L.C. and Asgaard, G. and Botros, N.'
+        end
+
         it "tokenizes 'Aa Bb, Cc Dd, and E F G'" do
-          Normalizer.instance.normalize_names('Aa Bb, C D, and E F G').should == 'Bb, Aa and D, C. and G, E. F.'
+          Normalizer.instance.normalize_names('Aa Bb, Cc Dd, and E F G').should == 'Bb, Aa and Dd, Cc and G, E. F.'
         end
 
         [

data/spec/anystyle/parser/parser_spec.rb

@@ -63,6 +63,9 @@ module Anystyle::Parser
           subject.prepare('hello, world!', true)[0].map { |t| t[/\S+$/] }.should == %w{ unknown unknown }
         end
 
+        it 'converts xml entitites' do
+          subject.prepare("<note>&gt;&gt; &amp; foo</note>", true)[0].map { |t| t[/\S+/] }.should == %w{ >> & foo }
+        end
       end
     end
 

data/spec/fixtures/train_dps.txt

@@ -1,9 +1,9 @@
 <author>Williams, J.B., J.D. Shorthouse and R.E. Lee, Jr.</author> <date>2002.</date> <title>Extreme resistance to desiccation and microclimate-related differences in cold-hardiness of gall wasps (Hymenoptera; Cynipidae) overwintering on roses in southern Canada.</title> <journal>The Journal of Experimental Biology</journal> <volume>205:</volume> <pages>2115–2124.</pages>
-<author>MACDONALD, S., & FENNIAK T.</author> <date>(2007).</date> <title>Understory plant communities of boreal mixedwood forests in western Canada: Natural patterns and response to variable-retention harvesting.</title> <journal>Forest Ecology and Management.</journal> <volume>242(1):</volume> <pages>34–48.</pages>
+<author>MACDONALD, S., &amp; FENNIAK T.</author> <date>(2007).</date> <title>Understory plant communities of boreal mixedwood forests in western Canada: Natural patterns and response to variable-retention harvesting.</title> <journal>Forest Ecology and Management.</journal> <volume>242(1):</volume> <pages>34–48.</pages>
 <author>Harris, P. and J.D. Shorthouse.</author> <date>1996.</date> <title>Effectiveness of gall inducers in weed biological control.</title> <journal>The Canadian Entomologist</journal> <volume>128:</volume> <pages>1021–1055.</pages>
 <author>Williams, J.B., J.D. Shorthouse and R.E. Lee, Jr.</author> <date>2003.</date> <title>Deleterious effects of mild simulated overwintering temperatures on survival and potential fecundity of rose-galling Diplolepis wasps (Hymenoptera: Cynipidae).</title> <journal>Journal of Experimental Zoology</journal> <volume>298A:</volume> <pages>23–31.</pages>
 <author>Shorthouse, J.D., H. Goulet and D.P. Shorthouse.</author> <date>2003.</date> <title>Notes on cynipid galls, ground beetles and ground-dwelling spiders collected at Fort Severn, Ontario.</title> <journal>Arctic</journal> <volume>56:</volume> <pages>159–167.</pages>
-<author>Epling, C., Lewis H., & Ball F. M.</author> <date>(1960).</date> <title>The Breeding Group and Seed Storage: A Study in Population Dynamics.</title> <journal>Evolution.</journal> <volume>14,</volume> <pages>238-255.</pages>
+<author>Epling, C., Lewis H., &amp; Ball F. M.</author> <date>(1960).</date> <title>The Breeding Group and Seed Storage: A Study in Population Dynamics.</title> <journal>Evolution.</journal> <volume>14,</volume> <pages>238-255.</pages>
 <author>Bagatto, Giuseppe, Louise C. Paquette, and Joseph D. Shorthouse.</author> <date>1995.</date> <title>Influence of galls of Phanacis taraxaci on carbon partitioning within common dandelion, Taraxacum officinale.</title> <journal>Entomologia Experimentalis et Applicata</journal> <volume>79:</volume> <pages>111–117.</pages>
 <author>Shorthouse, J.D.</author> <date>1993.</date> <title>Adaptations of gall wasps of the genus Diplolepis (Hymenoptera: Cynipidae) and the role of gall anatomy in cynipid systematics.</title> <journal>Memoirs of the Entomological Society of Canada</journal> <volume>165:</volume> <pages>139–163.</pages>
 <author>Bagatto, G., and J.D. Shorthouse.</author> <date>2000.</date> <title>Evaluation of municipal solid waste (MSW) compost as a soil amendment for acidic, metalliferous mine tailings.</title> <journal>International Journal of Surface Mining, Reclamation and Environment</journal> <volume>14:</volume> <pages>205–214.</pages>

data/spec/spec_helper.rb

@@ -38,6 +38,6 @@ RSpec.configure do |config|
   config.include Fixtures
 
   def strip_tags(string)
-    string.gsub(/<[^>]+>/, '')
+    Anystyle.parser.send :decode_xml_text, string.gsub(/<[^>]+>/, '')
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: anystyle-parser
 version: !ruby/object:Gem::Version
-  version: 0.6.2
+  version: 0.6.3
 platform: ruby
 authors:
 - Sylvester Keil
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-04-03 00:00:00.000000000 Z
+date: 2014-05-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bibtex-ruby
@@ -65,6 +65,9 @@ dependencies:
     - - ~>
       - !ruby/object:Gem::Version
         version: '0.8'
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: 0.8.6
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -72,6 +75,9 @@ dependencies:
     - - ~>
       - !ruby/object:Gem::Version
         version: '0.8'
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: 0.8.6
 description: A sophisticated parser for academic reference lists and bibliographies
   based on machine learning algorithms using conditional random fields.
 email: