datacite-mapping 0.2.2 → 0.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 6962dc851d06101de0b83fbc18ccba19cdb56c02
-  data.tar.gz: 2a11a09bb26c835049db959b4451aeebb9cabde2
+  metadata.gz: 94b1cd9d6fa3c9f73d4e01b8932735ed5531de96
+  data.tar.gz: 1f641d5ea6c21711480d806a4fc6e07b07df2697
 SHA512:
-  metadata.gz: 67d2bb08cd1a9d50f65ba8daa5308513a8b4363f72815599208423e620fe4a3666adb3c8fdd8a20ed3e70315bbdcbddb023d7885c97ae4f0a4a5bceea7e117f1
-  data.tar.gz: 6d3af828854f9caffbb9b56f5f92c2074600bb956f940febfefd2776aec878e8e078aa6901772c833e7ad21f7a3958cad7587a4ad667734da81a0a3f11766ef1
+  metadata.gz: 8958bbce65cce405ec02251e111902de276955f4f822a60409ef62c0532cc3315869db205f2491b95d8701579a623da711513c1938f637ab9c59ea84646668e9
+  data.tar.gz: 91bfefea1df89cf6067fa6f9a680c885ee49f76ff700849e88ada850016f42626fe3eec484da19efd1e1520ea16480dbea41afcbf5c75ec447d9b36237a8bbfe

data/CHANGES.md

@@ -1,3 +1,8 @@
+## 0.2.3 (5 October 2016)
+
+- Allow empty `<identifier/>` tags on read, but not write
+- Allow but ignore empty `<subject/>` and `<description/>` tags on read
+
 ## 0.2.2 (4 October 2016)
 
 - Fixed issue where `<geoLocation>` child elements would be written in Datacite 4

data/lib/datacite/mapping/empty_filtering_nodes.rb

@@ -0,0 +1,32 @@
+require 'xml/mapping_extensions'
+
+module Datacite
+  module Mapping
+
+    module EmptyNodeUtils
+      def not_empty(element)
+        return unless element
+        text = element.text
+        empty = text.nil? || text.strip.empty?
+        warn "Ignoring empty element #{element}" if empty
+        !empty
+      end
+    end
+
+    # An {XML::Mapping::ArrayNode} that ignores empty tags, including tags
+    # containing only blank text.
+    class EmptyFilteringArrayNode < XML::Mapping::ArrayNode
+      include EmptyNodeUtils
+      def extract_attr_value(xml)
+        elements = default_when_xpath_err { @reader_path.all(xml) }
+        non_empty_elements = elements.select { |e| not_empty(e) }
+        non_empty_elements.map { |e| unmarshal(e) }
+      end
+
+      def unmarshal(element)
+        @unmarshaller.call(element)
+      end
+    end
+    XML::Mapping.add_node_class EmptyFilteringArrayNode
+  end
+end

data/lib/datacite/mapping/identifier.rb

@@ -1,4 +1,5 @@
 require 'xml/mapping'
+require 'datacite/mapping/empty_filtering_nodes'
 
 module Datacite
   module Mapping
@@ -60,8 +61,10 @@ module Datacite
     # Custom node to warn (but not blow up) if we read an XML `<resource/>` that's
     # missing its `<identifier/>`.
     class IdentifierNode < XML::Mapping::ObjectNode
+      include EmptyNodeUtils
       def xml_to_obj(_obj, xml)
-        super if has_element?(xml)
+        return super if (element = has_element?(xml)) && not_empty(element)
+        warn 'Identifier not found; add a valid Identifier to the Resource before saving'
       end
 
       private
@@ -69,7 +72,7 @@ module Datacite
       def has_element?(xml) # rubocop:disable Style/PredicateName
         @path.first(xml)
       rescue XML::XXPathError
-        warn '<identifier/> not found; add a valid Identifier to the Resource before saving'
+        false
       end
     end
     XML::Mapping.add_node_class IdentifierNode

data/lib/datacite/mapping/module_info.rb

@@ -4,7 +4,7 @@ module Datacite
     NAME = 'datacite-mapping'.freeze
 
     # The version of this gem
-    VERSION = '0.2.2'.freeze
+    VERSION = '0.2.3'.freeze
 
     # The copyright notice for this gem
     COPYRIGHT = 'Copyright (c) 2016 The Regents of the University of California'.freeze

data/lib/datacite/mapping/resource.rb

@@ -184,7 +184,7 @@ module Datacite
 
       # @!attribute [rw] subjects
       #   @return [Array<Subject>] subjects, keywords, classification codes, or key phrases describing the resource.
-      array_node :subjects, 'subjects', 'subject', class: Subject, default_value: []
+      empty_filtering_array_node :subjects, 'subjects', 'subject', class: Subject, default_value: []
 
       # @!attribute [rw] fundingReferences
       #   @return [Array<FundingReference>] information about financial support (funding) for the resource being registered.
@@ -232,7 +232,7 @@ module Datacite
 
       # @!attribute [rw] descriptions
       #   @return [Array<Description>] all additional information that does not fit in any of the other categories.
-      array_node :descriptions, 'descriptions', 'description', class: Description, default_value: []
+      empty_filtering_array_node :descriptions, 'descriptions', 'description', class: Description, default_value: []
 
       # @!attribute [rw] geo_locations
       #   @return [Array<GeoLocations>] spatial region or named place where the data was gathered or about which the data is focused.

data/spec/unit/datacite/mapping/resource_spec.rb

@@ -938,7 +938,8 @@ module Datacite
         def normalize(xml_str)
           r0 = xml_str
           r1 = r0.gsub(%r{&lt;br\s+/&gt;}, '<br/>') # entity-de-escape <br/> tags
-          r2 = r1.gsub(%r{<(?!br)[^>]+/>}, '') # remove empty tags
+          # r2 = r1.gsub(%r{<(?!br)[^>]+/>}, '') # remove empty tags
+          r2 = r1
           r3 = r2.gsub(/<resource (xmlns:xsi="[^"]+")\s+(xsi:schemaLocation="[^"]+")>/, "<resource \\2 \\1 xmlns=\"http://datacite.org/schema/kernel-3\">") # fix missing namespace
           r4 = r3.gsub(%r{(<identifier[^>]+>)\s*([^ ]+)\s*(</identifier>)}, '\\1\\2\\3') # trim identifiers
           r5 = r4.gsub(%r{<([^>]+tude)>([0-9.-]+?)(0?)0+</\1>}, '<\\1>\\2\\3</\\1>') # strip trailing coordinate zeroes
@@ -966,21 +967,12 @@ module Datacite
           # - missing DOI
           # - empty tags
           # - nested contributors instead of contributorNames
-          # - empty descriptions
-          # TODO: handle empty descriptions like empty subjects
 
           r0 = xml_str
           r1 = r0.gsub(%r{<(?!br)[^>]+/>}, '') # remove empty tags
-          r2 = r1.gsub(%r{<description[^/>]*/>}, '') # strip empty descriptions
-          r3 = r2.gsub(%r{<description[^/>]*></description>}, '') # strip empty descriptions
-          r4 = r3.gsub(%r{<([A-Za-z]*)[^>]*>\s*</\1>}, '') # remove empty tag pairs
-          r5 = r4.gsub(%r{(<date[^>]*>)(\d{4})-(\d{4})(</date>)}, '\\1\\2/\\3\\4') # fix date ranges
-          r6 = r5.gsub(%r{(<contributor[^>/]+>\s*)<contributor>([^<]+)</contributor>(\s*</contributor>)}, '\\1<contributorName>\\2</contributorName>\\3') # fix broken contributors
-          # if r6.include?('&lt;br')
-          #   trace = [r0, r1, r2, r3, r4, r5, r6].map { |r| r.include?('&lt;br') }
-          #   puts trace
-          # end
-          r6
+          r2 = r1.gsub(%r{<([A-Za-z]*)[^>]*>\s*</\1>}, '') # remove empty tag pairs
+          r3 = r2.gsub(%r{(<date[^>]*>)(\d{4})-(\d{4})(</date>)}, '\\1\\2/\\3\\4') # fix date ranges
+          r3.gsub(%r{(<contributor[^>/]+>\s*)<contributor>([^<]+)</contributor>(\s*</contributor>)}, '\\1<contributorName>\\2</contributorName>\\3') # fix broken contributors
         end
 
         def it_round_trips(file:, mapping: :_default, fix_dash1: false) # rubocop:disable Metrics/AbcSize
@@ -1165,6 +1157,118 @@ module Datacite
           end
         end
       end
+
+      describe '#save_to_xml' do
+
+        attr_reader :resource
+
+        before(:each) do
+          @identifier = Identifier.new(value: '10.14749/1407399495')
+
+          @creators = [
+            Creator.new(
+              name: 'Hedy Lamarr',
+              identifier: NameIdentifier.new(scheme: 'ISNI', scheme_uri: URI('http://isni.org/'), value: '0000-0001-1690-159X'),
+              affiliations: ['United Artists', 'Metro-Goldwyn-Mayer']
+            ),
+            Creator.new(
+              name: 'Herschlag, Natalie',
+              identifier: NameIdentifier.new(scheme: 'ISNI', scheme_uri: URI('http://isni.org/'), value: '0000-0001-0907-8419'),
+              affiliations: ['Gaumont Buena Vista International', '20th Century Fox']
+            )
+          ]
+
+          @titles = [
+            Title.new(value: 'An Account of a Very Odd Monstrous Calf', language: 'en-emodeng'),
+            Title.new(type: TitleType::SUBTITLE, value: 'And a Contest between Two Artists about Optick Glasses, &c', language: 'en-emodeng')
+          ]
+
+          @publisher = 'California Digital Library'
+          @publication_year = 2015
+
+          @resource = Resource.new(
+            identifier: identifier,
+            creators: creators,
+            titles: titles,
+            publisher: publisher,
+            publication_year: publication_year
+          )
+        end
+
+        it 'sets the DC4 namespace by default' do
+          xml = resource.save_to_xml
+          expect(xml).to be_a(REXML::Element)
+          expect(xml.namespace).to eq(DATACITE_4_NAMESPACE.uri)
+        end
+
+        it 'sets the DC3 namespace for the :datacite_3 mapping' do
+          xml = resource.save_to_xml(mapping: :datacite_3)
+          expect(xml).to be_a(REXML::Element)
+          expect(xml.namespace).to eq(DATACITE_3_NAMESPACE.uri)
+        end
+
+        it 'fails on nil identifiers' do
+          resource.instance_variable_set(:@identifier, nil)
+          expect { resource.save_to_xml }.to raise_error(XML::MappingError)
+        end
+      end
+
+      describe '#parse_xml' do
+
+        attr_reader :xml_text
+
+        before(:each) do
+          @xml_text = "<resource xsi:schemaLocation='http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' xmlns='http://datacite.org/schema/kernel-4'>
+            <identifier identifierType='DOI'>10.14749/1407399495</identifier>
+            <creators>
+              <creator>
+                <creatorName>Hedy Lamarr</creatorName>
+                <nameIdentifier nameIdentifierScheme='ISNI' schemeURI='http://isni.org/'>0000-0001-1690-159X</nameIdentifier>
+                <affiliation>United Artists</affiliation>
+                <affiliation>Metro-Goldwyn-Mayer</affiliation>
+              </creator>
+              <creator>
+                <creatorName>Herschlag, Natalie</creatorName>
+                <nameIdentifier nameIdentifierScheme='ISNI' schemeURI='http://isni.org/'>0000-0001-0907-8419</nameIdentifier>
+                <affiliation>Gaumont Buena Vista International</affiliation>
+                <affiliation>20th Century Fox</affiliation>
+              </creator>
+            </creators>
+            <titles>
+              <title xml:lang='en-emodeng'>An Account of a Very Odd Monstrous Calf</title>
+              <title xml:lang='en-emodeng' titleType='Subtitle'>And a Contest between Two Artists about Optick Glasses, &amp;c</title>
+            </titles>
+            <publisher>California Digital Library</publisher>
+            <publicationYear>2015</publicationYear>
+            <subjects>
+              <subject xml:lang='en-us' schemeURI='http://id.loc.gov/authorities/subjects' subjectScheme='LCSH'>Mammals--Embryology</subject>
+            </subjects>
+            <descriptions>
+              <description xml:lang='en-us' descriptionType='Abstract'>foo</description>
+            </descriptions>
+          </resource>"
+        end
+
+        it 'skips empty identifiers' do
+          sketchy_xml = xml_text.gsub(%r{<identifier.*/identifier>}, '<identifier/>')
+          resource = Resource.parse_xml(sketchy_xml)
+          expect(resource).to be_a(Resource)
+        end
+
+        it 'skips empty subjects' do
+          sketchy_xml = xml_text.gsub(%r{>[^<]+</subject>}, '/>')
+          resource = Resource.parse_xml(sketchy_xml)
+          expect(resource).to be_a(Resource)
+          expect(resource.subjects).to eq([])
+        end
+
+        it 'skips empty descriptions' do
+          sketchy_xml = xml_text.gsub(%r{>[^<]+</description>}, '/>')
+          resource = Resource.parse_xml(sketchy_xml)
+          expect(resource).to be_a(Resource)
+          expect(resource.descriptions).to eq([])
+        end
+      end
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: datacite-mapping
 version: !ruby/object:Gem::Version
-  version: 0.2.2
+  version: 0.2.3
 platform: ruby
 authors:
 - David Moles
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-10-04 00:00:00.000000000 Z
+date: 2016-10-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: typesafe_enum
@@ -217,6 +217,7 @@ files:
 - lib/datacite/mapping/date.rb
 - lib/datacite/mapping/date_value.rb
 - lib/datacite/mapping/description.rb
+- lib/datacite/mapping/empty_filtering_nodes.rb
 - lib/datacite/mapping/funding_reference.rb
 - lib/datacite/mapping/geo_location.rb
 - lib/datacite/mapping/geo_location_box.rb