datacite-mapping 0.2.1 → 0.2.2

data/spec/data/dash1/ucsf-ark+=b7272=q6sf2t3q-mrt-datacite.xml

@@ -1,5 +1,6 @@
-<?xml version="1.0" encoding="utf-8"?><resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
-<identifier identifierType="DOI"/>
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+<identifier identifierType="DOI"></identifier>
 <creators>
 <creator><creatorName>Early Childhood Caries Collaborating Centers</creatorName></creator>
 </creators>
@@ -15,7 +16,7 @@
 </contributors>
 <resourceType resourceTypeGeneral="Dataset">Dataset</resourceType>
 <descriptions>
-<description descriptionType="Abstract"/>
-<description descriptionType="Methods"/>
+<description descriptionType="Abstract"></description>
+<description descriptionType="Methods"></description>
 </descriptions>
-</resource>
+</resource>

data/spec/data/dash1/ucsf-ark+=b7272=q6td9v7j-mrt-datacite.xml

@@ -1,5 +1,6 @@
-<?xml version="1.0" encoding="utf-8"?><resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
-<identifier identifierType="DOI"/>
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+<identifier identifierType="DOI"></identifier>
 <creators>
 <creator><creatorName>Megan Laurance</creatorName></creator>
 </creators>
@@ -26,4 +27,4 @@
 <description descriptionType="SeriesInformation">The original microarray data that was used for this reanalysis use case is described in the GEO record GSE27255.</description>
 <description descriptionType="SeriesInformation">http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE27255 </description>
 </descriptions>
-</resource>
+</resource>

data/spec/data/dash1/ucsf-ark+=b7272=q6v985zk-mrt-datacite.xml

@@ -0,0 +1,52 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+  <identifier identifierType="DOI"/>
+  <creators>
+    <creator>
+      <creatorName>Upadhyay, Ushma D.</creatorName>
+    </creator>
+    <creator>
+      <creatorName>Johns, Nicole E.</creatorName>
+    </creator>
+    <creator>
+      <creatorName>Combellick, Sarah L.</creatorName>
+    </creator>
+    <creator>
+      <creatorName>Kohn, Julia E.</creatorName>
+    </creator>
+    <creator>
+      <creatorName>Keder, Lisa M.</creatorName>
+    </creator>
+    <creator>
+      <creatorName>Roberts, Sarah C.M.</creatorName>
+    </creator>
+  </creators>
+  <titles>
+    <title>Comparison of Outcomes before and after Ohio's Law Mandating Use of the FDA Approved Protocol for Medication Abortion: A Retrospective Cohort Study</title>
+  </titles>
+  <publisher>UC San Francisco</publisher>
+  <publicationYear>2016</publicationYear>
+  <subjects/>
+  <contributors>
+    <contributor contributorType="DataManager">
+      <contributorName>Upadhyay, Ushma</contributorName>
+    </contributor>
+  </contributors>
+  <relatedIdentifiers>
+    <relatedIdentifier relatedIdentifierType="URL" relationType="Cites">http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002110</relatedIdentifier>
+  </relatedIdentifiers>
+  <resourceType resourceTypeGeneral="Dataset">Dataset</resourceType>
+  <sizes>
+    <size>1432064</size>
+  </sizes>
+  <rightsList>
+    <rights rightsURI="https://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution 4.0 International (CC-BY 4.0)</rights>
+  </rightsList>
+  <descriptions>
+    <description descriptionType="Abstract">Background: In February 2011, an Ohio law took effect mandating use of the United States Food and Drug Administration (FDA)-approved protocol for mifepristone, which is used with misoprostol for medication abortion.  Other state legislatures have passed or enacted similar laws requiring use of the FDA-approved protocol for medication abortion. The objective of this study is to examine the association of this legal change with medication abortion outcomes and utilization.
+
+Methods and Findings: We used a retrospective cohort design, comparing outcomes of medication abortion patients in the prelaw period to those in the postlaw period. Sociodemographic and clinical chart data were abstracted from all medication abortion patients from 1 y prior to the law’s implementation (January 2010–January 2011) to 3 y post implementation (February 2011–October 2014) at four abortion-providing health care facilities in Ohio. Outcome data were analyzed for all women undergoing abortion at ≤49 d gestation during the study period. The main outcomes were as follows: need for additional intervention following medication abortion (such as aspiration, repeat misoprostol, and blood transfusion), frequency of continuing pregnancy, reports of side effects, and the proportion of abortions that were medication abortions (versus other abortion procedures). Among the 2,783 medication abortions ≤49 d gestation, 4.9% (95% CI: 3.7%–6.2%) in the prelaw and 14.3% (95% CI: 12.6%–16.0%) in the postlaw period required one or more additional interventions. Women obtaining a medication abortion in the postlaw period had three times the odds of requiring an additional intervention as women in the prelaw period (adjusted odds ratio [AOR] = 3.11, 95% CI: 2.27–4.27). In a mixed effects multivariable model that uses facility-months as the unit of analysis to account for lack of independence by site, we found that the law change was associated with a 9.4% (95% CI: 4.0%–18.4%) absolute increase in the rate of requiring an additional intervention.  The most common subsequent intervention in both periods was an additional misoprostol dose and was most commonly administered to treat incomplete abortion. The percentage of women requiring two or more follow-up visits increased from 4.2% (95% CI: 3.0%–5.3%) in the prelaw period to 6.2% (95% CI: 5.5%–8.0%) in the postlaw period (p = 0.003). Continuing pregnancy was rare (0.3%). Overall, 12.6% of women reported at least one side effect during their medication abortion: 8.4% (95% CI: 6.8%–10.0%) in the prelaw period and 15.6% (95% CI: 13.8%–17.3%) in the postlaw period (p &lt; 0.001). Medication abortions fell from 22% (95% CI: 20.8%–22.3%) of all abortions the year before the law went into effect (2010) to 5% (95% CI: 4.8%–5.6%) 3 y after (2014) (p &lt; 0.001). The average patient charge increased from US$426 in 2010 to US$551 in 2014, representing a 16% increase after adjusting for inflation in medical prices.  The primary limitation to the study is that it was a pre/post-observational study with no control group that was not exposed to the law.  
+Conclusions: Ohio law required use of a medication abortion protocol that is associated with a greater need for additional intervention, more visits, more side effects, and higher costs for women relative to the evidence-based protocol.  There is no evidence that the change in law led to improved abortion outcomes.  Indeed, our findings suggest the opposite. In March 2016, the FDA-protocol was updated, so Ohio providers may now legally provide current evidence-based protocols. However, this law is still in place and bans physicians from using mifepristone based on any new developments in clinical research as best practices continue to be updated.
+</description>
+  </descriptions>
+</resource>

data/spec/data/dash1/ucsf-ark+=b7272=q6w66hpn-mrt-datacite.xml

@@ -1,5 +1,6 @@
-<?xml version="1.0" encoding="utf-8"?><resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
-<identifier identifierType="DOI"/>
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+<identifier identifierType="DOI"></identifier>
 <creators>
 <creator><creatorName>Robert Rodriguez</creatorName></creator>
 </creators>
@@ -17,7 +18,7 @@
 </contributors>
 <resourceType resourceTypeGeneral="Dataset">Dataset</resourceType>
 <descriptions>
-<description descriptionType="Abstract"/>
-<description descriptionType="Methods"/>
+<description descriptionType="Abstract"></description>
+<description descriptionType="Methods"></description>
 </descriptions>
-</resource>
+</resource>

data/spec/data/dash1/ucsf-ark+=b7272=q6x63jt1-mrt-datacite.xml

@@ -1,5 +1,6 @@
-<?xml version="1.0" encoding="utf-8"?><resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
-<identifier identifierType="DOI"/>
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+<identifier identifierType="DOI"></identifier>
 <creators>
 <creator><creatorName>McDaniel, Patricia A</creatorName></creator>
 <creator><creatorName>Malone, Ruth E</creatorName></creator>
@@ -24,4 +25,4 @@
 <description descriptionType="SeriesInformation">McDaniel PA, Malone RE. Why California retailers stop selling tobacco products, and what their customers and employees think about it when they do: case studies. BMC Public Health. 2011 Nov 8;11(1):848. </description>
 <description descriptionType="SeriesInformation">McDaniel PA, Malone RE. “People over Profits”: Retailers Who Voluntarily Ended Tobacco Sales. PLoS ONE 2014 9(1): e85751. </description>
 </descriptions>
-</resource>
+</resource>

data/spec/data/dash1/ucsf-ark+=b7272=q6z60kzd-mrt-datacite.xml

@@ -1,45 +1,46 @@
-<?xml version="1.0" encoding="utf-8"?><resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
-<identifier identifierType="DOI">10.7272/Q6Z60KZD</identifier>
-<creators>
-<creator>
-<creatorName>Weiner, Michael W.</creatorName>
-</creator>
-<creator>
-<creatorName>Chui, Helena</creatorName>
-</creator>
-</creators>
-<titles>
-<title>SIVD (Subcortical Ischemic Vascular Dementia)</title>
-</titles>
-<publisher>University of California, San Francisco</publisher>
-<publicationYear>2013</publicationYear>
-<contributors>
-<contributor contributorType="ResearchGroup">
-<contributorName>UCSF Center for Imaging of Neurodegenerative Diseases</contributorName>
-</contributor>
-</contributors>
-<resourceType resourceTypeGeneral="Dataset">dataset</resourceType>
-<relatedIdentifiers>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11756602</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10369354</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10782769</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11796772</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12040300</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12058091</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12297568</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12913198</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15277612</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15595356</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15653183</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15929035</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">17192928</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">14872024</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10369354</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10782769</relatedIdentifier>
-<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11113215</relatedIdentifier>
-</relatedIdentifiers>
-<descriptions>
-<description descriptionType="Abstract">The overall goal of this project was to determine the extent to which subcortical ischemia and infaraction, and Alzheimer's disease, contribute to dementia. This project aimed to provide a greater understanding of the pathogenesis of dementia in patients with SIVD and AD.</description>
-<description descriptionType="Methods">Data were acquired at several sites: ABS = Aging Brain San Francisco (SFVAMC), ABD = Aging Brain UC Davis, and ABL = Aging Brain Los Angeles (USC).</description>
-</descriptions>
+<?xml version="1.0" encoding="UTF-8"?>
+<resource xmlns="http://datacite.org/schema/kernel-3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://datacite.org/schema/kernel-3 http://schema.datacite.org/meta/kernel-3/metadata.xsd">
+<identifier identifierType="DOI">10.7272/Q6Z60KZD</identifier>
+<creators>
+<creator>
+<creatorName>Weiner, Michael W.</creatorName>
+</creator>
+<creator>
+<creatorName>Chui, Helena</creatorName>
+</creator>
+</creators>
+<titles>
+<title>SIVD (Subcortical Ischemic Vascular Dementia)</title>
+</titles>
+<publisher>University of California, San Francisco</publisher>
+<publicationYear>2013</publicationYear>
+<contributors>
+<contributor contributorType="ResearchGroup">
+<contributorName>UCSF Center for Imaging of Neurodegenerative Diseases</contributorName>
+</contributor>
+</contributors>
+<resourceType resourceTypeGeneral="Dataset">dataset</resourceType>
+<relatedIdentifiers>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11756602</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10369354</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10782769</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11796772</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12040300</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12058091</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12297568</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">12913198</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15277612</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15595356</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15653183</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">15929035</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">17192928</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">14872024</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10369354</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">10782769</relatedIdentifier>
+<relatedIdentifier relatedIdentifierType="PMID" relationType="References">11113215</relatedIdentifier>
+</relatedIdentifiers>
+<descriptions>
+<description descriptionType="Abstract">The overall goal of this project was to determine the extent to which subcortical ischemia and infaraction, and Alzheimer's disease, contribute to dementia. This project aimed to provide a greater understanding of the pathogenesis of dementia in patients with SIVD and AD.</description>
+<description descriptionType="Methods">Data were acquired at several sites: ABS = Aging Brain San Francisco (SFVAMC), ABD = Aging Brain UC Davis, and ABL = Aging Brain Los Angeles (USC).</description>
+</descriptions>
 </resource>

data/spec/rspec_custom_matchers.rb

@@ -38,8 +38,8 @@ RSpec::Matchers.define :be_xml do |expected|
     expected_string = to_pretty(to_nokogiri(expected))
     actual_string = to_pretty(to_nokogiri(actual)) || actual
 
-    File.open('tmp/actual.xml', 'w') { |f| f.write(actual_string) }
-    File.open('tmp/expected.xml', 'w') { |f| f.write(expected_string) }
+    # File.open('tmp/actual.xml', 'w') { |f| f.write(actual_string) }
+    # File.open('tmp/expected.xml', 'w') { |f| f.write(expected_string) }
 
     "expected XML:\n#{expected_string}\n\nbut was:\n#{actual_string}"
   end

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

@@ -89,6 +89,35 @@ module Datacite
         end
       end
 
+      describe '#<=>' do
+        it 'reports equal values as equal' do
+          d1 = Date.new(value: DateTime.new(1914, 8, 4, 11, 1, 6.0123, '+1'), type: DateType::AVAILABLE)
+          d2 = Date.new(value: '1914-08-04T11:01:06.0123+01:00', type: DateType::AVAILABLE)
+          expect(d1).to eq(d2)
+          expect(d2).to eq(d1)
+          expect(d1.hash).to eq(d2.hash)
+          expect(d2.hash).to eq(d1.hash)
+        end
+
+        it 'reports unequal values as unequal' do
+          d1 = Date.new(value: DateTime.new(1914, 8, 4, 11, 1, 6.0123, '+1'), type: DateType::AVAILABLE)
+          d2 = Date.new(value: '1914-08-04T11:01+01:00', type: DateType::AVAILABLE)
+          expect(d1).not_to eq(d2)
+          expect(d2).not_to eq(d1)
+          expect(d1.hash).not_to eq(d2.hash)
+          expect(d2.hash).not_to eq(d1.hash)
+        end
+
+        it 'reports different types as different' do
+          d1 = Date.new(value: '1914-08-04T11:01+01:00', type: DateType::VALID)
+          d2 = Date.new(value: '1914-08-04T11:01+01:00', type: DateType::AVAILABLE)
+          expect(d1).not_to eq(d2)
+          expect(d2).not_to eq(d1)
+          expect(d1.hash).not_to eq(d2.hash)
+          expect(d2.hash).not_to eq(d1.hash)
+        end
+      end
+
       describe '#load_from_xml' do
         it 'reads XML' do
           xml = '<date dateType="Available">1914-08-04T11:01:06.0123+01:00</date>'

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

@@ -254,6 +254,13 @@ module Datacite
             expect(actual_xml).to be_xml(expected_xml)
             expect_warning(loc.polygon.to_s, 1)
           end
+
+          it 'writes DC3 in XSD-defined order: point, box, place' do
+            actual_xml = loc.save_to_xml(mapping: :datacite_3)
+            expected_order = %w(geoLocationPoint geoLocationBox geoLocationPlace)
+            actual_order = actual_xml.children.map(&:name)
+            expect(actual_order).to eq(expected_order)
+          end
         end
       end
     end

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

@@ -717,6 +717,27 @@ module Datacite
           '
             expect(resource.version).to eq('9.2.2')
           end
+          it 'accepts a float' do
+            resource = Resource.new(args)
+            resource.version = 9.2
+            expect(resource.version).to eq('9.2')
+          end
+          it 'accepts an integer' do
+            resource = Resource.new(args)
+            resource.version = 9
+            expect(resource.version).to eq('9')
+          end
+          it 'can be initialized' do
+            args[:version] = '9.2.2'
+            resource = Resource.new(args)
+            expect(resource.version).to eq('9.2.2')
+          end
+          it 'can be set to nil' do
+            args[:version] = '9.2.2'
+            resource = Resource.new(args)
+            resource.version = nil
+            expect(resource.version).to be_nil
+          end
         end
 
         describe '#rights_list' do
@@ -915,12 +936,29 @@ module Datacite
         end
 
         def normalize(xml_str)
-          xml_str
-            .gsub(%r{<([^>]+tude)>([0-9.-]+?)(0?)0+</\1>}, '<\\1>\\2\\3</\\1>') # strip trailing coordinate zeroes
-            .gsub(/ "([^"]+)"/, ' &quot;\\1&quot;') # entity-escape quotes in text
-            .gsub('&lt;br /&gt;', '<br/>') # entity-de-escape <br/> tags
-            .gsub('"', "'") # swap double for single quotes
-            .gsub(%r{<(geoLocation[^>]+)>[^<]+</\1>}) { |loc| loc.gsub(/([0-9\-]+\.[0-9]+?)0+([^0-9])/, '\\1\\2') } # strip trailing coordinate zeroes
+          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
+          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
+          r6 = r5.gsub(%r{<(geoLocation[^>]+)>[^<]+</\1>}) { |loc| loc.gsub(/([0-9\-]+\.[0-9]+?)0+([^0-9])/, '\\1\\2') } # strip trailing coordinate zeroes
+          r7 = r6.gsub(%r{<([A-Za-z]*)[^>]*>\s*</\1>}, '') # remove empty tag pairs
+          r8 = to_pretty(r7)
+          # if r8.include?('&lt;br')
+          #   trace = [r0, r1, r2, r3, r4, r5, r6, r7, r8].map { |r| r.include?('&lt;br') }
+          #   puts trace
+          # end
+          r8
+        end
+
+        def to_pretty(xml_text)
+          xml = REXML::Document.new(xml_text).root
+          formatter = REXML::Formatters::Pretty.new
+          formatter.compact = true
+          io = ::StringIO.new
+          formatter.write(xml, io)
+          io.string
         end
 
         def fix_dash1(xml_str)
@@ -928,22 +966,33 @@ module Datacite
           # - missing DOI
           # - empty tags
           # - nested contributors instead of contributorNames
-          xml_str
-            .gsub(%r{<[^>]+/>}, '') # remove empty tags
-            .gsub(%r{<([^>]+)>\s+</\1>}, '') # remove empty tag pairs
-            .gsub(%r{(<date[^>]*>)(\d{4})-(\d{4})(</date>)}, '\\1\\2/\\3\\4') # fix date ranges
-            .gsub(%r{(<contributor[^>/]+>\s*)<contributor>([^<]+)</contributor>(\s*</contributor>)}, '\\1<contributorName>\\2</contributorName>\\3') # fix broken contributors
+          # - 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
         end
 
         def it_round_trips(file:, mapping: :_default, fix_dash1: false) # rubocop:disable Metrics/AbcSize
           options = { mapping: mapping }
           basename = File.basename(file)
           xml_text = xml_text_from(file, fix_dash1)
-          resource = parse_file(xml_text, basename)
-          actual_xml = write_xml(resource, basename, options)
           expected_xml = normalize(xml_text)
+          resource = parse_file(expected_xml, basename)
+          actual_xml = write_xml(resource, basename, options)
           expected_xml.gsub!(/(<resource[^>]+>)\s+(<creators>)/, "\\1\n  <identifier identifierType=\"DOI\">10.5555/12345678</identifier>\n  \\2") if fix_dash1
           begin
+            # actual_xml = actual_xml.gsub('&apos;', "'")
             expect(actual_xml).to be_xml(expected_xml)
           rescue Exception # rubocop:disable Lint/RescueException
             File.open("tmp/#{basename}-expected.xml", 'w') { |t| t.write(expected_xml) }

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: datacite-mapping
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.2.2
 platform: ruby
 authors:
 - David Moles
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-09-28 00:00:00.000000000 Z
+date: 2016-10-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: typesafe_enum
@@ -237,22 +237,38 @@ files:
 - spec/.rubocop.yml
 - spec/data/dash1/dataone-ark+=c5146=r3059p-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r31017-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r31s3x-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3201j-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r32s37-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r33w26-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r34s3v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r36k55-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r36p4t-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r38c7g-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r39884-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r39g6f-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3cc7d-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3d309-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3d59d-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3f30m-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3f59q-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3g591-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3hs38-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3k016-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3ks3w-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3kw2j-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3mw2v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3nk56-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3np4v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3pk5h-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3pp45-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3qg65-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3rg6g-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3rp4s-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3t88s-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3tg63-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3v883-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3x30z-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1159q-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d17g6j-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1c88g-mrt-datacite.xml
@@ -262,6 +278,7 @@ files:
 - spec/data/dash1/ucb-ark+=b6078=d1mw2k-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1rg66-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1rp4h-mrt-datacite.xml
+- spec/data/dash1/ucb-ark+=b6078=d1v88t-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1wc7s-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1001p-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1059f-mrt-datacite.xml
@@ -307,6 +324,7 @@ files:
 - spec/data/dash1/uci-ark+=b7280=d1qp47-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1rg67-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1rp4j-mrt-datacite.xml
+- spec/data/dash1/uci-ark+=b7280=d1s887-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1sg6j-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1t88j-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1tg6v-mrt-datacite.xml
@@ -314,6 +332,7 @@ files:
 - spec/data/dash1/uci-ark+=b7280=d1vc7h-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1w885-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1wc7t-mrt-datacite.xml
+- spec/data/dash1/uci-ark+=b7280=d1x30q-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1x59t-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1z594-mrt-datacite.xml
 - spec/data/dash1/ucla-ark+=b5060=d2qr4v2t-mrt-datacite.xml
@@ -336,12 +355,15 @@ files:
 - spec/data/dash1/ucsc-ark+=b7291=d1h59d-mrt-datacite.xml
 - spec/data/dash1/ucsc-ark+=b7291=d1mw2x-mrt-datacite.xml
 - spec/data/dash1/ucsc-ark+=b7291=d1wc74-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q600001h-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6057cv6-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6154f00-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q61z429d-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q62z13fs-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q63r0qr4-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q65q4t1r-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q66q1v54-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q67h1ggv-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q67p8w9z-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q68g8hmp-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6bg2kwf-mrt-datacite.xml
@@ -354,11 +376,13 @@ files:
 - spec/data/dash1/ucsf-ark+=b7272=q6ms3qnx-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6mw2f2n-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6pn93h6-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q6qj7f78-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6qn64nk-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6rf5rzx-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6rn35sz-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6sf2t3q-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6td9v7j-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q6v985zk-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6w66hpn-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6x63jt1-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6z60kzd-mrt-datacite.xml
@@ -436,22 +460,38 @@ test_files:
 - spec/.rubocop.yml
 - spec/data/dash1/dataone-ark+=c5146=r3059p-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r31017-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r31s3x-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3201j-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r32s37-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r33w26-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r34s3v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r36k55-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r36p4t-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r38c7g-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r39884-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r39g6f-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3cc7d-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3d309-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3d59d-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3f30m-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3f59q-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3g591-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3hs38-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3k016-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3ks3w-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3kw2j-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3mw2v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3nk56-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3np4v-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3pk5h-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3pp45-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3qg65-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3rg6g-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3rp4s-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3t88s-mrt-datacite.xml
 - spec/data/dash1/dataone-ark+=c5146=r3tg63-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3v883-mrt-datacite.xml
+- spec/data/dash1/dataone-ark+=c5146=r3x30z-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1159q-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d17g6j-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1c88g-mrt-datacite.xml
@@ -461,6 +501,7 @@ test_files:
 - spec/data/dash1/ucb-ark+=b6078=d1mw2k-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1rg66-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1rp4h-mrt-datacite.xml
+- spec/data/dash1/ucb-ark+=b6078=d1v88t-mrt-datacite.xml
 - spec/data/dash1/ucb-ark+=b6078=d1wc7s-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1001p-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1059f-mrt-datacite.xml
@@ -506,6 +547,7 @@ test_files:
 - spec/data/dash1/uci-ark+=b7280=d1qp47-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1rg67-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1rp4j-mrt-datacite.xml
+- spec/data/dash1/uci-ark+=b7280=d1s887-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1sg6j-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1t88j-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1tg6v-mrt-datacite.xml
@@ -513,6 +555,7 @@ test_files:
 - spec/data/dash1/uci-ark+=b7280=d1vc7h-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1w885-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1wc7t-mrt-datacite.xml
+- spec/data/dash1/uci-ark+=b7280=d1x30q-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1x59t-mrt-datacite.xml
 - spec/data/dash1/uci-ark+=b7280=d1z594-mrt-datacite.xml
 - spec/data/dash1/ucla-ark+=b5060=d2qr4v2t-mrt-datacite.xml
@@ -535,12 +578,15 @@ test_files:
 - spec/data/dash1/ucsc-ark+=b7291=d1h59d-mrt-datacite.xml
 - spec/data/dash1/ucsc-ark+=b7291=d1mw2x-mrt-datacite.xml
 - spec/data/dash1/ucsc-ark+=b7291=d1wc74-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q600001h-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6057cv6-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6154f00-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q61z429d-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q62z13fs-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q63r0qr4-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q65q4t1r-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q66q1v54-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q67h1ggv-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q67p8w9z-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q68g8hmp-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6bg2kwf-mrt-datacite.xml
@@ -553,11 +599,13 @@ test_files:
 - spec/data/dash1/ucsf-ark+=b7272=q6ms3qnx-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6mw2f2n-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6pn93h6-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q6qj7f78-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6qn64nk-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6rf5rzx-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6rn35sz-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6sf2t3q-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6td9v7j-mrt-datacite.xml
+- spec/data/dash1/ucsf-ark+=b7272=q6v985zk-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6w66hpn-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6x63jt1-mrt-datacite.xml
 - spec/data/dash1/ucsf-ark+=b7272=q6z60kzd-mrt-datacite.xml