hydra-pbcore 2.3.0 → 2.4.0

data/lib/hydra_pbcore/conversions.rb

@@ -39,18 +39,16 @@ module HydraPbcore::Conversions
     self.ng_xml = self.ng_xml.xpath("//pbcoreInstantiation").to_xml
   end
 
-  # Corrects errors in HydraPbcore::Datastream::Deprecated::Document and HydraPbcore::Datastream::Deprecated::DigitalDocument
+  # Corrects errors in HydraPbcore::Datastream::Deprecated::Document
   # - removes all pbcoreRelation nodes, except those that define event_series
   # - removes orphaned pbcoreRelationIdentifier nodes
   # - corrects invalid usage of event_place and event_series terms
-  #
-  # Note: Since pbcoreRelation is used to indicated archival collection and series, this information should be copied to another kind
-  # of datastream using an RDF relationship in Hydra.
   def clean_document xml = self.ng_xml
     xml.search("//pbcoreRelation").each do |node|  
-      node.remove unless is_event_series?(node)
+      node.remove unless is_relation?("Event Series", node)
     end
     xml.search("/pbcoreDescriptionDocument/pbcoreRelationIdentifier").collect {|n| n.remove}
+    xml.search("/pbcoreDescriptionDocument/instantiationRelationIdentifier").collect {|n| n.remove}
     xml.search("//pbcoreCoverage").each do |node|
       node.children.each do |c|
         if c.attribute("annotation").nil? and c.name == "coverage"
@@ -61,10 +59,36 @@ module HydraPbcore::Conversions
     end
   end
 
-  # Determines if the given node defines an event_series term
-  def is_event_series? node
+  # Corrects errors in HydraPbcore::Datastream::Deprecated::DigitalDocument
+  # - finds the current archival collection
+  # - removes all pbcoreRelation nodes, except those that define event_series
+  # - re-inserts the collection
+  # - removes orphaned pbcoreRelationIdentifier nodes
+  # - corrects invalid usage of event_place and event_series terms
+  def clean_digital_document xml = self.ng_xml
+    collection = xml.search("//pbcoreRelationIdentifier[@annotation='Archival Collection']").text
+    xml.search("//pbcoreRelation").each do |node|  
+      unless is_relation?("Accession Number", node) or is_relation?("Event Series", node)
+        node.remove
+      end
+    end
+    xml.search("/pbcoreDescriptionDocument/pbcoreRelationIdentifier").collect {|n| n.remove}
+    xml.search("/pbcoreDescriptionDocument/instantiationRelationIdentifier").collect {|n| n.remove}
+    self.insert_relation(collection, "Archival Collection") unless collection.nil?
+    xml.search("//pbcoreCoverage").each do |node|
+      node.children.each do |c|
+        if c.attribute("annotation").nil? and c.name == "coverage"
+          self.send(("insert_"+coverage_type(node)), c.text)
+          c.remove
+        end
+      end
+    end
+  end
+
+  # Determines if the given node defines a relation of a give type
+  def is_relation? type, node
     unless node.at_xpath("pbcoreRelationIdentifier").nil?
-      if node.at_xpath("pbcoreRelationIdentifier").attribute("annotation").to_s == "Event Series"
+      if node.at_xpath("pbcoreRelationIdentifier").attribute("annotation").to_s == type
         return true
       end
     end

data/lib/hydra_pbcore/version.rb

@@ -1,3 +1,3 @@
 module HydraPbcore
-  VERSION = "2.3.0"
+  VERSION = "2.4.0"
 end

data/spec/fixtures/converted_digital_document_rrhof_1904.xml

@@ -7,6 +7,14 @@
     <pbcoreRelationType ref="http://pbcore.org/vocabularies/relationType#is-part-of" source="PBCore relationType">Is Part Of</pbcoreRelationType>
     <pbcoreRelationIdentifier annotation="Event Series">Oral history project (Rock and Roll Hall of Fame and Museum)</pbcoreRelationIdentifier>
   </pbcoreRelation>
+  <pbcoreRelation>
+    <pbcoreRelationType source="PBCore relationType" ref="http://pbcore.org/vocabularies/relationType#is-part-of">Is Part Of</pbcoreRelationType>
+    <pbcoreRelationIdentifier annotation="Archival Collection">Rock and Roll Hall of Fame and Museum records. Education and Public Programs Division. </pbcoreRelationIdentifier>
+  </pbcoreRelation>
+  <pbcoreRelation>
+    <pbcoreRelationType ref="http://pbcore.org/vocabularies/relationType#is-part-of" source="PBCore relationType">Is Part Of</pbcoreRelationType>
+    <pbcoreRelationIdentifier annotation="Accession Number"/>
+  </pbcoreRelation>
   <pbcoreCoverage>
     <coverage annotation="Event Place">Oklahoma City, OK</coverage>
     <coverageType>Spatial</coverageType>

data/spec/fixtures/converted_digital_document_rrhof_2405.xml

@@ -7,6 +7,14 @@
     <pbcoreRelationType ref="http://pbcore.org/vocabularies/relationType#is-part-of" source="PBCore relationType">Is Part Of</pbcoreRelationType>
     <pbcoreRelationIdentifier annotation="Event Series">Oral history project (Rock and Roll Hall of Fame and Museum)</pbcoreRelationIdentifier>
   </pbcoreRelation>
+  <pbcoreRelation>
+    <pbcoreRelationType ref="http://pbcore.org/vocabularies/relationType#is-part-of" source="PBCore relationType">Is Part Of</pbcoreRelationType>
+    <pbcoreRelationIdentifier annotation="Accession Number">LA.2012.05.024</pbcoreRelationIdentifier>
+  </pbcoreRelation>
+  <pbcoreRelation>
+    <pbcoreRelationType ref="http://pbcore.org/vocabularies/relationType#is-part-of" source="PBCore relationType">Is Part Of</pbcoreRelationType>
+    <pbcoreRelationIdentifier annotation="Archival Collection">Rock and Roll Hall of Fame and Museum records. Education and Public Programs Division.</pbcoreRelationIdentifier>
+  </pbcoreRelation>
   <pbcoreCoverage>
     <coverage annotation="Event Place">Pound Ridge, New York</coverage>
     <coverageType>Spatial</coverageType>

data/spec/fixtures/integration/digital_document_rrhof_1904.xml

@@ -1,3 +1,4 @@
+
 <pbcoreDescriptionDocument xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.pbcore.org/PBCore/PBCoreNamespace.html">
   <pbcoreIdentifier annotation="PID" source="Rock and Roll Hall of Fame and Museum"></pbcoreIdentifier>
   <pbcoreTitle titleType="Main">Oral history project. Wanda Jackson.</pbcoreTitle>
@@ -60,4 +61,7 @@
   <pbcoreRelationIdentifier annotation="Archival Collection">Rock and Roll Hall of Fame and Museum records. Education and Public Programs Division. </pbcoreRelationIdentifier>
   <pbcoreRelationIdentifier annotation="Collection Number">RG.0004</pbcoreRelationIdentifier>
   <pbcoreSubject ref="http://id.loc.gov/authorities/subjects.html" source="Library of Congress Subject Headings">Inductee</pbcoreSubject>
+  <instantiationRelationIdentifier annotation="Archival collection">Rock and Roll Hall of Fame and Museum records. Education and Public Programs Division</instantiationRelationIdentifier>
+  <instantiationRelationIdentifier annotation="Archival Series">Oral history project</instantiationRelationIdentifier>
+  <instantiationRelationIdentifier annotation="Collection Number">RG.0004</instantiationRelationIdentifier>
 </pbcoreDescriptionDocument>

data/spec/fixtures/integration/digital_document_rrhof_2405.xml

@@ -1,3 +1,4 @@
+
 <pbcoreDescriptionDocument xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.pbcore.org/PBCore/PBCoreNamespace.html">
   <pbcoreIdentifier annotation="PID" source="Rock and Roll Hall of Fame and Museum"></pbcoreIdentifier>
   <pbcoreTitle titleType="Main">Oral History project. Clive Davis.</pbcoreTitle>
@@ -59,4 +60,7 @@
     <contributorRole source="MARC relator terms">producer</contributorRole>
   </pbcoreContributor>
   <pbcoreSubject ref="http://id.loc.gov/authorities/subjects.html" source="Library of Congress Subject Headings">Inductee</pbcoreSubject>
+  <instantiationRelationIdentifier annotation="Archival collection">Rock and Roll Hall of Fame and Museum records. Education and Public Programs Division</instantiationRelationIdentifier>
+  <instantiationRelationIdentifier annotation="Archival Series">Oral history project</instantiationRelationIdentifier>
+  <instantiationRelationIdentifier annotation="Collection Number">RG.0004</instantiationRelationIdentifier>
 </pbcoreDescriptionDocument>

data/spec/integration/conversions_spec.rb

@@ -47,9 +47,10 @@ describe "Converting" do
   end
 
   describe "#clean_document" do
-    it "should correct invalid pbcoreDigitalDocuments" do
-      ["digital_document_rrhof_1904.xml","digital_document_rrhof_2405.xml"].each do |file|
-        doc = HydraPbcore::Datastream::Deprecated::DigitalDocument.from_xml(integration_fixture file)
+    it "should correct invalid pbcoreDocuments" do
+      ["document_rrhof_524.xml","document_rrhof_2439.xml"].each do |file|
+        doc = HydraPbcore::Datastream::Deprecated::Document.from_xml(integration_fixture file)
+        doc.to_document
         doc.clean_document
         save_template doc.to_xml, ("converted_"+file)
         tmp_file_should_match_exemplar(("converted_"+file))
@@ -57,11 +58,13 @@ describe "Converting" do
         doc.valid?.should == []
       end
     end
-    it "should correct invalid pbcoreDocuments" do
-      ["document_rrhof_524.xml","document_rrhof_2439.xml"].each do |file|
-        doc = HydraPbcore::Datastream::Deprecated::Document.from_xml(integration_fixture file)
-        doc.to_document
-        doc.clean_document
+  end
+
+  describe "#clean_digital_document" do
+    it "should correct invalid pbcoreDigitalDocuments" do
+      ["digital_document_rrhof_1904.xml","digital_document_rrhof_2405.xml"].each do |file|
+        doc = HydraPbcore::Datastream::Deprecated::DigitalDocument.from_xml(integration_fixture file)
+        doc.clean_digital_document
         save_template doc.to_xml, ("converted_"+file)
         tmp_file_should_match_exemplar(("converted_"+file))
         save_template doc.to_pbcore_xml, "valid_pbcore.xml"

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hydra-pbcore
 version: !ruby/object:Gem::Version
-  version: 2.3.0
+  version: 2.4.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-21 00:00:00.000000000 Z
+date: 2013-05-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: active-fedora
@@ -219,7 +219,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 3539419655098167212
+      hash: 3882171645005853524
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -228,7 +228,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: 3539419655098167212
+      hash: 3882171645005853524
 requirements: []
 rubyforge_project: 
 rubygems_version: 1.8.23