dor-services 7.2.4 → 8.0.0

data/lib/dor/services/public_desc_metadata_service.rb

@@ -1,184 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class PublicDescMetadataService
-    extend Deprecation
-
-    attr_reader :object
-
-    NOKOGIRI_DEEP_COPY = 1
-
-    def initialize(object)
-      Deprecation.warn(self, 'Dor::PublicDescMetadataService has moved to dor-services-app')
-      @object = object
-    end
-
-    # @return [Nokogiri::XML::Document] A copy of the descriptiveMetadata of the object, to be modified
-    def doc
-      @doc ||= object.descMetadata.ng_xml.dup(NOKOGIRI_DEEP_COPY)
-    end
-
-    # @return [String] Public descriptive medatada XML
-    def to_xml(include_access_conditions: true)
-      ng_xml(include_access_conditions: include_access_conditions).to_xml
-    end
-
-    # @return [Nokogiri::XML::Document]
-    def ng_xml(include_access_conditions: true)
-      @ng_xml ||= begin
-        add_collection_reference!
-        add_access_conditions! if include_access_conditions
-        add_constituent_relations!
-        strip_comments!
-
-        new_doc = Nokogiri::XML(doc.to_xml, &:noblanks)
-        new_doc.encoding = 'UTF-8'
-        new_doc
-      end
-    end
-
-    private
-
-    def strip_comments!
-      doc.xpath('//comment()').remove
-    end
-
-    # Create MODS accessCondition statements from rightsMetadata
-    def add_access_conditions!
-      # clear out any existing accessConditions
-      doc.xpath('//mods:accessCondition', 'mods' => 'http://www.loc.gov/mods/v3').each(&:remove)
-      rights = object.datastreams['rightsMetadata'].ng_xml
-
-      rights.xpath('//use/human[@type="useAndReproduction"]').each do |use|
-        txt = use.text.strip
-        next if txt.empty?
-
-        doc.root.element_children.last.add_next_sibling doc.create_element('accessCondition', txt, type: 'useAndReproduction')
-      end
-      rights.xpath('//copyright/human[@type="copyright"]').each do |cr|
-        txt = cr.text.strip
-        next if txt.empty?
-
-        doc.root.element_children.last.add_next_sibling doc.create_element('accessCondition', txt, type: 'copyright')
-      end
-      rights.xpath("//use/machine[#{ci_compare('type', 'creativecommons')}]").each do |lic_type|
-        next if lic_type.text =~ /none/i
-
-        lic_text = rights.at_xpath("//use/human[#{ci_compare('type', 'creativecommons')}]").text.strip
-        next if lic_text.empty?
-
-        new_text = "CC #{lic_type.text}: #{lic_text}"
-        doc.root.element_children.last.add_next_sibling doc.create_element('accessCondition', new_text, type: 'license')
-      end
-      rights.xpath("//use/machine[#{ci_compare('type', 'opendatacommons')}]").each do |lic_type|
-        next if lic_type.text =~ /none/i
-
-        lic_text = rights.at_xpath("//use/human[#{ci_compare('type', 'opendatacommons')}]").text.strip
-        next if lic_text.empty?
-
-        new_text = "ODC #{lic_type.text}: #{lic_text}"
-        doc.root.element_children.last.add_next_sibling doc.create_element('accessCondition', new_text, type: 'license')
-      end
-    end
-
-    # expand constituent relations into relatedItem references -- see JUMBO-18
-    # @return [Void]
-    def add_constituent_relations!
-      object.relationships(:is_constituent_of).each do |parent|
-        # fetch the parent object to get title
-        druid = parent.gsub(/^info:fedora\//, '')
-        parent_item = Dor.find(druid)
-
-        # create the MODS relation
-        relatedItem = doc.create_element 'relatedItem'
-        relatedItem['type'] = 'host'
-        relatedItem['displayLabel'] = 'Appears in'
-
-        # load the title from the parent's DC.title
-        titleInfo = doc.create_element 'titleInfo'
-        title = doc.create_element 'title'
-        title.content = parent_item.full_title
-        titleInfo << title
-        relatedItem << titleInfo
-
-        # point to the PURL for the parent
-        location = doc.create_element 'location'
-        url = doc.create_element 'url'
-        url.content = "http://#{Dor::Config.stacks.document_cache_host}/#{druid.split(':').last}"
-        location << url
-        relatedItem << location
-
-        # finish up by adding relation to public MODS
-        doc.root << relatedItem
-      end
-    end
-
-    # Adds to desc metadata a relatedItem with information about the collection this object belongs to.
-    # For use in published mods and mods-to-DC conversion.
-    # @return [Void]
-    def add_collection_reference!
-      collections = object.relationships(:is_member_of_collection)
-      return if collections.empty?
-
-      remove_related_item_nodes_for_collections!
-
-      collections.each do |collection_uri|
-        collection_druid = collection_uri.gsub('info:fedora/', '')
-        add_related_item_node_for_collection! collection_druid
-      end
-    end
-
-    # Remove existing relatedItem entries for collections from descMetadata
-    def remove_related_item_nodes_for_collections!
-      doc.search('/mods:mods/mods:relatedItem[@type="host"]/mods:typeOfResource[@collection=\'yes\']', 'mods' => 'http://www.loc.gov/mods/v3').each do |node|
-        node.parent.remove
-      end
-    end
-
-    def add_related_item_node_for_collection!(collection_druid)
-      begin
-        collection_obj = Dor.find(collection_druid)
-      rescue ActiveFedora::ObjectNotFoundError
-        return nil
-      end
-
-      title_node         = Nokogiri::XML::Node.new('title', doc)
-      title_node.content = collection_obj.full_title
-
-      title_info_node = Nokogiri::XML::Node.new('titleInfo', doc)
-      title_info_node.add_child(title_node)
-
-      # e.g.:
-      #   <location>
-      #     <url>http://purl.stanford.edu/rh056sr3313</url>
-      #   </location>
-      loc_node = doc.create_element('location')
-      url_node = doc.create_element('url')
-      url_node.content = "https://#{Dor::Config.stacks.document_cache_host}/#{collection_druid.split(':').last}"
-      loc_node << url_node
-
-      type_node = Nokogiri::XML::Node.new('typeOfResource', doc)
-      type_node['collection'] = 'yes'
-
-      related_item_node = Nokogiri::XML::Node.new('relatedItem', doc)
-      related_item_node['type'] = 'host'
-
-      related_item_node.add_child(title_info_node)
-      related_item_node.add_child(loc_node)
-      related_item_node.add_child(type_node)
-
-      doc.root.add_child(related_item_node)
-    end
-
-    private
-
-    # Builds case-insensitive xpath translate function call that will match the attribute to a value
-    def ci_compare(attribute, value)
-      "translate(
-        @#{attribute},
-        'ABCDEFGHIJKLMNOPQRSTUVWXYZ',
-        'abcdefghijklmnopqrstuvwxyz'
-       ) = '#{value}' "
-    end
-  end
-end

data/lib/dor/services/purl_client.rb

@@ -1,42 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # Calls the purl service and returns the XML document
-  class PurlClient
-    def initialize(host:, pid:)
-      @host = host
-      @pid = pid
-    end
-
-    # Get XML from the purl service
-    # Fetches purl xml for a druid
-    # @raise [OpenURI::HTTPError]
-    # @return [Nokogiri::HTML::Document] parsed XML for the druid or an empty document if no purl is found
-    def fetch
-      handler = proc do |exception, attempt_number, total_delay|
-        # We assume a 404 means the document has never been published before and thus has no purl
-        Dor.logger.warn "[Attempt #{attempt_number}] GET #{url} -- #{exception.class}: #{exception.message}; #{total_delay} seconds elapsed."
-        raise exception unless exception.is_a? OpenURI::HTTPError
-        return Nokogiri::HTML::Document.new if exception.io.status.first == '404' # ["404", "Not Found"] from OpenURI::Meta.status
-      end
-
-      with_retries(max_retries: 3, base_sleep_seconds: 3, max_sleep_seconds: 5, handler: handler) do |attempt|
-        # If you change the method used for opening the webpage, you can change the :rescue param to handle the new method's errors
-        Dor.logger.debug "[Attempt #{attempt}] GET #{url}"
-        return Nokogiri::XML(OpenURI.open_uri(url))
-      end
-    end
-
-    private
-
-    # Take the and create the entire purl url that will usable for the open method in open-uri, returns http
-    # @return [String] the full url
-    def url
-      @url ||= "https://#{@host}/#{druid_without_prefix}.xml"
-    end
-
-    def druid_without_prefix
-      PidUtils.remove_druid_prefix(@pid)
-    end
-  end
-end

data/lib/dor/services/thumbnail_service.rb

@@ -1,59 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # Responsible for finding a path to a thumbnail based on the contentMetadata of an object
-  class ThumbnailService
-    # allow the mimetype attribute to be lower or camelcase when searching to make it more robust
-    MIME_TYPE_FINDER = "@mimetype='image/jp2' or @mimeType='image/jp2'"
-
-    # these are the finders we will use to search for a thumb resource in contentMetadata, they will be searched in the order provided, stopping when one is reached
-    THUMB_XPATH_FINDERS = [
-      # first find a file of mimetype jp2 explicitly marked as a thumb in the resource type and with a thumb=yes attribute
-      { image_type: 'local', finder: "/contentMetadata/resource[@type='thumb' and @thumb='yes']/file[#{MIME_TYPE_FINDER}]" },
-      # same thing for external files
-      { image_type: 'external', finder: "/contentMetadata/resource[@type='thumb' and @thumb='yes']/externalFile[#{MIME_TYPE_FINDER}]" },
-      # next find any image or page resource types with the thumb=yes attribute of mimetype jp2
-      { image_type: 'local', finder: "/contentMetadata/resource[(@type='page' or @type='image') and @thumb='yes']/file[#{MIME_TYPE_FINDER}]" },
-      # same thing for external file
-      { image_type: 'external', finder: "/contentMetadata/resource[(@type='page' or @type='image') and @thumb='yes']/externalFile[#{MIME_TYPE_FINDER}]" },
-      # next find a file of mimetype jp2 and resource type=thumb but not marked with the thumb directive
-      { image_type: 'local', finder: "/contentMetadata/resource[@type='thumb']/file[#{MIME_TYPE_FINDER}]" },
-      # same thing for external file
-      { image_type: 'external', finder: "/contentMetadata/resource[@type='thumb']/externalFile[#{MIME_TYPE_FINDER}]" },
-      # finally find the first page or image resource of mimetype jp2
-      { image_type: 'local', finder: "/contentMetadata/resource[@type='page' or @type='image']/file[#{MIME_TYPE_FINDER}]" },
-      # same thing for external file
-      { image_type: 'external', finder: "/contentMetadata/resource[@type='page' or @type='image']/externalFile[#{MIME_TYPE_FINDER}]" }
-    ].freeze
-
-    # @param [Dor::Item] object
-    def initialize(object)
-      @object = object
-    end
-
-    attr_reader :object
-
-    # @return [String] the computed thumb filename, with the druid prefix and a slash in front of it, e.g. oo000oo0001/filenamewith space.jp2
-    def thumb
-      return unless object.respond_to?(:contentMetadata) && object.contentMetadata.present?
-
-      cm = object.contentMetadata.ng_xml
-      thumb_image = nil
-
-      THUMB_XPATH_FINDERS.each do |search_path|
-        thumb_files = cm.xpath(search_path[:finder]) # look for a thumb
-        next if thumb_files.empty?
-
-        # if we find one, return the filename based on whether it is a local file or external file
-        thumb_image = if search_path[:image_type] == 'local'
-                        "#{PidUtils.remove_druid_prefix(object.pid)}/#{thumb_files[0]['id']}"
-                      else
-                        "#{PidUtils.remove_druid_prefix(thumb_files[0]['objectId'])}/#{thumb_files[0]['fileId']}"
-                      end
-        break # break out of the loop so we stop searching
-      end
-
-      thumb_image
-    end
-  end
-end