dor-services 6.6.2 → 6.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 68389296f84eb896aaff7f6916569e2fc2a39fee502218fa49cc6f96ccdf3793
-  data.tar.gz: d3bb35c8b8b3a5f3554731dca051bdd07a848b0e8721841411b40226e9a6e09e
+  metadata.gz: 49efa35d39eee0748121dad21f44e871d6ea38eff28dc1719f7d13061eaab269
+  data.tar.gz: d1e0418a853a958966e7247e9c2ca0ceaf0739784ee3da2e3a0ab1a4f8a2f67d
 SHA512:
-  metadata.gz: 2d921dfca2a8c757421582696d32c62059fb7723b6350a449a3bbc32cb8c8021b00ed663458cc9d8f97b0c7eef2dc22b0bb0443bc62e957c570bf73467754a1b
-  data.tar.gz: bdad1ef88346944263a4e39adb245c4843fbae6343ed00339c9889e806bb60008aaa02c8d3079eefae91ca973a9c0e8e5cbf0824982ca473968a643a3cc83ded
+  metadata.gz: 37ac6a9d98625fb2272a5e541d6cf99d4b3f6d7e615c000b22aa6c51b49ed917b912a30f77c293da356107f2418fa60d5c70a5a3b84f31bb52fa149b13f3601d
+  data.tar.gz: 2401c8bfef4aa070e5672227e67c53a29dee353dbe58fa73922f4a3409d1b23f0ad8baf9d5e30fa4c4e0d87062e0c38d37fbc0ac74cf31fd740d6d33fba84fea

data/config/dev_console_env.rb.example

@@ -37,7 +37,7 @@ Dor::Config.configure do
   end
 
   solr.url     'https://host/solr'
-  workflow.url     'https://host/workflow/'
+  workflow.url     'https://workflow.example.edu/'
   dor_services.url 'https://host/dor/v1'
 
   cleanup do

data/lib/dor-services.rb

@@ -91,6 +91,7 @@ module Dor
     autoload :ProcessableIndexer
     autoload :ReleasableIndexer
     autoload :WorkflowIndexer
+    autoload :WorkflowsIndexer
   end
 
   # datastreams
@@ -144,6 +145,7 @@ module Dor
       autoload :Collection
       autoload :AdminPolicyObject
       autoload :WorkflowObject
+      autoload :WorkflowSolrDocument
     end
   end
 
@@ -155,6 +157,7 @@ module Dor
     autoload :CreateWorkflowService
     autoload :CreativeCommonsLicenseService
     autoload :DatastreamBuilder
+    autoload :DecommissionService
     autoload :DigitalStacksService
     autoload :DublinCoreService
     autoload :FileMetadataMergeService

data/lib/dor/datastreams/workflow_ds.rb

@@ -35,7 +35,7 @@ module Dor
     end
 
     def get_workflow(wf, repo = 'dor')
-      xml = Dor::Config.workflow.client.get_workflow_xml(repo, pid, wf)
+      xml = Dor::Config.workflow.client.workflow_xml(repo, pid, wf)
       xml = Nokogiri::XML(xml)
       return nil if xml.xpath('workflow').length == 0
 
@@ -55,14 +55,16 @@ module Dor
     # service directly
     def content(refresh = false)
       @content = nil if refresh
-      @content ||= Dor::Config.workflow.client.get_workflow_xml 'dor', pid, nil
+      @content ||= Dor::Config.workflow.client.all_workflows_xml pid
     rescue Dor::WorkflowException => e
+      # TODO: I don't understand when this would be useful as this block ends up calling the workflow service too.
+      # Why not just raise an exception here?
       Dor.logger.warn "Unable to connect to the workflow service #{e}. Falling back to placeholder XML"
       xml = Nokogiri::XML(%(<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n<workflows objectId="#{pid}"/>))
       digital_object.datastreams.keys.each do |dsid|
         next unless dsid =~ /WF$/
 
-        ds_content = Nokogiri::XML(Dor::Config.workflow.client.get_workflow_xml('dor', pid, dsid))
+        ds_content = Nokogiri::XML(Dor::Config.workflow.client.workflow_xml('dor', pid, dsid))
         xml.root.add_child(ds_content.root)
       end
       @content ||= xml.to_xml
@@ -83,7 +85,7 @@ module Dor
     end
 
     def to_solr(solr_doc = {}, *_args)
-      # noop - indexing is done by the WorkflowIndexer
+      # noop - indexing is done by the WorkflowsIndexer
       solr_doc
     end
 

data/lib/dor/exceptions.rb

@@ -11,6 +11,9 @@ module Dor
   # rubocop:disable Lint/InheritException
   # See https://github.com/rubocop-hq/rubocop/issues/6770
   class VersionAlreadyOpenError < Exception; end
+
+  # Raised when we can't get a response from the catalog
+  class BadResponseFromCatalog < Exception; end
   # rubocop:enable Lint/InheritException
 
   class DuplicateIdError < RuntimeError

data/lib/dor/indexers/workflow_indexer.rb

@@ -3,25 +3,82 @@
 module Dor
   # Indexes the objects position in workflows
   class WorkflowIndexer
-    include SolrDocHelper
+    ERROR_OMISSION = '... (continued)'
+    private_constant :ERROR_OMISSION
 
-    attr_reader :resource
-    def initialize(resource:)
-      @resource = resource
+    # see https://lucene.apache.org/core/7_3_1/core/org/apache/lucene/util/BytesRefHash.MaxBytesLengthExceededException.html
+    MAX_ERROR_LENGTH = 32_768 - 2 - ERROR_OMISSION.length
+    private_constant :MAX_ERROR_LENGTH
+
+    # @param [Dor::WorkflowDocument] document the workflow document to index
+    def initialize(document:)
+      @document = document
     end
 
-    # @return [Hash] the partial solr document for workflow concerns
+    # @return [Hash] the partial solr document for the workflow document
     def to_solr
-      {}.tap do |solr_doc|
-        workflows.each { |wf| solr_doc = wf.to_solr(solr_doc) }
+      WorkflowSolrDocument.new do |solr_doc|
+        wf_name = document.workflowId.first
+        solr_doc.name = wf_name
+        errors = processes.count(&:error?)
+
+        repo = document.repository.first
+        solr_doc.status = [wf_name, workflow_status, errors, repo].join('|')
+
+        processes.each do |process|
+          index_process(solr_doc, wf_name, process)
+        end
       end
     end
 
     private
 
-    # @return [Array<Dor::WorkflowDocument>]
-    def workflows
-      resource.workflows.workflows
+    attr_reader :document
+    delegate :processes, to: :document
+
+    def index_process(solr_doc, wf_name, process)
+      return unless process.status.present?
+
+      # add a record of the robot having operated on this item, so we can track robot activity
+      solr_doc.add_process_time(wf_name, process.name, Time.parse(process.date_time)) if process_has_time?(process)
+
+      index_error_message(solr_doc, wf_name, process)
+
+      # workflow name, process status then process name
+      solr_doc.add_wps("#{wf_name}:#{process.status}", "#{wf_name}:#{process.status}:#{process.name}")
+
+      # workflow name, process name then process status
+      solr_doc.add_wps("#{wf_name}:#{process.name}", "#{wf_name}:#{process.name}:#{process.status}")
+
+      # process status, workflowname then process name
+      solr_doc.add_swp(process.status.to_s, "#{process.status}:#{wf_name}", "#{process.status}:#{wf_name}:#{process.name}")
+      return if process.state == process.status
+
+      solr_doc.add_wsp("#{wf_name}:#{process.state}:#{process.name}")
+      solr_doc.add_wps("#{wf_name}:#{process.name}:#{process.state}")
+      solr_doc.add_swp(process.state.to_s, "#{process.state}:#{wf_name}", "#{process.state}:#{wf_name}:#{process.name}")
+    end
+
+    def process_has_time?(process)
+      !process.date_time.blank? && process.status && (process.status == 'completed' || process.status == 'error')
+    end
+
+    def workflow_status
+      return 'empty' if processes.empty?
+
+      workflow_should_show_completed?(processes) ? 'completed' : 'active'
+    end
+
+    def workflow_should_show_completed?(processes)
+      processes.all? { |p| ['skipped', 'completed', '', nil].include?(p.status) }
+    end
+
+    # index the error message without the druid so we hopefully get some overlap
+    # truncate to avoid org.apache.lucene.util.BytesRefHash$MaxBytesLengthExceededException
+    def index_error_message(solr_doc, wf_name, process)
+      return unless process.error_message
+
+      solr_doc.error = "#{wf_name}:#{process.name}:#{process.error_message}".truncate(MAX_ERROR_LENGTH, omission: ERROR_OMISSION)
     end
   end
 end

data/lib/dor/indexers/workflows_indexer.rb

@@ -0,0 +1,28 @@
+# frozen_string_literal: true
+
+module Dor
+  # Indexes the objects position in workflows
+  class WorkflowsIndexer
+    attr_reader :resource
+    def initialize(resource:)
+      @resource = resource
+    end
+
+    # @return [Hash] the partial solr document for workflow concerns
+    def to_solr
+      WorkflowSolrDocument.new do |combined_doc|
+        workflows.each do |wf|
+          doc = WorkflowIndexer.new(document: wf).to_solr
+          combined_doc.merge!(doc)
+        end
+      end.to_h
+    end
+
+    private
+
+    # @return [Array<Dor::WorkflowDocument>]
+    def workflows
+      resource.workflows.workflows
+    end
+  end
+end

data/lib/dor/models/admin_policy_object.rb

@@ -15,7 +15,7 @@ module Dor
       EditableIndexer,
       IdentifiableIndexer,
       ProcessableIndexer,
-      WorkflowIndexer
+      WorkflowsIndexer
     )
 
     CREATIVE_COMMONS_USE_LICENSES = ActiveSupport::Deprecation::DeprecatedConstantProxy.new('CREATIVE_COMMONS_USE_LICENSES', 'Dor::CreativeCommonsLicenseService')

data/lib/dor/models/collection.rb

@@ -12,7 +12,7 @@ module Dor
       IdentifiableIndexer,
       ProcessableIndexer,
       ReleasableIndexer,
-      WorkflowIndexer
+      WorkflowsIndexer
     )
   end
 end

data/lib/dor/models/concerns/contentable.rb

@@ -170,27 +170,16 @@ module Dor
     # Clears RELS-EXT relationships, sets the isGovernedBy relationship to the SDR Graveyard APO
     # @param [String] tag optional String of text that is concatenated to the identityMetadata/tag "Decommissioned : "
     def decommission(tag)
-      # remove isMemberOf and isMemberOfCollection relationships
-      clear_relationship :is_member_of
-      clear_relationship :is_member_of_collection
-      # remove isGovernedBy relationship
-      clear_relationship :is_governed_by
-      # add isGovernedBy to graveyard APO druid:sw909tc7852
-      # SEARCH BY dc title for 'SDR Graveyard'
-      add_relationship :is_governed_by, ActiveFedora::Base.find(Dor::SearchService.sdr_graveyard_apo_druid)
-      # eliminate contentMetadata. set it to <contentMetadata/> ?
-      contentMetadata.content = '<contentMetadata/>'
-      # eliminate rightsMetadata. set it to <rightsMetadata/> ?
-      rightsMetadata.content = '<rightsMetadata/>'
-      TagService.add self, "Decommissioned : #{tag}"
+      DecommissionService.new(self).decommission(tag)
     end
+    deprecation_deprecate decommission: 'Use DecommissionService#decommission instead'
 
-    # TODO: Move to Dor-Utils.
     # Adds a RELS-EXT constituent relationship to the given druid
     # @param [String] druid the parent druid of the constituent relationship
     #   e.g.: <fedora:isConstituentOf rdf:resource="info:fedora/druid:hj097bm8879" />
     def add_constituent(druid)
       add_relationship :is_constituent_of, ActiveFedora::Base.find(druid)
     end
+    deprecation_deprecate add_constituent: 'Use add_relationship :is_constituent_of instead'
   end
 end

data/lib/dor/models/concerns/versionable.rb

@@ -49,9 +49,9 @@ module Dor
     # @return [Boolean] true if the object is in a state that allows it to be modified.
     #  States that will allow modification are: has not been submitted for accessioning, has an open version or has sdr-ingest set to hold
     def allows_modification?
-      if Dor::Config.workflow.client.get_lifecycle('dor', pid, 'submitted') &&
+      if Dor::Config.workflow.client.lifecycle('dor', pid, 'submitted') &&
          !VersionService.new(self).open? &&
-         Dor::Config.workflow.client.get_workflow_status('dor', pid, 'accessionWF', 'sdr-ingest-transfer') != 'hold'
+         Dor::Config.workflow.client.workflow_status('dor', pid, 'accessionWF', 'sdr-ingest-transfer') != 'hold'
         false
       else
         true

data/lib/dor/models/item.rb

@@ -19,7 +19,7 @@ module Dor
       IdentifiableIndexer,
       ProcessableIndexer,
       ReleasableIndexer,
-      WorkflowIndexer
+      WorkflowsIndexer
     )
 
     has_metadata name: 'technicalMetadata', type: TechnicalMetadataDS, label: 'Technical Metadata', control_group: 'M'

data/lib/dor/models/set.rb

@@ -12,7 +12,7 @@ module Dor
       DescribableIndexer,
       IdentifiableIndexer,
       ProcessableIndexer,
-      WorkflowIndexer
+      WorkflowsIndexer
     )
   end
 end

data/lib/dor/models/workflow_object.rb

@@ -15,7 +15,7 @@ module Dor
       DescribableIndexer,
       IdentifiableIndexer,
       ProcessableIndexer,
-      WorkflowIndexer
+      WorkflowsIndexer
     )
 
     def self.find_by_name(name)

data/lib/dor/models/workflow_solr_document.rb

@@ -0,0 +1,93 @@
+# frozen_string_literal: true
+
+module Dor
+  # Represents that part of the solr document that holds workflow data
+  class WorkflowSolrDocument
+    WORKFLOW_SOLR = 'wf_ssim'
+    # field that indexes workflow name, process status then process name
+    WORKFLOW_WPS_SOLR = 'wf_wps_ssim'
+    # field that indexes workflow name, process name then process status
+    WORKFLOW_WSP_SOLR = 'wf_wsp_ssim'
+    # field that indexes process status, workflowname then process name
+    WORKFLOW_SWP_SOLR = 'wf_swp_ssim'
+    WORKFLOW_ERROR_SOLR = 'wf_error_ssim'
+    WORKFLOW_STATUS_SOLR = 'workflow_status_ssim'
+
+    KEYS_TO_MERGE = [
+      WORKFLOW_SOLR,
+      WORKFLOW_WPS_SOLR,
+      WORKFLOW_WSP_SOLR,
+      WORKFLOW_SWP_SOLR,
+      WORKFLOW_STATUS_SOLR,
+      WORKFLOW_ERROR_SOLR
+    ].freeze
+
+    def initialize
+      @data = empty_document
+      yield self if block_given?
+    end
+
+    def name=(wf_name)
+      data[WORKFLOW_SOLR] += [wf_name]
+      data[WORKFLOW_WPS_SOLR] += [wf_name]
+      data[WORKFLOW_WSP_SOLR] += [wf_name]
+    end
+
+    def status=(status)
+      data[WORKFLOW_STATUS_SOLR] += [status]
+    end
+
+    def error=(message)
+      data[WORKFLOW_ERROR_SOLR] += [message]
+    end
+
+    # Add to the field that indexes workflow name, process status then process name
+    def add_wps(*messages)
+      data[WORKFLOW_WPS_SOLR] += messages
+    end
+
+    # Add to the field that indexes workflow name, process name then process status
+    def add_wsp(*messages)
+      data[WORKFLOW_WSP_SOLR] += messages
+    end
+
+    # Add to the field that indexes process status, workflow name then process name
+    def add_swp(*messages)
+      data[WORKFLOW_SWP_SOLR] += messages
+    end
+
+    # Add the processes data_time attribute to the solr document
+    # @param [String] wf_name
+    # @param [String] process_name
+    # @param [Time] time
+    def add_process_time(wf_name, process_name, time)
+      data["wf_#{wf_name}_#{process_name}_dttsi"] = time.utc.iso8601
+    end
+
+    def to_h
+      KEYS_TO_MERGE.each { |k| data[k].uniq! }
+      data
+    end
+
+    delegate :except, :[], to: :data
+
+    # @param [WorkflowSolrDocument] doc
+    def merge!(doc)
+      # This is going to get the date fields, e.g. `wf_assemblyWF_jp2-create_dttsi'
+      @data.merge!(doc.except(*KEYS_TO_MERGE))
+
+      # Combine the non-unique fields together
+      KEYS_TO_MERGE.each do |k|
+        data[k] += doc[k]
+      end
+    end
+
+    private
+
+    attr_reader :data
+
+    def empty_document
+      KEYS_TO_MERGE.each_with_object({}) { |k, obj| obj[k] = [] }
+    end
+  end
+end

data/lib/dor/services/cleanup_reset_service.rb

@@ -19,7 +19,7 @@ module Dor
       last_version = druid_obj.current_version.to_i
 
       # if the current version is still open, avoid this versioned directory
-      last_version -= 1 if Dor::Config.workflow.client.get_lifecycle('dor', druid, 'accessioned').nil?
+      last_version -= 1 if Dor::Config.workflow.client.lifecycle('dor', druid, 'accessioned').nil?
       last_version
     end
 

data/lib/dor/services/cleanup_service.rb

@@ -70,7 +70,7 @@ module Dor
 
     def self.remove_active_workflows(druid)
       %w(dor sdr).each do |repo|
-        dor_wfs = Dor::Config.workflow.client.get_workflows(druid, repo)
+        dor_wfs = Dor::Config.workflow.client.workflows(druid, repo)
         dor_wfs.each { |wf| Dor::Config.workflow.client.delete_workflow(repo, druid, wf) }
       end
     end

data/lib/dor/services/decommission_service.rb

@@ -0,0 +1,31 @@
+# frozen_string_literal: true
+
+module Dor
+  # Responsible for decommissioning objects
+  class DecommissionService
+    # @param [Dor::Item] object
+    def initialize(object)
+      @object = object
+    end
+
+    attr_reader :object
+
+    # Clears RELS-EXT relationships, sets the isGovernedBy relationship to the SDR Graveyard APO
+    # @param [String] tag optional String of text that is concatenated to the identityMetadata/tag "Decommissioned : "
+    def decommission(tag)
+      # remove isMemberOf and isMemberOfCollection relationships
+      object.clear_relationship :is_member_of
+      object.clear_relationship :is_member_of_collection
+      # remove isGovernedBy relationship
+      object.clear_relationship :is_governed_by
+      # add isGovernedBy to graveyard APO druid:sw909tc7852
+      # SEARCH BY dc title for 'SDR Graveyard'
+      object.add_relationship :is_governed_by, ActiveFedora::Base.find(Dor::SearchService.sdr_graveyard_apo_druid)
+      # eliminate contentMetadata. set it to <contentMetadata/> ?
+      object.contentMetadata.content = '<contentMetadata/>'
+      # eliminate rightsMetadata. set it to <rightsMetadata/> ?
+      object.rightsMetadata.content = '<rightsMetadata/>'
+      TagService.add object, "Decommissioned : #{tag}"
+    end
+  end
+end

data/lib/dor/services/metadata_handlers/catalog_handler.rb

@@ -2,12 +2,15 @@
 
 require 'rest-client'
 
-handler = Class.new do
+class CatalogHandler
   def fetch(prefix, identifier)
     client = RestClient::Resource.new(Dor::Config.metadata.catalog.url,
                                       Dor::Config.metadata.catalog.user,
                                       Dor::Config.metadata.catalog.pass)
-    client["?#{prefix.chomp}=#{identifier.chomp}"].get
+    params = "?#{prefix.chomp}=#{identifier.chomp}"
+    client[params].get
+  rescue RestClient::Exception => e
+    raise BadResponseFromCatalog, "#{e.class} - when contacting (with BasicAuth hidden): #{Dor::Config.metadata.catalog.url}#{params}"
   end
 
   def label(metadata)
@@ -20,5 +23,3 @@ handler = Class.new do
     %w(catkey barcode)
   end
 end
-
-Dor::MetadataService.register(handler)

data/lib/dor/services/metadata_service.rb

@@ -1,37 +1,15 @@
 # frozen_string_literal: true
 
 require 'cache'
+require 'dor/services/metadata_handlers/catalog_handler'
 
 module Dor
   class MetadataError < RuntimeError; end
 
-  #  class MetadataHandler
-  #
-  #    def fetch(prefix, identifier)
-  #      ### Return metadata for prefix/identifier combo
-  #    end
-  #
-  #    def label(metadata)
-  #      ### Return a Fedora-compatible label from the metadata format returned by #fetch
-  #    end
-  #
-  #  end
-
   class MetadataService
     class << self
       @@cache = Cache.new(nil, nil, 250, 300)
 
-      def register(handler_class)
-        %w(fetch label prefixes).each do |method|
-          raise TypeError, "Metadata handlers must define ##{method}" unless handler_class.instance_methods.include?(method) || handler_class.instance_methods.include?(method.to_sym)
-        end
-        handler = handler_class.new
-        handler.prefixes.each do |prefix|
-          handlers[prefix.to_sym] = handler
-        end
-        handler
-      end
-
       def known_prefixes
         handlers.keys
       end
@@ -62,7 +40,7 @@ module Dor
 
       def handler_for(prefix)
         handler = handlers[prefix.to_sym]
-        raise MetadataError, "Unkown metadata prefix: #{prefix}" if handler.nil?
+        raise MetadataError, "Unknown metadata prefix: #{prefix}" if handler.nil?
 
         handler
       end
@@ -70,12 +48,17 @@ module Dor
       private
 
       def handlers
-        @handlers ||= {}
+        @handlers ||= {}.tap do |md_handlers|
+          # There's only one. If additional handlers are added, will need to be registered here.
+          register(CatalogHandler.new, md_handlers)
+        end
+      end
+
+      def register(handler, md_handlers)
+        handler.prefixes.each do |prefix|
+          md_handlers[prefix.to_sym] = handler
+        end
       end
     end
   end
 end
-
-Dir[File.join(File.dirname(__FILE__), 'metadata_handlers', '*.rb')].each do |handler_file|
-  load handler_file
-end

data/lib/dor/services/status_service.rb

@@ -76,7 +76,7 @@ module Dor
     end
 
     def milestones
-      @milestones ||= Dor::Config.workflow.client.get_milestones('dor', work.pid)
+      @milestones ||= Dor::Config.workflow.client.milestones('dor', work.pid)
     end
 
     private

data/lib/dor/services/version_service.rb

@@ -24,9 +24,9 @@ module Dor
     def open(opts = {})
       # During local development, we need a way to open a new version even if the object has not been accessioned.
       raise(Dor::Exception, 'Object net yet accessioned') unless
-        opts[:assume_accessioned] || Dor::Config.workflow.client.get_lifecycle('dor', work.pid, 'accessioned')
+        opts[:assume_accessioned] || Dor::Config.workflow.client.lifecycle('dor', work.pid, 'accessioned')
       raise Dor::VersionAlreadyOpenError, 'Object already opened for versioning' if open?
-      raise Dor::Exception, 'Object currently being accessioned' if Dor::Config.workflow.client.get_active_lifecycle('dor', work.pid, 'submitted')
+      raise Dor::Exception, 'Object currently being accessioned' if Dor::Config.workflow.client.active_lifecycle('dor', work.pid, 'submitted')
 
       sdr_version = Sdr::Client.current_version work.pid
 
@@ -67,14 +67,14 @@ module Dor
 
       raise Dor::Exception, 'latest version in versionMetadata requires tag and description before it can be closed' unless work.versionMetadata.current_version_closeable?
       raise Dor::Exception, 'Trying to close version on an object not opened for versioning' unless open?
-      raise Dor::Exception, 'accessionWF already created for versioned object' if Dor::Config.workflow.client.get_active_lifecycle('dor', work.pid, 'submitted')
+      raise Dor::Exception, 'accessionWF already created for versioned object' if Dor::Config.workflow.client.active_lifecycle('dor', work.pid, 'submitted')
 
       Dor::Config.workflow.client.close_version 'dor', work.pid, opts.fetch(:start_accession, true) # Default to creating accessionWF when calling close_version
     end
 
     # @return [Boolean] true if 'opened' lifecycle is active, false otherwise
     def open?
-      return true if Dor::Config.workflow.client.get_active_lifecycle('dor', work.pid, 'opened')
+      return true if Dor::Config.workflow.client.active_lifecycle('dor', work.pid, 'opened')
 
       false
     end

data/lib/dor/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Dor
-  VERSION = '6.6.2'
+  VERSION = '6.7.0'
 end

data/lib/dor/workflow/document.rb

@@ -4,21 +4,8 @@ module Dor
   module Workflow
     class Document
       extend Deprecation
-      include SolrDocHelper
       include ::OM::XML::Document
 
-      ERROR_OMISSION = '... (continued)'
-      private_constant :ERROR_OMISSION
-
-      # see https://lucene.apache.org/core/7_3_1/core/org/apache/lucene/util/BytesRefHash.MaxBytesLengthExceededException.html
-      MAX_ERROR_LENGTH = 32_768 - 2 - ERROR_OMISSION.length
-      private_constant :MAX_ERROR_LENGTH
-
-      WF_SOLR_TYPE = :string
-      private_constant :WF_SOLR_TYPE
-      WF_SOLR_ATTRS = [:symbol].freeze
-      private_constant :WF_SOLR_ATTRS
-
       set_terminology do |t|
         t.root(path: 'workflow')
         t.repository(path: { attribute: 'repository' })
@@ -93,69 +80,9 @@ module Dor
           end
       end
 
-      def workflow_should_show_completed?(processes)
-        processes.all? { |p| ['skipped', 'completed', '', nil].include?(p.status) }
-      end
-
-      def to_solr(solr_doc = {}, *_args)
-        wf_name = workflowId.first
-        repo = repository.first
-
-        add_solr_value(solr_doc, 'wf',     wf_name, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-        add_solr_value(solr_doc, 'wf_wps', wf_name, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-        add_solr_value(solr_doc, 'wf_wsp', wf_name, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-        status = processes.empty? ? 'empty' : (workflow_should_show_completed?(processes) ? 'completed' : 'active')
-        errors = processes.count(&:error?)
-        add_solr_value(solr_doc, 'workflow_status', [wf_name, status, errors, repo].join('|'), WF_SOLR_TYPE, WF_SOLR_ATTRS)
-
-        processes.each do |process|
-          next unless process.status.present?
-
-          # add a record of the robot having operated on this item, so we can track robot activity
-          if !process.date_time.blank? && process.status && (process.status == 'completed' || process.status == 'error')
-            solr_doc["wf_#{wf_name}_#{process.name}_dttsi"] = Time.parse(process.date_time).utc.iso8601
-          end
-
-          index_error_message(solr_doc, wf_name, process)
-
-          add_solr_value(solr_doc, 'wf_wsp', "#{wf_name}:#{process.status}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_wsp', "#{wf_name}:#{process.status}:#{process.name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_wps', "#{wf_name}:#{process.name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_wps', "#{wf_name}:#{process.name}:#{process.status}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', process.status.to_s, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', "#{process.status}:#{wf_name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', "#{process.status}:#{wf_name}:#{process.name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          next unless process.state != process.status
-
-          add_solr_value(solr_doc, 'wf_wsp', "#{wf_name}:#{process.state}:#{process.name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_wps', "#{wf_name}:#{process.name}:#{process.state}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', process.state.to_s, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', "#{process.state}:#{wf_name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-          add_solr_value(solr_doc, 'wf_swp', "#{process.state}:#{wf_name}:#{process.name}", WF_SOLR_TYPE, WF_SOLR_ATTRS)
-        end
-
-        solr_doc[Solrizer.solr_name('wf_wps', :symbol)]&.uniq!
-        solr_doc[Solrizer.solr_name('wf_wsp', :symbol)]&.uniq!
-        solr_doc[Solrizer.solr_name('wf_swp', :symbol)]&.uniq!
-        solr_doc['workflow_status']&.uniq!
-
-        solr_doc
-      end
-
       def inspect
         "#<#{self.class.name}:#{object_id}>"
       end
-
-      private
-
-      # index the error message without the druid so we hopefully get some overlap
-      # truncate to avoid org.apache.lucene.util.BytesRefHash$MaxBytesLengthExceededException
-      def index_error_message(solr_doc, wf_name, process)
-        return unless process.error_message
-
-        error_message = "#{wf_name}:#{process.name}:#{process.error_message}".truncate(MAX_ERROR_LENGTH, omission: ERROR_OMISSION)
-        add_solr_value(solr_doc, 'wf_error', error_message, WF_SOLR_TYPE, WF_SOLR_ATTRS)
-      end
     end
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: dor-services
 version: !ruby/object:Gem::Version
-  version: 6.6.2
+  version: 6.7.0
 platform: ruby
 authors:
 - Michael Klein
@@ -14,7 +14,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-03-13 00:00:00.000000000 Z
+date: 2019-04-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: active-fedora
@@ -290,20 +290,6 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.6'
-- !ruby/object:Gem::Dependency
-  name: uuidtools
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 2.1.4
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 2.1.4
 - !ruby/object:Gem::Dependency
   name: retries
   requirement: !ruby/object:Gem::Requirement
@@ -344,20 +330,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.0'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 2.0.1
+        version: '2.11'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.0'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 2.0.1
+        version: '2.11'
 - !ruby/object:Gem::Dependency
   name: druid-tools
   requirement: !ruby/object:Gem::Requirement
@@ -624,6 +604,7 @@ files:
 - lib/dor/indexers/processable_indexer.rb
 - lib/dor/indexers/releasable_indexer.rb
 - lib/dor/indexers/workflow_indexer.rb
+- lib/dor/indexers/workflows_indexer.rb
 - lib/dor/models/abstract.rb
 - lib/dor/models/admin_policy_object.rb
 - lib/dor/models/agreement.rb
@@ -647,6 +628,7 @@ files:
 - lib/dor/models/item.rb
 - lib/dor/models/set.rb
 - lib/dor/models/workflow_object.rb
+- lib/dor/models/workflow_solr_document.rb
 - lib/dor/rest_resource_factory.rb
 - lib/dor/services/ability.rb
 - lib/dor/services/cleanup_reset_service.rb
@@ -654,6 +636,7 @@ files:
 - lib/dor/services/create_workflow_service.rb
 - lib/dor/services/creative_commons_license_service.rb
 - lib/dor/services/datastream_builder.rb
+- lib/dor/services/decommission_service.rb
 - lib/dor/services/digital_stacks_service.rb
 - lib/dor/services/dublin_core_service.rb
 - lib/dor/services/file_metadata_merge_service.rb
@@ -667,7 +650,6 @@ files:
 - lib/dor/services/public_desc_metadata_service.rb
 - lib/dor/services/public_xml_service.rb
 - lib/dor/services/publish_metadata_service.rb
-- lib/dor/services/registration_service.rb
 - lib/dor/services/release_tag_service.rb
 - lib/dor/services/reset_workspace_service.rb
 - lib/dor/services/sdr_ingest_service.rb

data/lib/dor/services/registration_service.rb

@@ -1,202 +0,0 @@
-# frozen_string_literal: true
-
-require 'uuidtools'
-
-module Dor
-  class RegistrationService
-    class << self
-      # @TODO: Why isn't all this logic in, for example, Dor::Item.create? or Dor::Base.create? or Dor::Creatable.create?
-      # @TODO: these duplicate checks could be combined into 1 query
-
-      # @param [String] pid an ID to check, if desired.  If not passed (or nil), a new ID is minted
-      # @return [String] a pid you can use immidately, either freshly minted or your checked value
-      # @raise [Dor::DuplicateIdError]
-      def unduplicated_pid(pid = nil)
-        return Dor::SuriService.mint_id unless pid
-
-        existing_pid = SearchService.query_by_id(pid).first
-        raise Dor::DuplicateIdError.new(existing_pid), "An object with the PID #{pid} has already been registered." unless existing_pid.nil?
-
-        pid
-      end
-
-      # @param [String] source_id_string a fully qualified source:val or empty string
-      # @return [String] the same qualified source:id for immediate use
-      # @raise [Dor::DuplicateIdError]
-      def check_source_id(source_id_string)
-        return '' if source_id_string == ''
-        unless SearchService.query_by_id(source_id_string.to_s).first.nil?
-          raise Dor::DuplicateIdError.new(source_id_string), "An object with the source ID '#{source_id_string}' has already been registered."
-        end
-
-        source_id_string
-      end
-
-      # @param [Hash{Symbol => various}] params
-      # @option params [String] :object_type required
-      # @option params [String] :label required
-      # @option params [String] :admin_policy required
-      # @option params [String] :metadata_source
-      # @option params [String] :rights
-      # @option params [String] :collection
-      # @option params [Hash{String => String}] :source_id Primary ID from another system, max one key/value pair!
-      # @option params [Hash] :other_ids including :uuid if known
-      # @option params [String] :pid Fully qualified PID if you don't want one generated for you
-      # @option params [Integer] :workflow_priority]
-      # @option params [Array<String>] :seed_datastream datastream_names
-      # @option params [Array<String>] :initiate_workflow workflow_ids
-      # @option params [Array] :tags
-      def register_object(params = {})
-        %i[object_type label].each do |required_param|
-          raise Dor::ParameterError, "#{required_param.inspect} must be specified in call to #{name}.register_object" unless params[required_param]
-        end
-        metadata_source = params[:metadata_source]
-        raise Dor::ParameterError, "label cannot be empty to call #{name}.register_object" if params[:label].length < 1 && %w[label none].include?(metadata_source)
-
-        object_type = params[:object_type]
-        item_class = Dor.registered_classes[object_type]
-        raise Dor::ParameterError, "Unknown item type: '#{object_type}'" if item_class.nil?
-
-        # content_model = params[:content_model]
-        # parent        = params[:parent]
-        label         = params[:label]
-        source_id     = params[:source_id] || {}
-        other_ids     = params[:other_ids] || {}
-        tags          = params[:tags] || []
-        collection    = params[:collection]
-
-        # Check for sourceId conflict *before* potentially minting PID
-        source_id_string = check_source_id [source_id.keys.first, source_id[source_id.keys.first]].compact.join(':')
-        pid = unduplicated_pid(params[:pid])
-
-        raise ArgumentError, ":source_id Hash can contain at most 1 pair: recieved #{source_id.size}" if source_id.size > 1
-
-        rights = nil
-        if params[:rights]
-          rights = params[:rights]
-          raise Dor::ParameterError, "Unknown rights setting '#{rights}' when calling #{name}.register_object" unless rights == 'default' || RightsMetadataDS.valid_rights_type?(rights)
-        end
-
-        other_ids[:uuid] = UUIDTools::UUID.timestamp_create.to_s if (other_ids.key?(:uuid) || other_ids.key?('uuid')) == false
-        apo_object = Dor.find(params[:admin_policy])
-        new_item = item_class.new(pid: pid)
-        new_item.label = label.length > 254 ? label[0, 254] : label
-        idmd = new_item.identityMetadata
-        idmd.sourceId = source_id_string
-        idmd.add_value(:objectId, pid)
-        idmd.add_value(:objectCreator, 'DOR')
-        idmd.add_value(:objectLabel, label)
-        idmd.add_value(:objectType, object_type)
-        other_ids.each_pair { |name, value| idmd.add_otherId("#{name}:#{value}") }
-        tags.each { |tag| idmd.add_value(:tag, tag) }
-        new_item.admin_policy_object = apo_object
-
-        apo_object.administrativeMetadata.ng_xml.xpath('/administrativeMetadata/relationships/*').each do |rel|
-          short_predicate = ActiveFedora::RelsExtDatastream.short_predicate rel.namespace.href + rel.name
-          if short_predicate.nil?
-            ix = 0
-            ix += 1 while ActiveFedora::Predicates.predicate_mappings[rel.namespace.href].key?(short_predicate = :"extra_predicate_#{ix}")
-            ActiveFedora::Predicates.predicate_mappings[rel.namespace.href][short_predicate] = rel.name
-          end
-          new_item.add_relationship short_predicate, rel['rdf:resource']
-        end
-        new_item.add_collection(collection) if collection
-        if rights && %w(item collection).include?(object_type)
-          rights_xml = apo_object.defaultObjectRights.ng_xml
-          new_item.datastreams['rightsMetadata'].content = rights_xml.to_s
-          new_item.set_read_rights(rights) unless rights == 'default' # already defaulted to default!
-        end
-        # create basic mods from the label
-        build_desc_metadata_from_label(new_item, label) if metadata_source == 'label'
-
-        workflow_priority = params[:workflow_priority] ? params[:workflow_priority].to_i : 0
-
-        seed_datastreams(Array(params[:seed_datastream]), new_item)
-        initiate_workflow(workflows: Array(params[:initiate_workflow]), item: new_item, priority: workflow_priority)
-
-        new_item.class.ancestors.select { |x| x.respond_to?(:to_class_uri) && x != ActiveFedora::Base }.each do |parent_class|
-          new_item.add_relationship(:has_model, parent_class.to_class_uri)
-        end
-
-        new_item.save
-        new_item
-      end
-
-      # @param [Hash] params
-      # @see register_object similar but different
-      def create_from_request(params)
-        other_ids = Array(params[:other_id]).map do |id|
-          if id =~ /^symphony:(.+)$/
-            "#{$1.length < 14 ? 'catkey' : 'barcode'}:#{$1}"
-          else
-            id
-          end
-        end
-
-        if params[:label] == ':auto'
-          params.delete(:label)
-          params.delete('label')
-          metadata_id = Dor::MetadataService.resolvable(other_ids).first
-          params[:label] = Dor::MetadataService.label_for(metadata_id)
-        end
-
-        dor_params = {
-          pid: params[:pid],
-          admin_policy: params[:admin_policy],
-          content_model: params[:model],
-          label: params[:label],
-          object_type: params[:object_type],
-          other_ids: ids_to_hash(other_ids),
-          parent: params[:parent],
-          source_id: ids_to_hash(params[:source_id]),
-          tags: params[:tag] || [],
-          seed_datastream: params[:seed_datastream],
-          initiate_workflow: Array(params[:initiate_workflow]) + Array(params[:workflow_id]),
-          rights: params[:rights],
-          metadata_source: params[:metadata_source],
-          collection: params[:collection],
-          workflow_priority: params[:workflow_priority]
-        }
-        dor_params.delete_if { |_k, v| v.nil? }
-
-        dor_obj = register_object(dor_params)
-        pid = dor_obj.pid
-        location = URI.parse(Dor::Config.fedora.safeurl.sub(/\/*$/, '/')).merge("objects/#{pid}").to_s
-        dor_params.dup.merge(location: location, pid: pid)
-      end
-
-      private
-
-      def ids_to_hash(ids)
-        return nil if ids.nil?
-
-        Hash[Array(ids).map { |id| id.split(':', 2) }]
-      end
-
-      def seed_datastreams(names, item)
-        names.each do |datastream_name|
-          item.build_datastream(datastream_name)
-        end
-      end
-
-      def initiate_workflow(workflows:, item:, priority:)
-        workflows.each do |workflow_id|
-          Dor::CreateWorkflowService.create_workflow(item, name: workflow_id,
-                                                           create_ds: !item.new_record?,
-                                                           priority: priority)
-        end
-      end
-
-      def build_desc_metadata_from_label(new_item, label)
-        builder = Nokogiri::XML::Builder.new do |xml|
-          xml.mods(Dor::DescMetadataDS::MODS_HEADER_CONFIG) do
-            xml.titleInfo do
-              xml.title label
-            end
-          end
-        end
-        new_item.descMetadata.content = builder.to_xml
-      end
-    end
-  end
-end