dor-services 8.3.0 → 8.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 122c50609b2053d3b4083b985fcc128fd89726687a5bdf1230a4bbfafb55355f
-  data.tar.gz: 5c375fe22a648e6a697902a056b6752962cd78d7ecaa6c58d3b7c7049b1181fc
+  metadata.gz: 211ce0eefc758823b5dced495a74a980025d82884639a720a30ff97f20179cf6
+  data.tar.gz: f21f78c40a74231051b08762cea0c4c49f7cb47663ec5c6b9f24bc4189fef4f0
 SHA512:
-  metadata.gz: 073f7a22dfe87e282a156ea6f1b14ad1971bafa79e29334d95da14a89c094b0c0e68f7b706519b0ea5d928de665383846c33563e8f4c98a6d9e7b291192c07aa
-  data.tar.gz: 8c5424ada3a28d7acee8061d398a436f8cc9b214a299d40e4ad15a8337967ff176f5a376b59eed55239a0e412c0f99ef0870b3865e515627fd2a12f655bfe338
+  metadata.gz: 36a41ca02293ad622536768705b960ffa68bacb291cf6ef787b3181fcf0b5222050f29b0f5fd23ad350ba64a815cbcce9ddcb37c71004ebfc8cc96c1f666c527
+  data.tar.gz: ab9e98f31624be15f00701d24184e357014c2e5f9f0fdc3d7042293610fe6939b1323865bd882733638f98e2584df2749ac209e8e07770f30efa43273e1de9e7

data/lib/dor-services.rb

@@ -57,6 +57,7 @@ module Dor
     autoload :EditableIndexer
     autoload :IdentifiableIndexer
     autoload :ProcessableIndexer
+    autoload :ProcessIndexer
     autoload :ReleasableIndexer
     autoload :WorkflowIndexer
     autoload :WorkflowsIndexer

data/lib/dor/indexers/identifiable_indexer.rb

@@ -25,6 +25,7 @@ module Dor
       solr_doc[INDEX_VERSION_FIELD] = Dor::VERSION
       solr_doc['indexed_at_dtsi'] = Time.now.utc.xmlschema
       resource.datastreams.values.each do |ds|
+        # This is used to draw the table of datastreams in Argo
         add_solr_value(solr_doc, 'ds_specs', ds.datastream_spec_string, :string, [:symbol]) unless ds.new?
       end
 

data/lib/dor/indexers/process_indexer.rb

@@ -0,0 +1,58 @@
+# frozen_string_literal: true
+
+module Dor
+  # Indexes the process for a workflow
+  class ProcessIndexer
+    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
+
+    # @param [WorkflowSolrDocument] solr_doc
+    # @param [String] workflow_name
+    # @param [Dor::Workflow::Response::Process] process
+    def initialize(solr_doc:, workflow_name:, process:)
+      @solr_doc = solr_doc
+      @workflow_name = workflow_name
+      @process = process
+    end
+
+    # @return [Hash] the partial solr document for the workflow document
+    def to_solr
+      return unless status
+
+      # add a record of the robot having operated on this item, so we can track robot activity
+      solr_doc.add_process_time(workflow_name, name, Time.parse(process.datetime)) if has_time?
+
+      index_error_message
+
+      # workflow name, process status then process name
+      solr_doc.add_wsp("#{workflow_name}:#{status}", "#{workflow_name}:#{status}:#{name}")
+
+      # workflow name, process name then process status
+      solr_doc.add_wps("#{workflow_name}:#{name}", "#{workflow_name}:#{name}:#{status}")
+
+      # process status, workflowname then process name
+      solr_doc.add_swp(process.status.to_s, "#{status}:#{workflow_name}", "#{status}:#{workflow_name}:#{name}")
+    end
+
+    private
+
+    attr_reader :process, :workflow_name, :solr_doc
+    delegate :status, :name, :state, :error_message, :datetime, to: :process
+
+    def has_time?
+      datetime && (status == 'completed' || status == 'error')
+    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
+      return unless error_message
+
+      solr_doc.error = "#{workflow_name}:#{name}:#{error_message}".truncate(MAX_ERROR_LENGTH, omission: ERROR_OMISSION)
+    end
+  end
+end

data/lib/dor/indexers/workflow_indexer.rb

@@ -3,13 +3,6 @@
 module Dor
   # Indexes the objects position in workflows
   class WorkflowIndexer
-    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
-
     # @param [Workflow::Response::Workflow] workflow the workflow document to index
     def initialize(workflow:)
       @workflow = workflow
@@ -18,20 +11,14 @@ module Dor
     # @return [Hash] the partial solr document for the workflow document
     def to_solr
       WorkflowSolrDocument.new do |solr_doc|
-        definition = Dor::Config.workflow.client.workflow_template(workflow_name)
         solr_doc.name = workflow_name
-        definition_process_names = definition['processes'].map { |p| p['name'] }
 
         errors = 0 # The error count is used by the Report class in Argo
-        processes = definition_process_names.map do |process_name|
-          workflow.process_for_recent_version(name: process_name)
-        end
-
         processes.each do |process|
-          index_process(solr_doc, process)
+          ProcessIndexer.new(solr_doc: solr_doc, workflow_name: workflow_name, process: process).to_solr
           errors += 1 if process.status == 'error'
         end
-        solr_doc.status = [workflow_name, workflow_status(workflow), errors, repository].join('|')
+        solr_doc.status = [workflow_name, workflow_status, errors, repository].join('|')
       end
     end
 
@@ -40,39 +27,21 @@ module Dor
     attr_reader :workflow
     delegate :workflow_name, :repository, to: :workflow
 
-    # @param [Workflow::Response::Process] process
-    def index_process(solr_doc, process)
-      return unless process.status
-
-      # add a record of the robot having operated on this item, so we can track robot activity
-      solr_doc.add_process_time(workflow_name, process.name, Time.parse(process.datetime)) if process_has_time?(process)
-
-      index_error_message(solr_doc, process)
-
-      # workflow name, process status then process name
-      solr_doc.add_wsp("#{workflow_name}:#{process.status}", "#{workflow_name}:#{process.status}:#{process.name}")
-
-      # workflow name, process name then process status
-      solr_doc.add_wps("#{workflow_name}:#{process.name}", "#{workflow_name}:#{process.name}:#{process.status}")
-
-      # process status, workflowname then process name
-      solr_doc.add_swp(process.status.to_s, "#{process.status}:#{workflow_name}", "#{process.status}:#{workflow_name}:#{process.name}")
+    def definition_process_names
+      @definition_process_names ||= begin
+        definition = Dor::Config.workflow.client.workflow_template(workflow_name)
+        definition['processes'].map { |p| p['name'] }
+      end
     end
 
-    def process_has_time?(process)
-      process.datetime && process.status && (process.status == 'completed' || process.status == 'error')
+    def processes
+      @processes ||= definition_process_names.map do |process_name|
+        workflow.process_for_recent_version(name: process_name)
+      end
     end
 
-    def workflow_status(workflow)
+    def workflow_status
       workflow.complete? ? 'completed' : 'active'
     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, process)
-      return unless process.error_message
-
-      solr_doc.error = "#{workflow_name}:#{process.name}:#{process.error_message}".truncate(MAX_ERROR_LENGTH, omission: ERROR_OMISSION)
-    end
   end
 end

data/lib/dor/release_tags/identity_metadata.rb

@@ -181,6 +181,8 @@ module Dor
       # @return [Hash] same form as new_tags, with all missing tags not in new_tags, but in current_tag_names, added in with a Boolean value of false
       def add_tags_from_purl(new_tags)
         missing_tags = release_tags_from_purl.map(&:downcase) - new_tags.keys.map(&:downcase)
+        Honeybadger.notify("Found missing release tags on PURL for #{pid}: #{missing_tags.inspect}") if missing_tags.present? && defined? Honeybadger
+
         missing_tags.each do |missing_tag|
           new_tags[missing_tag.capitalize] = { 'release' => false }
         end

data/lib/dor/rest_resource_factory.rb

@@ -5,34 +5,20 @@ module Dor
   class RestResourceFactory
     include Singleton
 
-    # @param type [Symbol] the type of connection to create (e.g. :fedora)
+    # @param url [String] the url to connect to
     # @return [RestClient::Resource]
-    def self.create(type)
-      instance.create(type)
+    def self.create(url)
+      instance.create(url)
     end
 
-    # @param type [Symbol] the type of connection to create (e.g. :fedora)
+    # @param url [String] the url to connect to
     # @return [RestClient::Resource]
-    def create(type)
-      RestClient::Resource.new(url_for(type), connection_options)
+    def create(url)
+      RestClient::Resource.new(url, connection_options)
     end
 
     private
 
-    # @param type [Symbol] the type of connection to create (e.g. :fedora)
-    # @return [String] the url to connect to.
-    def url_for(type)
-      connection_configuration(type).url
-    end
-
-    # @param type [Symbol] the type of connection to create (e.g. :fedora)
-    # @return [#url] the configuration for the connection
-    def connection_configuration(type)
-      Dor::Config.fetch(type)
-    rescue KeyError
-      raise "ERROR: Unable to find a configuration for #{type}"
-    end
-
     # @return [Hash] options for creating a RestClient::Resource
     def connection_options
       {}

data/lib/dor/services/release_tag_service.rb

@@ -12,8 +12,6 @@ module Dor
     def initialize(item)
       @identity_metadata_service = ReleaseTags::IdentityMetadata.new(item)
       @purl_service = ReleaseTags::Purl.new(pid: item.pid, purl_host: Dor::Config.stacks.document_cache_host)
-
-      @item = item
     end
 
     # Called in Dor::UpdateMarcRecordService (in dor-services-app too)

data/lib/dor/static_config/fedora_config.rb

@@ -15,7 +15,7 @@ module Dor
       end
 
       def client
-        CertificateAuthenticatedRestResourceFactory.create(:fedora)
+        CertificateAuthenticatedRestResourceFactory.create(url)
       end
 
       def url(new_value = nil)

data/lib/dor/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Dor
-  VERSION = '8.3.0'
+  VERSION = '8.4.0'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: dor-services
 version: !ruby/object:Gem::Version
-  version: 8.3.0
+  version: 8.4.0
 platform: ruby
 authors:
 - Michael Klein
@@ -20,7 +20,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-01-08 00:00:00.000000000 Z
+date: 2020-02-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: active-fedora
@@ -483,6 +483,7 @@ files:
 - lib/dor/indexers/describable_indexer.rb
 - lib/dor/indexers/editable_indexer.rb
 - lib/dor/indexers/identifiable_indexer.rb
+- lib/dor/indexers/process_indexer.rb
 - lib/dor/indexers/processable_indexer.rb
 - lib/dor/indexers/releasable_indexer.rb
 - lib/dor/indexers/workflow_indexer.rb