cul_hydra 1.0.6 → 1.0.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b2394e77a1eb550f0a7345256e447bf195ecf191
-  data.tar.gz: d2283027faf5a423ff68c2971b71a65e6c0fceb6
+  metadata.gz: 01f266d6762f6bf8a2ce0d8c498a2fa414ac775d
+  data.tar.gz: b307b9f693f468d48c3b503332b9df090428cf40
 SHA512:
-  metadata.gz: 82937f824a67b93f93ab736899f30dda395ede7288436d0478ec59870289d44cd5a530361685ac1e54671ce6a9d8967cb2823cfcb53a49f308fc8924bcc27f6e
-  data.tar.gz: ab7a330d7a7ff0f32588a6aaf76a3363cd26ca2e6ab5e28ad900bcf65d4ce039d8bdc11f3c7220565b287625e7a6c010b439e67f9173a913f26ad9d6aa86e080
+  metadata.gz: 9a7d2446ef89b32a22d2e14945fb104fff5be8c1cd12562ab7e128fa90c5e09708c72a0a9d198ac37fa8b833f21cd8e972f383a267614de38b8337de5e5c5d6a
+  data.tar.gz: 3a1a0fb360f335f9138abbc0827b0339347023405980536002515ddfcf4496c4866d2eddef81eee782e03b6151c515e77e04cff13c4c48b30d4e34cb3e634f5b

data/app/models/cul/hydra/datastreams/struct_metadata.rb

@@ -156,7 +156,9 @@ class StructMetadata < ::ActiveFedora::Datastream
     current = node
     labels = []
     while (current.name == 'div')
-      labels.unshift URI.escape(current['LABEL'])
+      label = URI.escape(current['LABEL'])
+      label = URI.escape(current['ORDER']) if label.blank?
+      labels.unshift label
       current = current.parent
     end
     labels

data/fixtures/spec/STRUCTMAP/structmap-unlabeled-seq.xml

@@ -0,0 +1,4 @@
+<mets:structMap xmlns:mets="http://www.loc.gov/METS/">
+<mets:div CONTENTIDS="apt://columbia.edu/columbia.jay/data/jjbw/01026/01026001.tif" ORDER="1" LABEL=""/>
+<mets:div CONTENTIDS="apt://columbia.edu/columbia.jay/data/jjbw/01026/01026002.tif" ORDER="2" LABEL=""/>
+</mets:structMap>

data/lib/cul_hydra/version.rb

@@ -1,6 +1,6 @@
 module Cul
   module Hydra
-    VERSION = '1.0.6'
+    VERSION = '1.0.7'
     def self.version
       VERSION
     end

data/lib/cul_hydra/version.rb~

@@ -1,6 +1,6 @@
 module Cul
   module Hydra
-    VERSION = '1.0.5'
+    VERSION = '1.0.6'
     def self.version
       VERSION
     end

data/lib/tasks/index.rake

@@ -1,3 +1,5 @@
+require 'thread/pool'
+
 namespace :cul_hydra do
 
   namespace :index do
@@ -51,21 +53,58 @@ namespace :cul_hydra do
         puts 'Please specify a project PID (e.g. PID=cul:123)'
         next
       end
+      
+      if ENV['THREADS'].present?
+        thread_pool_size = ENV['THREADS'].to_i
+        puts "Number of threads: #{thread_pool_size}"
+      else
+        thread_pool_size = 1
+        puts "Number of threads: #{thread_pool_size}"
+      end
 
       skip_generic_resources = (ENV['skip_generic_resources'] == 'true')
+      
+      ### Stop excessive ActiveFedora logging ###
+      # initialize the fedora connection if necessary
+      connection = (ActiveFedora::Base.fedora_connection[0] ||= ActiveFedora::RubydoraConnection.new(ActiveFedora.config.credentials)).connection
+      # the logger accessor is private
+      (connection.api.send :logger).level = Logger::INFO
 
       start_time = Time.now
       pids = Cul::Hydra::RisearchMembers.get_project_constituent_pids(project_pid, true)
       total = pids.length
       puts "Found #{total} project members."
       counter = 0
+      
+      # We run into autoloading issues when running in a multithreaded context,
+      # so we'll have the application eager load all classes now.
+      Rails.application.eager_load!
+      # Hack: Force load of classes that are giving autoload errors by referencing them below
+      BagAggregator.to_s
+      ContentAggregator.to_s
+      GenericResource.to_s
+      
+      ###########################################
+      pool = Thread.pool(thread_pool_size)
+      mutex = Mutex.new
 
       pids.each do |pid|
-        Cul::Hydra::Indexer.index_pid(pid, skip_generic_resources, false)
-        counter += 1
-        puts "Indexed #{counter} of #{total} | #{Time.now - start_time} seconds"
+        pool.process {
+          
+          Cul::Hydra::Indexer.index_pid(pid, skip_generic_resources, false)
+          
+          mutex.synchronize do
+            counter += 1
+            puts "Indexed #{counter} of #{total} | #{Time.now - start_time} seconds"
+          end
+        }
       end
 
+      pool.shutdown
+      ###########################################
+      
+      puts 'Done'
+
     end
     
     task :by_publish_target_pid => :environment do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: cul_hydra
 version: !ruby/object:Gem::Version
-  version: 1.0.6
+  version: 1.0.7
 platform: ruby
 authors:
 - Benjamin Armintor
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-05-26 00:00:00.000000000 Z
+date: 2015-06-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: blacklight
@@ -137,6 +137,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: thread
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rdf
   requirement: !ruby/object:Gem::Requirement
@@ -432,6 +446,7 @@ files:
 - fixtures/spec/STRUCTMAP/structmap-nested.xml
 - fixtures/spec/STRUCTMAP/structmap-recto.xml
 - fixtures/spec/STRUCTMAP/structmap-seq.xml
+- fixtures/spec/STRUCTMAP/structmap-unlabeled-seq.xml
 - fixtures/spec/STRUCTMAP/structmap-unordered-seq.xml
 - lib/cul_hydra.rb
 - lib/cul_hydra/access_controls_enforcement.rb