sufia-models 4.3.1 → 5.0.0.beta1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c4b6c01683739e17c53f491882eb8c50ead27577
-  data.tar.gz: 2b780079c2b52e9f076df2e50e61835bc1f3ab2a
+  metadata.gz: d1a8c6e6170dab340eb52e7102f29c63de547850
+  data.tar.gz: 68b0a675fb176f9132a01f791ec3ea7302acd716
 SHA512:
-  metadata.gz: 8fdfc3fdd81834097e68d9180cd860edb2f9813226532dd6fe06d643dddce743513556fb1a5f1af7e6bae93a662d6438ab0a6ef25d7521569127c4fcb321a489
-  data.tar.gz: 3249a331ab148fad1979b43b27fee9bf56ef1299061fde4f4ac0489effb9bee1a4d8357464de85350b2b78996d5f9e584565ffca94da43628b44405f895b3c7a
+  metadata.gz: e90c6e067dae4e9aa8c10bbf7d914e4cc28b18355822244cf46be1622d2c7068aba56730e004bea8112fa4083fe77d66635b45e9de8f84c1ce765695db1f15a3
+  data.tar.gz: c2cf9171d0cfaa461c0afc944e250bf9adf0b240f10be8d36d840b48429b73c823d9d170f8514453d9b66b3eef6d820e449ea019bd0bf6b1ff3ce27521a38b7c

data/app/actors/sufia/generic_file/actor.rb

@@ -20,7 +20,7 @@ module Sufia::GenericFile
       generic_file.creator = [user.name]
 
       if batch_id
-        generic_file.batch_id = Sufia::Noid.namespaceize(batch_id)
+        generic_file.batch_id = batch_id
       else
         ActiveFedora::Base.logger.warn "unable to find batch to attach to"
       end
@@ -28,8 +28,7 @@ module Sufia::GenericFile
     end
 
     def create_content(file, file_name, dsid)
-      fname = generic_file.label.blank? ? file_name.truncate(255) : generic_file.label
-      generic_file.add_file(file, dsid, fname)
+      generic_file.add_file(file, dsid, file_name.truncate(255))
       save_characterize_and_record_committer do
         if Sufia.config.respond_to?(:after_create_content)
           Sufia.config.after_create_content.call(generic_file, user)
@@ -37,9 +36,8 @@ module Sufia::GenericFile
       end
     end
 
-    def revert_content(revision_id, datastream_id)
-      revision = generic_file.content.get_version(revision_id)
-      generic_file.add_file(revision.content, datastream_id, revision.label)
+    def revert_content(revision_id)
+      generic_file.content.restore_version(revision_id)
       save_characterize_and_record_committer do
         if Sufia.config.respond_to?(:after_revert_content)
           Sufia.config.after_revert_content.call(generic_file, user, revision_id)
@@ -69,11 +67,10 @@ module Sufia::GenericFile
     end
 
     def destroy
-      pid = generic_file.pid  #Work around for https://github.com/projecthydra/active_fedora/issues/422
       generic_file.destroy
-      FeaturedWork.where(generic_file_id: pid).destroy_all
+      FeaturedWork.where(generic_file_id: generic_file.id).destroy_all
       if Sufia.config.respond_to?(:after_destroy)
-        Sufia.config.after_destroy.call(pid, user)
+        Sufia.config.after_destroy.call(generic_file.id, user)
       end
     end
 
@@ -103,7 +100,7 @@ module Sufia::GenericFile
     end
 
     def push_characterize_job
-      Sufia.queue.push(CharacterizeJob.new(@generic_file.pid))
+      Sufia.queue.push(CharacterizeJob.new(@generic_file.id))
     end
 
     class << self

data/app/jobs/active_fedora_pid_based_job.rb

@@ -10,10 +10,9 @@ class ActiveFedoraPidBasedJob
   end
 
   def object
-    @object ||= ActiveFedora::Base.find(pid, cast: true).tap do |f|
-      f.reload_on_save = true
-    end
+    @object ||= ActiveFedora::Base.find(pid)
   end
+
   alias_method :generic_file, :object
   alias_method :generic_file_id, :pid
 

data/app/jobs/audit_job.rb

@@ -6,44 +6,41 @@ class AuditJob < ActiveFedoraPidBasedJob
   PASS = 'Passing Audit Run'
   FAIL = 'Failing Audit Run'
 
-  attr_accessor :pid, :datastream_id, :version_id
+  attr_accessor :uri, :pid, :path
 
-  def initialize(pid, datastream_id, version_id)
-    super(pid)
-    self.datastream_id = datastream_id
-    self.version_id = version_id
+  # URI of the resource to audit.
+  # This URI could include the actual resource (e.g. content) and the version to audit:
+  #     http://localhost:8983/fedora/rest/test/a/b/c/abcxyz/content/fcr:versions/version1
+  # but it could also just be:
+  #     http://localhost:8983/fedora/rest/test/a/b/c/abcxyz/content
+  def initialize(id, path, uri)
+    super(uri)
+    self.pid = id
+    self.path = path
+    self.uri = uri
   end
 
   def run
-    if generic_file
-      datastream = generic_file.datastreams[datastream_id]
-      if datastream
-        version =  datastream.versions.select { |v| v.versionID == version_id}.first
-        log = run_audit(version)
-
-        # look up the user for sending the message to
-        login = generic_file.depositor
-        if login
-          user = User.find_by_user_key(login)
-          ActiveFedora::Base.logger.warn "User '#{login}' not found" unless user
-          job_user = User.audituser()
-          # send the user a message about the failing audit
-          unless (log.pass == 1)
-            message = "The audit run at #{log.created_at} for #{log.pid}:#{log.dsid}:#{log.version} was #{log.pass == 1 ? 'passing' : 'failing'}."
-            subject = (log.pass == 1 ? PASS : FAIL)
-            job_user.send_message(user, message, subject)
-          end 
-        end
-      else
-        ActiveFedora::Base.logger.warn "No datastream for audit!!!!! pid: #{pid} dsid: #{datastream_id}"
-      end
-    else
-      ActiveFedora::Base.logger.warn "No generic file for data stream audit!!!!! pid: #{pid} dsid: #{datastream_id}"
+    fixity_ok = false
+    log = run_audit(pid, path, uri)
+    fixity_ok = (log.pass == 1)
+    unless fixity_ok
+      # send the user a message about the failing audit
+      login = generic_file.depositor
+      user = User.find_by_user_key(login)
+      ActiveFedora::Base.logger.warn "User '#{login}' not found" unless user
+      job_user = User.audituser()
+      file_title = generic_file.title.first
+      message = "The audit run at #{log.created_at} for #{file_title} (#{uri}) failed."
+      subject = FAIL
+      job_user.send_message(user, message, subject)
     end
+    fixity_ok
   end
 
   private
-  def run_audit(version)
-    object.class.run_audit(version)
+  def run_audit(id, path, uri)
+    object.class.run_audit(id, path, uri)
   end
+
 end

data/app/jobs/batch_update_job.rb

@@ -10,7 +10,7 @@ class BatchUpdateJob
 
   def initialize(login, params)
     self.login = login
-    self.title = params[:title]
+    self.title = params[:title] || {}
     self.file_attributes = params[:generic_file]
     self.visibility = params[:visibility]
     self.batch_id = params[:id]
@@ -25,7 +25,9 @@ class BatchUpdateJob
     batch.generic_files.each do |gf|
       update_file(gf, user)
     end
-    batch.update_attributes({status:["Complete"]})
+
+    batch.update(status: ["Complete"])
+
     if denied.empty?
       send_user_success_message(user, batch) unless saved.empty?
     else
@@ -35,12 +37,12 @@ class BatchUpdateJob
 
   def update_file(gf, user)
     unless user.can? :edit, gf
-      ActiveFedora::Base.logger.error "User #{user.user_key} DENIED access to #{gf.pid}!"
+      ActiveFedora::Base.logger.error "User #{user.user_key} DENIED access to #{gf.id}!"
       denied << gf
       return
     end
-    gf.title = title[gf.pid] if title[gf.pid] rescue gf.label
-    gf.attributes=file_attributes
+    gf.title = title[gf.id] if title[gf.id]
+    gf.attributes = file_attributes
     gf.visibility= visibility
 
     save_tries = 0
@@ -48,13 +50,13 @@ class BatchUpdateJob
       gf.save!
     rescue RSolr::Error::Http => error
       save_tries += 1
-      ActiveFedora::Base.logger.warn "BatchUpdateJob caught RSOLR error on #{gf.pid}: #{error.inspect}"
+      ActiveFedora::Base.logger.warn "BatchUpdateJob caught RSOLR error on #{gf.id}: #{error.inspect}"
       # fail for good if the tries is greater than 3
       raise error if save_tries >=3
       sleep 0.01
       retry
     end #
-    Sufia.queue.push(ContentUpdateEventJob.new(gf.pid, login))
+    Sufia.queue.push(ContentUpdateEventJob.new(gf.id, login))
     saved << gf
   end
 
@@ -67,5 +69,4 @@ class BatchUpdateJob
     message = denied.count > 1 ? multiple_failure(batch.noid, denied) : single_failure(batch.noid, denied.first)
     User.batchuser.send_message(user, message, failure_subject, sanitize_text = false)
   end
-  
 end

data/app/jobs/import_url_job.rb

@@ -11,12 +11,12 @@ class ImportUrlJob < ActiveFedoraPidBasedJob
   def run
     user = User.find_by_user_key(generic_file.depositor)
 
-    Tempfile.open(self.pid) do |f|
+    Tempfile.open(pid.gsub('/', '_')) do |f|
       path = copy_remote_file(generic_file.import_url, f)
       # attach downloaded file to generic file stubbed out
       if Sufia::GenericFile::Actor.new(generic_file, user).create_content(f, path, 'content')
         # add message to user for downloaded file
-        message = "The file (#{generic_file.content.label}) was successfully imported."
+        message = "The file (#{generic_file.label}) was successfully imported."
         job_user.send_message(user, message, 'File Import')
       else
         job_user.send_message(user, generic_file.errors.full_messages.join(', '), 'File Import Error')

data/app/models/batch.rb

@@ -3,24 +3,23 @@ class Batch < ActiveFedora::Base
   include Sufia::ModelMethods
   include Sufia::Noid
 
-  has_metadata name: "descMetadata", type: BatchRdfDatastream
+  has_many :generic_files, predicate: ActiveFedora::RDF::Fcrepo::RelsExt.isPartOf
 
-  belongs_to :user, property: "creator"
-  has_many :generic_files, property: :is_part_of
+  property :creator, predicate: ::RDF::DC.creator
+  property :title, predicate: ::RDF::DC.title
+  property :status, predicate: ::RDF::DC.type
 
-  has_attributes :title, :creator, :part, :status, datastream: :descMetadata, multiple: true
-
-  def self.find_or_create(pid)
+  def self.find_or_create(id)
     begin
-      Batch.find(pid)
+      Batch.find(id)
     rescue ActiveFedora::ObjectNotFoundError
-      Batch.create({pid: pid})
+      Batch.create(id: id)
     end
   end
 
-  def to_solr(solr_doc={}, opts={})
-    solr_doc = super(solr_doc, opts)
-    solr_doc[Solrizer.solr_name('noid', Sufia::GenericFile.noid_indexer)] = noid
-    return solr_doc
+  def to_solr(solr_doc={})
+    super.tap do |solr_doc|
+      solr_doc[Solrizer.solr_name('noid', Sufia::GenericFile.noid_indexer)] = noid
+    end
   end
 end

data/app/models/checksum_audit_log.rb

@@ -1,16 +1,15 @@
 class ChecksumAuditLog < ActiveRecord::Base
 
-  def ChecksumAuditLog.get_audit_log(version)
-    ChecksumAuditLog.find_or_create_by_pid_and_dsid_and_version(pid: version.pid,
-                                                                dsid: version.dsid,
-                                                                version: version.versionID)
+  def ChecksumAuditLog.get_audit_log(id, path, version_uri)
+    ChecksumAuditLog.find_or_create_by(pid: id, dsid: path, version: version_uri)
   end
 
-  def ChecksumAuditLog.prune_history(version)
-    ## Check to see if there are previous passing logs that we can delete
-    # we want to keep the first passing event after a failure, the most current passing event, and all failures so that this table doesn't grow too large
+  def ChecksumAuditLog.prune_history(id, path)
+    # Check to see if there are previous passing logs that we can delete
+    # we want to keep the first passing event after a failure, the most current passing event, 
+    # and all failures so that this table doesn't grow too large
     # Simple way (a little naieve): if the last 2 were passing, delete the first one
-    logs = GenericFile.load_instance_from_solr(version.pid).logs(version.dsid)
+    logs = GenericFile.load_instance_from_solr(id).logs(path)
     list = logs.limit(2)
     if list.size > 1 && (list[0].pass == 1) && (list[1].pass == 1)
       list[0].destroy

data/app/models/concerns/sufia/ability.rb

@@ -15,8 +15,8 @@ module Sufia
     end
 
     def proxy_deposit_abilities
-      can :transfer, String do |pid|
-        get_depositor_from_pid(pid) == current_user.user_key
+      can :transfer, String do |id|
+        depositor_for_document(id) == current_user.user_key
       end
       can :create, ProxyDepositRequest if user_groups.include? 'registered'
       can :accept, ProxyDepositRequest, receiving_user_id: current_user.id, status: 'pending'
@@ -47,10 +47,8 @@ module Sufia
 
     private
 
-    def get_depositor_from_pid(pid)
-      ::GenericFile.load_instance_from_solr(pid).depositor
-    rescue
-      nil
+    def depositor_for_document(document_id)
+      ::GenericFile.load_instance_from_solr(document_id).depositor
     end
   end
 end

data/app/models/concerns/sufia/collection.rb

@@ -11,7 +11,6 @@ module Sufia
       before_save :update_permissions
       validates :title, presence: true
 
-      has_metadata "properties", type: PropertiesDatastream
     end
 
     def terms_for_display
@@ -35,10 +34,10 @@ module Sufia
       noid
     end
 
-    def to_solr(solr_doc={}, opts={})
-      super(solr_doc, opts)
-      solr_doc[Solrizer.solr_name("noid", Sufia::GenericFile.noid_indexer)] = noid
-      return solr_doc
+    def to_solr(solr_doc={})
+      super.tap do |solr_doc|
+        solr_doc[Solrizer.solr_name("noid", Sufia::GenericFile.noid_indexer)] = noid
+      end
     end
 
     def update_permissions

data/app/models/concerns/sufia/file_stat_utils.rb

@@ -17,13 +17,13 @@ module Sufia
       {ga_start_date: ga_start_date, cached_stats: stats.to_a }
     end
 
-    def combined_stats file_id, start_date, object_method, ga_key, user_id=nil
+    def combined_stats file_id, start_date, object_method, ga_key
       stat_cache_info = cached_stats( file_id, start_date, object_method)
       stats = stat_cache_info[:cached_stats]
       if stat_cache_info[:ga_start_date] < Date.today
         ga_stats =  ga_statistics(stat_cache_info[:ga_start_date], file_id)
         ga_stats.each do |stat|
-          lstat = self.new file_id: file_id, date: stat[:date], object_method => stat[ga_key], user_id: user_id
+          lstat = self.new file_id:file_id, date: stat[:date], object_method => stat[ga_key]
           lstat.save unless Date.parse(stat[:date]) == Date.today
           stats << lstat
         end
@@ -32,4 +32,4 @@ module Sufia
     end
 
   end
-end
+end

data/app/models/concerns/sufia/generic_file.rb

@@ -17,17 +17,16 @@ module Sufia
     include Sufia::GenericFile::Metadata
     include Sufia::GenericFile::Versions
     include Sufia::GenericFile::VirusCheck
-    include Sufia::GenericFile::ReloadOnSave
     include Sufia::GenericFile::FullTextIndexing
     include Sufia::GenericFile::ProxyDeposit
     include Hydra::Collections::Collectible
 
     included do
-      belongs_to :batch, property: :is_part_of
+      belongs_to :batch, predicate: ActiveFedora::RDF::Fcrepo::RelsExt.isPartOf
 
-      around_save :retry_warming
+      # around_save :retry_warming
 
-      attr_accessible *(terms_for_display + [:part_of, :permissions])
+      attr_accessible *(terms_for_display + [:part_of, :permissions_attributes])
     end
 
     def persistent_url
@@ -37,6 +36,7 @@ module Sufia
     def retry_warming
       save_tries = 0
       conflict_tries = 0
+      etag_tries = 0
       begin
         yield
       rescue RSolr::Error::Http => error
@@ -46,6 +46,15 @@ module Sufia
         raise if save_tries >=3
         sleep 0.01
         retry
+      rescue Ldp::EtagMismatch
+        prev_changes = changes.dup
+        # There was a version conflict, so reload the previous version, then apply the changed attributes
+        reload
+        prev_changes.each do |key, (_, value)|
+          self[key] = value
+        end
+        raise if etag_tries >= 1
+        retry
       rescue  ActiveResource::ResourceConflict => error
         conflict_tries += 1
         logger.warn "Retry caught Active Resource Conflict #{self.pid}: #{error.inspect}"
@@ -81,21 +90,14 @@ module Sufia
       @noid_indexer ||= Solrizer::Descriptor.new(:text, :indexed, :stored)
     end
 
-    def to_solr(solr_doc={}, opts={})
-      super(solr_doc, opts).tap do |solr_doc|
+    def to_solr(solr_doc={})
+      super.tap do |solr_doc|
         solr_doc[Solrizer.solr_name('label')] = self.label
         solr_doc[Solrizer.solr_name('noid', Sufia::GenericFile.noid_indexer)] = noid
         solr_doc[Solrizer.solr_name('file_format')] = file_format
         solr_doc[Solrizer.solr_name('file_format', :facetable)] = file_format
         solr_doc['all_text_timv'] = full_text.content
-        solr_doc = index_collection_pids(solr_doc)
-      end
-    end
-
-    def label=(new_label)
-      @inner_object.label = new_label
-      if self.title.empty?
-        self.title = [new_label].compact
+        solr_doc = index_collection_ids(solr_doc)
       end
     end
 

data/app/models/concerns/sufia/generic_file/audit.rb

@@ -5,25 +5,43 @@ module Sufia
 
       NO_RUNS = 999
 
-      def audit(force = false)
-        logs = []
-        self.per_version do |ver|
-          logs << audit_each(ver, force)
+      # provides a human readable version of the audit status
+      def human_readable_audit_status
+        stat = audit_stat(false)
+        case stat
+          when 0
+            'failing'
+          when 1
+            'passing'
+          else
+            stat
         end
-        logs
       end
 
-      def per_version(&block)
-        self.datastreams.each do |dsid, ds|
-          next if ds == full_text
-          ds.versions.each do |ver|
-            block.call(ver)
-          end
+      # TODO: Run audits on all attached files. We're only audting "content" at tht moment
+      def audit force = false
+        @audit_log ||= Array.new
+        @force = force
+        audit_content
+        return @audit_log
+      end
+
+      def audit_content
+        if content.has_versions?
+          audit_file_versions("content")
+        else
+          @audit_log << audit_file("content", content.uri)
+        end
+      end
+
+      def audit_file_versions file
+        attached_files[file].versions.all.each do |version|
+          @audit_log << audit_file(file, version.uri,  version.label)
         end
       end
 
-      def logs(dsid)
-        ChecksumAuditLog.where(dsid: dsid, pid: self.pid).order('created_at desc, id desc')
+      def logs(file)
+        ChecksumAuditLog.where(pid: self.id, dsid: file).order('created_at desc, id desc')
       end
 
       def audit!
@@ -51,16 +69,11 @@ module Sufia
         end
       end
 
-      def audit_each(version, force = false)
-        latest_audit = logs(version.dsid).first
-        return latest_audit unless force || ::GenericFile.needs_audit?(version, latest_audit)
-
-        #  Resque.enqueue(AuditJob, version.pid, version.dsid, version.versionID)
-        Sufia.queue.push(AuditJob.new(version.pid, version.dsid, version.versionID))
-
-        # run the find just incase the job has finished already
-        latest_audit = logs(version.dsid).first
-        latest_audit = ChecksumAuditLog.new(pass: NO_RUNS, pid: version.pid, dsid: version.dsid, version: version.versionID) unless latest_audit
+      def audit_file(file, uri, label = nil)
+        latest_audit = logs(file).first
+        return latest_audit unless @force || ::GenericFile.needs_audit?(uri, latest_audit)
+        Sufia.queue.push(AuditJob.new(id, file, uri))
+        latest_audit ||= ChecksumAuditLog.new(pass: NO_RUNS, pid: id, dsid: file, version: label)
         latest_audit
       end
 
@@ -70,8 +83,9 @@ module Sufia
           ::GenericFile.audit(version, true)
         end
 
-        def audit(version, force = false)
-          latest_audit = self.find(version.pid).audit_each( version, force)
+        def audit(version_uri, force = false)
+          return { pass: true } # TODO Just skipping the audit for now
+          latest_audit = self.find(version_uri).audit_each( version, force)
         end
 
         def needs_audit?(version, latest_audit)
@@ -98,16 +112,21 @@ module Sufia
           ::GenericFile.audit_everything(true)
         end
 
-        def run_audit(version)
-          if version.dsChecksumValid
+        def run_audit(id, path, uri)
+          begin
+            fixity_ok = ActiveFedora::FixityService.new(uri).check
+          rescue Ldp::NotFound 
+            error_msg = "resource not found"
+          end
+
+          if fixity_ok
             passing = 1
-            ChecksumAuditLog.prune_history(version)
+            ChecksumAuditLog.prune_history(id, path)
           else
-            logger.warn "***AUDIT*** Audit failed for #{version.pid} #{version.versionID}"
+            logger.warn "***AUDIT*** Audit failed for #{uri} #{error_msg}"
             passing = 0
           end
-          check = ChecksumAuditLog.create!(pass: passing, pid: version.pid,
-                                           dsid: version.dsid, version: version.versionID)
+          check = ChecksumAuditLog.create!(pass: passing, pid: id, version: uri, dsid: path)
           check
         end
       end