curation_concerns-models 0.1.0

data/app/jobs/active_fedora_id_based_job.rb

@@ -0,0 +1,22 @@
+class ActiveFedoraIdBasedJob
+  def queue_name
+    :id_based
+  end
+
+  attr_accessor :id
+
+  def initialize(id)
+    self.id = id
+  end
+
+  def object
+    @object ||= ActiveFedora::Base.find(id)
+  end
+
+  alias_method :generic_file, :object
+  alias_method :generic_file_id, :id
+
+  def run
+    raise RuntimeError, "Define #run in a subclass"
+  end
+end

data/app/jobs/active_fedora_pid_based_job.rb

@@ -0,0 +1,6 @@
+class ActiveFedoraPidBasedJob < ActiveFedoraIdBasedJob
+  extend Deprecation
+  def self.extended(document)
+    Deprecation.warn ActiveFedoraPidBasedJob, "ActiveFedoraPidBasedJob is deprecated; use ActiveFedoraIdBasedJob instead."
+  end
+end

data/app/jobs/audit_job.rb

@@ -0,0 +1,58 @@
+class AuditJob < ActiveFedoraIdBasedJob
+  def queue_name
+    :audit
+  end
+
+  attr_accessor :uri, :id, :file_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
+  # @param [String] id of the parent object
+  # @param [String] file_id used to find the file within its parent object (usually "original_file")
+  # @param [String] uri of the specific file/version to be audited
+  def initialize(id, file_id, uri)
+    super(id)
+    self.file_id = file_id
+    self.uri = uri
+  end
+
+  def run
+    fixity_ok = false
+    log = run_audit
+    fixity_ok = (log.pass == 1)
+    unless fixity_ok
+      if CurationConcerns.config.respond_to?(:after_audit_failure)
+        login = generic_file.depositor
+        user = User.find_by_user_key(login)
+        CurationConcerns.config.after_audit_failure.call(generic_file, user, log.created_at)
+      end
+    end
+    fixity_ok
+  end
+
+  protected
+
+    def run_audit
+      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(id, file_id)
+      else
+        logger.warn "***AUDIT*** Audit failed for #{uri} #{error_msg}"
+        passing = 0
+      end
+      ChecksumAuditLog.create!(pass: passing, generic_file_id: id, version: uri, file_id: file_id)
+    end
+
+    def logger
+      ActiveFedora::Base.logger
+    end
+end

data/app/jobs/characterize_job.rb

@@ -0,0 +1,11 @@
+class CharacterizeJob < ActiveFedoraIdBasedJob
+  def queue_name
+    :characterize
+  end
+
+  def run
+    CurationConcerns::CharacterizationService.run(generic_file)
+    generic_file.save
+    CurationConcerns.queue.push(CreateDerivativesJob.new(generic_file.id))
+  end
+end

data/app/jobs/copy_permissions_job.rb

@@ -0,0 +1,24 @@
+class CopyPermissionsJob
+  def queue_name
+    :permissions
+  end
+
+  attr_accessor :id
+
+  def initialize(id)
+    self.id = id
+  end
+
+  def run
+    # work = ActiveFedora::Base.load_instance_from_solr(id)
+    work = ActiveFedora::Base.find(id)  # Hydra::Works::GenericWork is not compatible with load_instance_from_solr.  Using (slower) ActiveFedora::Base.find instead.
+    if work.respond_to?(:generic_files)
+      work.generic_files.each do |file|
+
+        work.permissions.each {|perm| file.permissions << Hydra::AccessControls::Permission.new(perm.to_hash)}
+        new_permissions = file.permissions.uniq! { |item| item.to_hash}
+        file.permissions = new_permissions unless new_permissions.blank?
+      end
+    end
+  end
+end

data/app/jobs/create_derivatives_job.rb

@@ -0,0 +1,15 @@
+class CreateDerivativesJob < ActiveFedoraIdBasedJob
+  def queue_name
+    :derivatives
+  end
+
+  def run
+    return unless generic_file.original_file.has_content?
+    if generic_file.video?
+      return unless CurationConcerns.config.enable_ffmpeg
+    end
+    
+    generic_file.create_derivatives
+    generic_file.save
+  end
+end

data/app/jobs/import_url_job.rb

@@ -0,0 +1,56 @@
+require 'net/https'
+require 'uri'
+require 'tempfile'
+
+class ImportUrlJob < ActiveFedoraIdBasedJob
+
+  def queue_name
+    :import_url
+  end
+
+  def run
+    user = User.find_by_user_key(generic_file.depositor)
+
+    Tempfile.open(id.gsub('/', '_')) do |f|
+      copy_remote_file(generic_file.import_url, f)
+      # attach downloaded file to generic file stubbed out
+      if CurationConcerns::GenericFileActor.new(generic_file, user).create_content(f)
+
+        # send message to user on download success
+        if CurationConcerns.config.respond_to?(:after_import_url_success)
+          CurationConcerns.config.after_import_url_success.call(generic_file, user)
+        end
+      else
+
+        # send message to user on download failure
+        if CurationConcerns.config.respond_to?(:after_import_url_failure)
+          CurationConcerns.config.after_import_url_failure.call(generic_file, user)
+        end
+      end
+    end
+  end
+
+  def copy_remote_file(import_url, f)
+    f.binmode
+    # download file from url
+    uri = URI(generic_file.import_url)
+    http = Net::HTTP.new(uri.host, uri.port)
+    http.use_ssl = uri.scheme == "https"  # enable SSL/TLS
+    http.verify_mode = OpenSSL::SSL::VERIFY_NONE
+    mime_type = nil
+
+    http.start do
+      http.request_get(uri.request_uri) do |resp|
+        mime_type = resp.content_type
+        resp.read_body do |segment|
+          f.write(segment)
+        end
+      end
+    end
+    f.rewind
+  end
+
+  def job_user
+    User.batchuser
+  end
+end

data/app/jobs/ingest_local_file_job.rb

@@ -0,0 +1,48 @@
+class IngestLocalFileJob
+  attr_accessor :directory, :filename, :user_key, :generic_file_id
+
+  def queue_name
+    :ingest
+  end
+
+  def initialize(generic_file_id, directory, filename, user_key)
+    self.generic_file_id = generic_file_id
+    self.directory = directory
+    self.filename = filename
+    self.user_key = user_key
+  end
+
+  def run
+    user = User.find_by_user_key(user_key)
+    raise "Unable to find user for #{user_key}" unless user
+    generic_file = GenericFile.find(generic_file_id)
+    generic_file.label ||= filename
+    path = File.join(directory, filename)
+
+    actor = CurationConcerns::GenericFileActor.new(generic_file, user)
+
+    if actor.create_content(File.open(path))
+      FileUtils.rm(path)
+
+      # send message to user on import success
+      if CurationConcerns.config.respond_to?(:after_import_local_file_success)
+        CurationConcerns.config.after_import_local_file_success.call(generic_file, user, filename)
+      end
+    else
+
+      # send message to user on import failure
+      if CurationConcerns.config.respond_to?(:after_import_local_file_failure)
+        CurationConcerns.config.after_import_local_file_failure.call(generic_file, user, filename)
+      end
+    end
+  end
+
+  def job_user
+    User.batchuser
+  end
+
+  def mime_type(file_name)
+    mime_types = MIME::Types.of(file_name)
+    mime_types.empty? ? "application/octet-stream" : mime_types.first.content_type
+  end
+end

data/app/jobs/resolrize_job.rb

@@ -0,0 +1,9 @@
+class ResolrizeJob
+  def queue_name
+    :resolrize
+  end
+
+  def run
+    ActiveFedora::Base.reindex_everything
+  end
+end

data/app/models/checksum_audit_log.rb

@@ -0,0 +1,21 @@
+class ChecksumAuditLog < ActiveRecord::Base
+
+  def self.get_audit_log(id, path, version_uri)
+    ChecksumAuditLog.find_or_create_by(generic_file_id: id, file_id: path, version: version_uri)
+  end
+
+  # 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
+  def self.prune_history(id, path)
+    list = logs_for(id, path).limit(2)
+    if list.size > 1 && (list[0].pass == 1) && (list[1].pass == 1)
+      list[0].destroy
+    end
+  end
+
+  def self.logs_for(id, path)
+    ChecksumAuditLog.where(generic_file_id: id, file_id: path).order('created_at desc, id desc')
+  end
+end

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

@@ -0,0 +1,34 @@
+module CurationConcerns
+  module Ability
+    extend ActiveSupport::Concern
+    included do
+      self.ability_logic += [:curation_concerns_permissions]
+    end
+
+    def curation_concerns_permissions
+
+      unless current_user.new_record?
+        can :create, CurationConcerns::ClassifyConcern
+        # TODO: Shouldn't this be in `everyone_can_create_curation_concerns` ?
+        can :create, ::GenericFile
+      end
+
+      if user_groups.include? 'admin'
+       can [:create, :discover, :show, :read, :edit, :update, :destroy], :all
+      end
+
+      can :collect, :all
+
+    end
+
+    # Add this to your ability_logic if you want all logged in users to be able
+    # to submit content
+    def everyone_can_create_curation_concerns
+      unless current_user.new_record?
+        can :create, [ CurationConcerns.configuration.curation_concerns ]
+        can :create, ::Collection
+      end
+    end
+
+  end
+end

data/app/models/concerns/curation_concerns/basic_metadata.rb

@@ -0,0 +1,87 @@
+module CurationConcerns
+  module BasicMetadata
+    extend ActiveSupport::Concern
+
+    included do
+
+      property :label, predicate: ActiveFedora::RDF::Fcrepo::Model.downloadFilename, multiple: false
+
+      property :depositor, predicate: ::RDF::URI.new("http://id.loc.gov/vocabulary/relators/dpt"), multiple: false do |index|
+        index.as :symbol, :stored_searchable
+      end
+
+      property :relative_path, predicate: ::RDF::URI.new('http://scholarsphere.psu.edu/ns#relativePath'), multiple: false
+
+      property :import_url, predicate: ::RDF::URI.new('http://scholarsphere.psu.edu/ns#importUrl'), multiple: false do |index|
+        index.as :symbol
+      end
+
+      property :part_of, predicate: ::RDF::DC.isPartOf
+      property :resource_type, predicate: ::RDF::DC.type do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :title, predicate: ::RDF::DC.title do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :creator, predicate: ::RDF::DC.creator do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :contributor, predicate: ::RDF::DC.contributor do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :description, predicate: ::RDF::DC.description do |index|
+        index.type :text
+        index.as :stored_searchable
+      end
+      property :tag, predicate: ::RDF::DC.relation do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :rights, predicate: ::RDF::DC.rights do |index|
+        index.as :stored_searchable
+      end
+      property :publisher, predicate: ::RDF::DC.publisher do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :date_created, predicate: ::RDF::DC.created do |index|
+        index.as :stored_searchable
+      end
+
+      # We reserve date_uploaded for the original creation date of the record.
+      # For example, when migrating data from a fedora3 repo to fedora4,
+      # fedora's system created date will reflect the date when the record
+      # was created in fedora4, but the date_uploaded will preserve the
+      # original creation date from the old repository.
+      property :date_uploaded, predicate: ::RDF::DC.dateSubmitted, multiple: false do |index|
+        index.type :date
+        index.as :stored_sortable
+      end
+
+      property :date_modified, predicate: ::RDF::DC.modified, multiple: false do |index|
+        index.type :date
+        index.as :stored_sortable
+      end
+      property :subject, predicate: ::RDF::DC.subject do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :language, predicate: ::RDF::DC.language do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :identifier, predicate: ::RDF::DC.identifier do |index|
+        index.as :stored_searchable
+      end
+      property :based_near, predicate: ::RDF::FOAF.based_near do |index|
+        index.as :stored_searchable, :facetable
+      end
+      property :related_url, predicate: ::RDF::RDFS.seeAlso do |index|
+        index.as :stored_searchable
+      end
+      property :bibliographic_citation, predicate: ::RDF::DC.bibliographicCitation do |index|
+        index.as :stored_searchable
+      end
+      property :source, predicate: ::RDF::DC.source do |index|
+        index.as :stored_searchable
+      end
+    end
+
+  end
+end

data/app/models/concerns/curation_concerns/collection_behavior.rb

@@ -0,0 +1,47 @@
+module CurationConcerns
+  module CollectionBehavior
+    extend ActiveSupport::Concern
+
+    include Hydra::AccessControls::WithAccessRight
+    include Hydra::Collection
+    include CurationConcerns::Noid
+    include CurationConcerns::HumanReadableType
+    include CurationConcerns::HasRepresentative
+    include CurationConcerns::Permissions
+
+    included do
+      validates :title, presence: true
+    end
+
+    def add_member(collectible)
+      if can_add_to_members?(collectible)
+        self.members << collectible
+        save
+      end
+    end
+
+    def to_s
+      title.present? ? title : "No Title"
+    end
+
+    def bytes
+      members.reduce(0) { |sum, gf| sum + gf.content.size.to_i }
+    end
+
+    def to_solr(solr_doc={})
+      super(solr_doc).tap do |solr_doc|
+        Solrizer.set_field(solr_doc, 'generic_type', human_readable_type, :facetable)
+      end
+    end
+
+    def can_be_member_of_collection?(collection)
+      collection == self ? false : true
+    end
+
+    private
+
+    def can_add_to_members?(collectible)
+      collectible.try(:can_be_member_of_collection?, self)
+    end
+  end
+end

data/app/models/concerns/curation_concerns/generic_file/belongs_to_works.rb

@@ -0,0 +1,53 @@
+module CurationConcerns
+  module GenericFile
+    module BelongsToWorks
+      extend ActiveSupport::Concern
+
+      included do
+        before_destroy :remove_representative_relationship
+      end
+
+      def generic_works
+        self.parent_objects  # parent_objects is provided by Hydra::PCDM::ObjectBehavior
+      end
+
+      def generic_work_ids
+        generic_works.map { |work| work.id }
+      end
+
+      # Returns the first parent object
+      # This is a hack to handle things like GenericFiles inheriting access controls from their parent.  (see CurationConcerns::ParentContainer in app/controllers/concerns/curation_concers/parent_container.rb)
+      def parent
+        self.parent_objects.first
+      end
+
+      # Returns the id of first parent object
+      # This is a hack to handle things like GenericFiles inheriting access controls from their parent.  (see CurationConcerns::ParentContainer in app/controllers/concerns/curation_concers/parent_container.rb)
+      def parent_id
+        parent.id
+      end
+
+      # Files with sibling relationships
+      # Returns all GenericFiles aggregated by any of the GenericWorks that aggregate the current object
+      def related_files
+        generic_works = self.generic_works
+        return [] if generic_works.empty?
+        generic_works.flat_map {|work| work.generic_files.select {|generic_file| generic_file.id != self.id } }
+      end
+
+      # If any parent works are pointing at this object as their representative, remove that pointer.
+      def remove_representative_relationship
+        generic_works = self.generic_works
+        return if generic_works.empty?
+        generic_works.each do |work|
+          if work.representative == self.id
+            work.representative = nil
+            work.save
+          end
+        end
+      end
+
+    end
+  end
+end
+

data/app/models/concerns/curation_concerns/generic_file/characterization.rb

@@ -0,0 +1,89 @@
+module CurationConcerns
+  module GenericFile
+    module Characterization
+      extend ActiveSupport::Concern
+      included do
+        contains "characterization", class_name: 'FitsDatastream'
+        property :mime_type,         delegate_to: 'characterization', multiple: false do |index|
+          index.as :stored_searchable
+        end
+        property :format_label,      delegate_to: 'characterization'
+        property :file_size,         delegate_to: 'characterization'
+        property :last_modified,     delegate_to: 'characterization'
+        property :filename,          delegate_to: 'characterization'
+        property :original_checksum, delegate_to: 'characterization'
+        property :rights_basis,      delegate_to: 'characterization'
+        property :copyright_basis,   delegate_to: 'characterization'
+        property :copyright_note,    delegate_to: 'characterization'
+        property :well_formed,       delegate_to: 'characterization'
+        property :valid,             delegate_to: 'characterization'
+        property :status_message,    delegate_to: 'characterization'
+        property :file_title,        delegate_to: 'characterization'
+        property :file_author,       delegate_to: 'characterization'
+        property :page_count,        delegate_to: 'characterization'
+        property :file_language,     delegate_to: 'characterization'
+        property :word_count,        delegate_to: 'characterization'
+        property :character_count,   delegate_to: 'characterization'
+        property :paragraph_count,   delegate_to: 'characterization'
+        property :line_count,        delegate_to: 'characterization'
+        property :table_count,       delegate_to: 'characterization'
+        property :graphics_count,    delegate_to: 'characterization'
+        property :byte_order,        delegate_to: 'characterization'
+        property :compression,       delegate_to: 'characterization'
+        property :color_space,       delegate_to: 'characterization'
+        property :profile_name,      delegate_to: 'characterization'
+        property :profile_version,   delegate_to: 'characterization'
+        property :orientation,       delegate_to: 'characterization'
+        property :color_map,         delegate_to: 'characterization'
+        property :image_producer,    delegate_to: 'characterization'
+        property :capture_device,    delegate_to: 'characterization'
+        property :scanning_software, delegate_to: 'characterization'
+        property :exif_version,      delegate_to: 'characterization'
+        property :gps_timestamp,     delegate_to: 'characterization'
+        property :latitude,          delegate_to: 'characterization'
+        property :longitude,         delegate_to: 'characterization'
+        property :character_set,     delegate_to: 'characterization'
+        property :markup_basis,      delegate_to: 'characterization'
+        property :markup_language,   delegate_to: 'characterization'
+        property :bit_depth,         delegate_to: 'characterization'
+        property :channels,          delegate_to: 'characterization'
+        property :data_format,       delegate_to: 'characterization'
+        property :offset,            delegate_to: 'characterization'
+        property :frame_rate,        delegate_to: 'characterization'
+
+      end
+
+      def width
+        characterization.width.blank? ? characterization.video_width : characterization.width
+      end
+
+      def height
+        characterization.height.blank? ? characterization.video_height : characterization.height
+      end
+
+      def duration
+        characterization.duration.blank? ? characterization.video_duration : characterization.duration
+      end
+
+      def sample_rate
+        characterization.sample_rate.blank? ? characterization.video_sample_rate : characterization.sample_rate
+      end
+
+      def characterization_terms
+        h = {}
+        self.characterization.class.terminology.terms.each_pair do |k, v|
+          next unless v.respond_to? :proxied_term
+          term = v.proxied_term
+          begin
+            value = self.send(term.name)
+            h[term.name] = value unless value.empty?
+          rescue NoMethodError
+            next
+          end
+        end
+        h
+      end
+
+    end
+  end
+end

data/app/models/concerns/curation_concerns/generic_file/content.rb

@@ -0,0 +1,8 @@
+module CurationConcerns
+  module GenericFile
+    module Content
+      extend ActiveSupport::Concern
+
+    end
+  end
+end