9dbcf7a4-1c56-4510-9614-74ad5a22cff6

Committed 31 Jul 2024 02:46PM UTC coverage: 1.08% (-95.1%) from 96.17%

Build # 9dbcf7a4-1c56-4510-9614-74ad5a22cff6

Build Type

push

circleci

Committed by

jrgriffiniii

Commit Message

wip

Run Details

52 of 4814 relevant lines covered (1.08%)

0.01 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/app/models/work.rb

# frozen_string_literal: true

# rubocop:disable Metrics/ClassLength
class Work < ApplicationRecord
  # Errors for cases where there is no valid Group
  class InvalidGroupError < ::ArgumentError; end

  has_many :work_activity, -> { order(updated_at: :desc) }, dependent: :destroy
  has_many :user_work, -> { order(updated_at: :desc) }, dependent: :destroy
  has_many :upload_snapshots, -> { order(updated_at: :desc) }, dependent: :destroy

  belongs_to :group, class_name: "Group"
  belongs_to :curator, class_name: "User", foreign_key: "curator_user_id", optional: true

  attribute :work_type, :string, default: "DATASET"
  attribute :profile, :string, default: "DATACITE"

  attr_accessor :user_entered_doi

  alias state_history user_work

  delegate :valid_to_submit, :valid_to_draft, :valid_to_approve, :valid_to_complete, to: :work_validator

  include AASM

  aasm column: :state do
    state :none, initial: true
    state :draft, :awaiting_approval, :approved, :withdrawn, :deletion_marker

    event :draft, after: :draft_doi do
      transitions from: :none, to: :draft, guard: :valid_to_draft
    end

    event :complete_submission do
      transitions from: :draft, to: :awaiting_approval, guard: :valid_to_complete
    end

    event :request_changes do
      transitions from: :awaiting_approval, to: :awaiting_approval, guard: :valid_to_submit
    end

    event :approve do
      transitions from: :awaiting_approval, to: :approved, guard: :valid_to_approve, after: :publish
    end

    event :withdraw do
      transitions from: [:draft, :awaiting_approval, :approved], to: :withdrawn
    end

    event :resubmit do
      transitions from: :withdrawn, to: :draft
    end

    event :remove do
      transitions from: :withdrawn, to: :deletion_marker
    end

    after_all_events :track_state_change
  end

  def state=(new_state)
    new_state_sym = new_state.to_sym
    valid_states = self.class.aasm.states.map(&:name)
    raise(StandardError, "Invalid state '#{new_state}'") unless valid_states.include?(new_state_sym)
    aasm_write_state_without_persistence(new_state_sym)
  end

  ##
  # Is this work editable by a given user?
  # A work is editable when:
  # * it is being edited by the person who made it
  # * it is being edited by a group admin of the group where is resides
  # * it is being edited by a super admin
  # @param [User]
  # @return [Boolean]
  def editable_by?(user)
    submitted_by?(user) || administered_by?(user)
  end

  def editable_in_current_state?(user)
    # anyone with edit privleges can edit a work while it is in draft
    return editable_by?(user) if draft?

    # Only admisitrators can edit a work in other states
    administered_by?(user)
  end

  def submitted_by?(user)
    created_by_user_id == user.id
  end

  def administered_by?(user)
    user.has_role?(:group_admin, group)
  end

  class << self
    def find_by_doi(doi)
      prefix = "10.34770/"
      doi = "#{prefix}#{doi}" unless doi.blank? || doi.start_with?(prefix)
      Work.find_by!("metadata @> ?", JSON.dump(doi:))
    end

    def find_by_ark(ark)
      prefix = "ark:/"
      ark = "#{prefix}#{ark}" unless ark.blank? || ark.start_with?(prefix)
      Work.find_by!("metadata @> ?", JSON.dump(ark:))
    end

    delegate :resource_type_general_values, to: PDCMetadata::Resource
  end

  include Rails.application.routes.url_helpers

  before_save do |work|
    # Ensure that the metadata JSONB postgres field is persisted properly
    work.metadata = JSON.parse(work.resource.to_json)
  end

  after_save do |work|
    if work.approved?
      work.reload
    end
  end

  validate do |_work|
    work_validator.valid?
  end

  # Overload ActiveRecord.reload method
  # https://apidock.com/rails/ActiveRecord/Base/reload
  #
  # NOTE: Usually `after_save` is a better place to put this kind of code:
  #
  #   after_save do |work|
  #     work.resource = nil
  #   end
  #
  # but that does not work in this case because the block points to a different
  # memory object for `work` than the we want we want to reload.
  def reload(options = nil)
    super
    # Force `resource` to be reloaded
    @resource = nil
    self
  end

  def title
    resource.main_title
  end

  def uploads_attributes
    return [] if approved? # once approved we no longer allow the updating of uploads via the application
    uploads.map do |upload|
      {
        id: upload.id,
        key: upload.key,
        filename: upload.filename.to_s,
        created_at: upload.created_at,
        url: upload.url
      }
    end
  end

  def form_attributes
    {
      uploads: uploads_attributes
    }
  end

  def draft_doi
    return if resource.doi.present?
    resource.doi = datacite_service.draft_doi
    save!
  end

  # Return the DOI formatted as a URL, so it can be used as a link on display pages
  # @return [String] A url formatted version of the DOI
  def doi_url
    return "https://doi.org/#{doi}" unless doi.starts_with?("https://doi.org")
    doi
  end

  def created_by_user
    User.find(created_by_user_id)
  rescue ActiveRecord::RecordNotFound
    nil
  end

  def resource=(resource)
    @resource = resource
    # Ensure that the metadata JSONB postgres field is persisted properly
    self.metadata = JSON.parse(resource.to_json)
  end

  def resource
    @resource ||= PDCMetadata::Resource.new_from_jsonb(metadata)
  end

  def url
    return unless persisted?

    @url ||= url_for(self)
  end

  def files_location_upload?
    files_location.blank? || files_location == "file_upload"
  end

  def files_location_cluster?
    files_location == "file_cluster"
  end

  def files_location_other?
    files_location == "file_other"
  end

  def change_curator(curator_user_id, current_user)
    if curator_user_id == "no-one"
      clear_curator(current_user)
    else
      update_curator(curator_user_id, current_user)
    end
  end

  def clear_curator(current_user)
    # Update the curator on the Work
    self.curator_user_id = nil
    save!

    # ...and log the activity
    WorkActivity.add_work_activity(id, "Unassigned existing curator", current_user.id, activity_type: WorkActivity::SYSTEM)
  end

  def update_curator(curator_user_id, current_user)
    # Update the curator on the Work
    self.curator_user_id = curator_user_id
    save!

    # ...and log the activity
    new_curator = User.find(curator_user_id)

    work_url = "[#{title}](#{Rails.application.routes.url_helpers.work_url(self)})"

    # Troubleshooting https://github.com/pulibrary/pdc_describe/issues/1783
    if work_url.include?("/describe/describe/")
      Rails.logger.error("URL #{work_url} included /describe/describe/ and was fixed. See https://github.com/pulibrary/pdc_describe/issues/1783")
      work_url = work_url.gsub("/describe/describe/", "/describe/")
    end

    message = if curator_user_id.to_i == current_user.id
                "Self-assigned @#{current_user.uid} as curator for work #{work_url}"
              else
                "Set curator to @#{new_curator.uid} for work #{work_url}"
              end
    WorkActivity.add_work_activity(id, message, current_user.id, activity_type: WorkActivity::SYSTEM)
  end

  def add_message(message, current_user_id)
    WorkActivity.add_work_activity(id, message, current_user_id, activity_type: WorkActivity::MESSAGE)
  end

  def add_provenance_note(date, note, current_user_id, change_label = "")
    WorkActivity.add_work_activity(id, { note:, change_label: }.to_json, current_user_id, activity_type: WorkActivity::PROVENANCE_NOTES, created_at: date)
    # WorkActivity.add_work_activity(id, note, current_user_id, activity_type: WorkActivity::PROVENANCE_NOTES, created_at: date)
  end

  def log_changes(resource_compare, current_user_id)
    return if resource_compare.identical?
    WorkActivity.add_work_activity(id, resource_compare.differences.to_json, current_user_id, activity_type: WorkActivity::CHANGES)
  end

  def log_file_changes(current_user_id)
    return if changes.count == 0
    WorkActivity.add_work_activity(id, changes.to_json, current_user_id, activity_type: WorkActivity::FILE_CHANGES)
  end

  def activities
    WorkActivity.activities_for_work(id, WorkActivity::MESSAGE_ACTIVITY_TYPES + WorkActivity::CHANGE_LOG_ACTIVITY_TYPES)
  end

  def new_notification_count_for_user(user_id)
    WorkActivityNotification.joins(:work_activity)
                            .where(user_id:, read_at: nil)
                            .where(work_activity: { work_id: id })
                            .count
  end

  # Marks as read the notifications for the given user_id in this work.
  # In practice, the user_id is the id of the current user and therefore this method marks the current's user
  # notifications as read.
  def mark_new_notifications_as_read(user_id)
    activities.each do |activity|
      unread_notifications = WorkActivityNotification.where(user_id:, work_activity_id: activity.id, read_at: nil)
      unread_notifications.each do |notification|
        notification.read_at = Time.now.utc
        notification.save
      end
    end
  end

  def current_transition
    aasm.current_event.to_s.humanize.delete("!")
  end

  # Retrieve the S3 file uploads associated with the Work
  # @return [Array<S3File>]
  def uploads
    return post_curation_uploads if approved?

    pre_curation_uploads
  end

  # Retrieve the S3 file uploads named "README"
  # @return [Array<S3File>]
  def readme_uploads
    uploads.select { |s3_file| s3_file.filename.include?("README") }
  end

  # Retrieve the S3 file uploads which are research artifacts proper (not README or other files providing metadata/documentation)
  # @return [Array<S3File>]
  def artifact_uploads
    uploads.reject { |s3_file| s3_file.filename.include?("README") }
  end

  # Returns the list of files for the work with some basic information about each of them.
  # This method is much faster than `uploads` because it does not return the actual S3File
  # objects to the client, instead it returns just a few selected data elements.
  def file_list
    s3_files = approved? ? post_curation_uploads : pre_curation_uploads
    files_info = s3_files.map do |s3_file|
      {
        "safe_id": s3_file.safe_id,
        "filename": s3_file.filename,
        "filename_display": s3_file.filename_display,
        "last_modified": s3_file.last_modified,
        "last_modified_display": s3_file.last_modified_display,
        "size": s3_file.size,
        "display_size": s3_file.display_size,
        "url": s3_file.url,
        "is_folder": s3_file.is_folder
      }
    end
    files_info
  end

  def total_file_size
    @total_file_size ||= begin
      total_size = 0
      file_list.each do |file|
        total_size += file[:size]
      end
      total_size
    end
  end

  # Fetches the data from S3 directly bypassing ActiveStorage
  def pre_curation_uploads
    s3_query_service.client_s3_files.sort_by(&:filename)
  end

  # Accesses post-curation S3 Bucket Objects
  def post_curation_s3_resources
    if approved?
      s3_resources
    else
      []
    end
  end

  # Returns the files in post-curation for the work
  def post_curation_uploads(force_post_curation: false)
    if force_post_curation
      # Always use the post-curation data regardless of the work's status
      post_curation_s3_query_service = S3QueryService.new(self, "postcuration")
      post_curation_s3_query_service.data_profile.fetch(:objects, [])
    else
      # Return the list based of files honoring the work status
      post_curation_s3_resources
    end
  end

  def s3_files
    pre_curation_uploads
  end

  def s3_client
    s3_query_service.client
  end

  delegate :bucket_name, :prefix, to: :s3_query_service
  delegate :doi_attribute_url, :curator_or_current_uid, to: :datacite_service

  # Generates the S3 Object key
  # @return [String]
  def s3_object_key
    "#{doi}/#{id}"
  end

  # Transmit a HEAD request for the S3 Bucket directory for this Work
  # @param bucket_name location to be checked to be found
  # @return [Aws::S3::Types::HeadObjectOutput]
  def find_post_curation_s3_dir(bucket_name:)
    # TODO: Directories really do not exists in S3
    #      if we really need this check then we need to do something else to check the bucket
    s3_client.head_object({
                            bucket: bucket_name,
                            key: s3_object_key
                          })
    true
  rescue Aws::S3::Errors::NotFound
    nil
  end

  # Generates the JSON serialized expression of the Work
  # @param args [Array<Hash>]
  # @option args [Boolean] :force_post_curation Force the request of AWS S3
  #   Resources, clearing the in-memory cache
  # @return [String]
  def as_json(*args)
    files = files_as_json(*args)

    # to_json returns a string of serialized JSON.
    # as_json returns the corresponding hash.
    {
      "resource" => resource.as_json,
      "files" => files,
      "group" => group.as_json.except("id"),
      "embargo_date" => embargo_date_as_json,
      "created_at" => format_date_for_solr(created_at),
      "updated_at" => format_date_for_solr(updated_at)
    }
  end

  # Format the date for Apache Solr
  # @param date [ActiveSupport::TimeWithZone]
  # @return [String]
  def format_date_for_solr(date)
    date.strftime("%Y-%m-%dT%H:%M:%SZ")
  end

  def pre_curation_uploads_count
    s3_query_service.file_count
  end

  delegate :ark, :doi, :resource_type, :resource_type=, :resource_type_general, :resource_type_general=,
           :to_xml, to: :resource

  # S3QueryService object associated with this Work
  # @return [S3QueryService]
  def s3_query_service
    mode = approved? ? "postcuration" : "precuration"
    @s3_query_service ||= S3QueryService.new(self, mode)
  end

  def past_snapshots
    UploadSnapshot.where(work: self)
  end

  # Build or find persisted UploadSnapshot models for this Work
  # @param [integer] user_id optional user to assign the snapshot to
  # @return [UploadSnapshot]
  def reload_snapshots(user_id: nil)
    work_changes = []
    s3_files = pre_curation_uploads
    s3_filenames = s3_files.map(&:filename)

    upload_snapshot = latest_snapshot

    upload_snapshot.snapshot_deletions(work_changes, s3_filenames)

    upload_snapshot.snapshot_modifications(work_changes, s3_files)

    # Create WorkActivity models with the set of changes
    unless work_changes.empty?
      new_snapshot = UploadSnapshot.new(work: self, url: s3_query_service.prefix)
      new_snapshot.store_files(s3_files)
      new_snapshot.save!
      WorkActivity.add_work_activity(id, work_changes.to_json, user_id, activity_type: WorkActivity::FILE_CHANGES)
    end
  end

  def self.presenter_class
    WorkPresenter
  end

  def presenter
    self.class.presenter_class.new(work: self)
  end

  def changes
    @changes ||= []
  end

  def track_change(action, filename)
    changes << { action:, filename: }
  end

  # rubocop:disable Naming/PredicateName
  def has_rights?(rights_id)
    resource.rights_many.index { |rights| rights.identifier == rights_id } != nil
  end
  # rubocop:enable Naming/PredicateName

  # This is the solr id / work show page in PDC Discovery
  def pdc_discovery_url
    "https://datacommons.princeton.edu/discovery/catalog/doi-#{doi.tr('/', '-').tr('.', '-')}"
  end

  # Determine whether or not the Work is under active embargo
  # @return [Boolean]
  def embargoed?
    return false if embargo_date.blank?

    current_date = Time.zone.now
    embargo_date >= current_date
  end

  def upload_count
    @upload_count ||= s3_query_service.count_objects
  end

  protected

    def work_validator
      @work_validator ||= WorkValidator.new(self)
    end

    # This must be protected, NOT private for ActiveRecord to work properly with this attribute.
    #   Protected will still keep others from setting the metatdata, but allows ActiveRecord the access it needs
    def metadata=(metadata)
      super
      @resource = PDCMetadata::Resource.new_from_jsonb(metadata)
    end

  private

    def publish(user)
      datacite_service.publish_doi(user)
      update_ark_information
      publish_precurated_files(user)
      save!
    end

    # Update EZID (our provider of ARKs) with the new information for this work.
    def update_ark_information
      # We only want to update the ark url under certain conditions.
      # Set this value in config/update_ark_url.yml
      if Rails.configuration.update_ark_url
        if ark.present?
          Ark.update(ark, datacite_service.doi_attribute_url)
        end
      end
    end

    def track_state_change(user, state = aasm.to_state)
      uw = UserWork.new(user_id: user.id, work_id: id, state:)
      uw.save!
      WorkActivity.add_work_activity(id, "marked as #{state.to_s.titleize}", user.id, activity_type: WorkActivity::SYSTEM)
      WorkStateTransitionNotification.new(self, user.id).send
    end

    # Request S3 Bucket Objects associated with this Work
    # @return [Array<S3File>]
    def s3_resources
      data_profile = s3_query_service.data_profile
      data_profile.fetch(:objects, [])
    end
    alias pre_curation_s3_resources s3_resources

    def s3_object_persisted?(s3_file)
      uploads_keys = uploads.map(&:key)
      uploads_keys.include?(s3_file.key)
    end

    def publish_precurated_files(user)
      # We need to explicitly check the to post-curation bucket here.
      s3_post_curation_query_service = S3QueryService.new(self, "postcuration")

      s3_dir = find_post_curation_s3_dir(bucket_name: s3_post_curation_query_service.bucket_name)
      raise(StandardError, "Attempting to publish a Work with an existing S3 Bucket directory for: #{s3_object_key}") unless s3_dir.nil?

      # Copy the pre-curation S3 Objects to the post-curation S3 Bucket...
      s3_query_service.publish_files(user)
    end

    def latest_snapshot
      return upload_snapshots.first unless upload_snapshots.empty?

      UploadSnapshot.new(work: self, files: [])
    end

    def datacite_service
      @datacite_service ||= PULDatacite.new(self)
    end

    def files_as_json(*args)
      return [] if embargoed?

      force_post_curation = args.any? { |arg| arg[:force_post_curation] == true }

      # Pre-curation files are not accessible externally,
      # so we are not interested in listing them in JSON.
      post_curation_uploads(force_post_curation:).map do |upload|
        {
          "filename": upload.filename,
          "size": upload.size,
          "display_size": upload.display_size,
          "url": upload.globus_url
        }
      end
    end

    def embargo_date_as_json
      if embargo_date.present?
        embargo_datetime = embargo_date.to_datetime
        embargo_date_iso8601 = embargo_datetime.iso8601
        # Apache Solr timestamps require the following format:
        # 1972-05-20T17:33:18Z
        # https://solr.apache.org/guide/solr/latest/indexing-guide/date-formatting-math.html
        embargo_date_iso8601.gsub(/\+.+$/, "Z")
      end
    end
end
# rubocop:enable Metrics/ClassLength

1	# frozen_string_literal: true
2
3	# rubocop:disable Metrics/ClassLength
4	class Work < ApplicationRecord	×
5	# Errors for cases where there is no valid Group
6	class InvalidGroupError < ::ArgumentError; end	×
7
8	has_many :work_activity, -> { order(updated_at: :desc) }, dependent: :destroy	×
9	has_many :user_work, -> { order(updated_at: :desc) }, dependent: :destroy	×
10	has_many :upload_snapshots, -> { order(updated_at: :desc) }, dependent: :destroy	×
11
12	belongs_to :group, class_name: "Group"	×
13	belongs_to :curator, class_name: "User", foreign_key: "curator_user_id", optional: true	×
14
15	attribute :work_type, :string, default: "DATASET"	×
16	attribute :profile, :string, default: "DATACITE"	×
17
18	attr_accessor :user_entered_doi	×
19
20	alias state_history user_work	×
21
22	delegate :valid_to_submit, :valid_to_draft, :valid_to_approve, :valid_to_complete, to: :work_validator	×
23
24	include AASM	×
25
26	aasm column: :state do	×
27	state :none, initial: true	×
28	state :draft, :awaiting_approval, :approved, :withdrawn, :deletion_marker	×
29
30	event :draft, after: :draft_doi do	×
31	transitions from: :none, to: :draft, guard: :valid_to_draft	×
32	end	×
33
34	event :complete_submission do	×
35	transitions from: :draft, to: :awaiting_approval, guard: :valid_to_complete	×
36	end	×
37
38	event :request_changes do	×
39	transitions from: :awaiting_approval, to: :awaiting_approval, guard: :valid_to_submit	×
40	end	×
41
42	event :approve do	×
43	transitions from: :awaiting_approval, to: :approved, guard: :valid_to_approve, after: :publish	×
44	end	×
45
46	event :withdraw do	×
47	transitions from: [:draft, :awaiting_approval, :approved], to: :withdrawn	×
48	end	×
49
50	event :resubmit do	×
51	transitions from: :withdrawn, to: :draft	×
52	end	×
53
54	event :remove do	×
55	transitions from: :withdrawn, to: :deletion_marker	×
56	end	×
57
58	after_all_events :track_state_change	×
59	end	×
60
61	def state=(new_state)	×
62	new_state_sym = new_state.to_sym	×
63	valid_states = self.class.aasm.states.map(&:name)	×
64	raise(StandardError, "Invalid state '#{new_state}'") unless valid_states.include?(new_state_sym)	×
65	aasm_write_state_without_persistence(new_state_sym)	×
66	end	×
67
68	##
69	# Is this work editable by a given user?
70	# A work is editable when:
71	# * it is being edited by the person who made it
72	# * it is being edited by a group admin of the group where is resides
73	# * it is being edited by a super admin
74	# @param [User]
75	# @return [Boolean]
76	def editable_by?(user)	×
77	submitted_by?(user) \|\| administered_by?(user)	×
78	end	×
79
80	def editable_in_current_state?(user)	×
81	# anyone with edit privleges can edit a work while it is in draft
82	return editable_by?(user) if draft?	×
83
84	# Only admisitrators can edit a work in other states
85	administered_by?(user)	×
86	end	×
87
88	def submitted_by?(user)	×
89	created_by_user_id == user.id	×
90	end	×
91
92	def administered_by?(user)	×
93	user.has_role?(:group_admin, group)	×
94	end	×
95
96	class << self	×
97	def find_by_doi(doi)	×
98	prefix = "10.34770/"	×
99	doi = "#{prefix}#{doi}" unless doi.blank? \|\| doi.start_with?(prefix)	×
100	Work.find_by!("metadata @> ?", JSON.dump(doi:))	×
101	end	×
102
103	def find_by_ark(ark)	×
104	prefix = "ark:/"	×
105	ark = "#{prefix}#{ark}" unless ark.blank? \|\| ark.start_with?(prefix)	×
106	Work.find_by!("metadata @> ?", JSON.dump(ark:))	×
107	end	×
108
109	delegate :resource_type_general_values, to: PDCMetadata::Resource	×
110	end	×
111
112	include Rails.application.routes.url_helpers	×
113
114	before_save do \|work\|	×
115	# Ensure that the metadata JSONB postgres field is persisted properly
116	work.metadata = JSON.parse(work.resource.to_json)	×
117	end	×
118
119	after_save do \|work\|	×
120	if work.approved?	×
121	work.reload	×
122	end	×
123	end	×
124
125	validate do \|_work\|	×
126	work_validator.valid?	×
127	end	×
128
129	# Overload ActiveRecord.reload method
130	# https://apidock.com/rails/ActiveRecord/Base/reload
131	#
132	# NOTE: Usually `after_save` is a better place to put this kind of code:
133	#
134	# after_save do \|work\|
135	# work.resource = nil
136	# end
137	#
138	# but that does not work in this case because the block points to a different
139	# memory object for `work` than the we want we want to reload.
140	def reload(options = nil)	×
141	super	×
142	# Force `resource` to be reloaded
143	@resource = nil	×
144	self	×
145	end	×
146
147	def title	×
148	resource.main_title	×
149	end	×
150
151	def uploads_attributes	×
152	return [] if approved? # once approved we no longer allow the updating of uploads via the application	×
153	uploads.map do \|upload\|	×
154	{	×
155	id: upload.id,	×
156	key: upload.key,	×
157	filename: upload.filename.to_s,	×
158	created_at: upload.created_at,	×
159	url: upload.url	×
160	}	×
161	end	×
162	end	×
163
164	def form_attributes	×
165	{	×
166	uploads: uploads_attributes	×
167	}	×
168	end	×
169
170	def draft_doi	×
171	return if resource.doi.present?	×
172	resource.doi = datacite_service.draft_doi	×
173	save!	×
174	end	×
175
176	# Return the DOI formatted as a URL, so it can be used as a link on display pages
177	# @return [String] A url formatted version of the DOI
178	def doi_url	×
179	return "https://doi.org/#{doi}" unless doi.starts_with?("https://doi.org")	×
180	doi	×
181	end	×
182
183	def created_by_user	×
184	User.find(created_by_user_id)	×
185	rescue ActiveRecord::RecordNotFound	×
186	nil	×
187	end	×
188
189	def resource=(resource)	×
190	@resource = resource	×
191	# Ensure that the metadata JSONB postgres field is persisted properly
192	self.metadata = JSON.parse(resource.to_json)	×
193	end	×
194
195	def resource	×
196	@resource \|\|= PDCMetadata::Resource.new_from_jsonb(metadata)	×
197	end	×
198
199	def url	×
200	return unless persisted?	×
201
202	@url \|\|= url_for(self)	×
203	end	×
204
205	def files_location_upload?	×
206	files_location.blank? \|\| files_location == "file_upload"	×
207	end	×
208
209	def files_location_cluster?	×
210	files_location == "file_cluster"	×
211	end	×
212
213	def files_location_other?	×
214	files_location == "file_other"	×
215	end	×
216
217	def change_curator(curator_user_id, current_user)	×
218	if curator_user_id == "no-one"	×
219	clear_curator(current_user)	×
220	else	×
221	update_curator(curator_user_id, current_user)	×
222	end	×
223	end	×
224
225	def clear_curator(current_user)	×
226	# Update the curator on the Work
227	self.curator_user_id = nil	×
228	save!	×
229
230	# ...and log the activity
231	WorkActivity.add_work_activity(id, "Unassigned existing curator", current_user.id, activity_type: WorkActivity::SYSTEM)	×
232	end	×
233
234	def update_curator(curator_user_id, current_user)	×
235	# Update the curator on the Work
236	self.curator_user_id = curator_user_id	×
237	save!	×
238
239	# ...and log the activity
240	new_curator = User.find(curator_user_id)	×
241
242	work_url = "[#{title}](#{Rails.application.routes.url_helpers.work_url(self)})"	×
243
244	# Troubleshooting https://github.com/pulibrary/pdc_describe/issues/1783
245	if work_url.include?("/describe/describe/")	×
246	Rails.logger.error("URL #{work_url} included /describe/describe/ and was fixed. See https://github.com/pulibrary/pdc_describe/issues/1783")	×
247	work_url = work_url.gsub("/describe/describe/", "/describe/")	×
248	end	×
249
250	message = if curator_user_id.to_i == current_user.id	×
251	"Self-assigned @#{current_user.uid} as curator for work #{work_url}"	×
252	else	×
253	"Set curator to @#{new_curator.uid} for work #{work_url}"	×
254	end	×
255	WorkActivity.add_work_activity(id, message, current_user.id, activity_type: WorkActivity::SYSTEM)	×
256	end	×
257
258	def add_message(message, current_user_id)	×
259	WorkActivity.add_work_activity(id, message, current_user_id, activity_type: WorkActivity::MESSAGE)	×
260	end	×
261
262	def add_provenance_note(date, note, current_user_id, change_label = "")	×
263	WorkActivity.add_work_activity(id, { note:, change_label: }.to_json, current_user_id, activity_type: WorkActivity::PROVENANCE_NOTES, created_at: date)	×
264	# WorkActivity.add_work_activity(id, note, current_user_id, activity_type: WorkActivity::PROVENANCE_NOTES, created_at: date)
265	end	×
266
267	def log_changes(resource_compare, current_user_id)	×
268	return if resource_compare.identical?	×
269	WorkActivity.add_work_activity(id, resource_compare.differences.to_json, current_user_id, activity_type: WorkActivity::CHANGES)	×
270	end	×
271
272	def log_file_changes(current_user_id)	×
273	return if changes.count == 0	×
274	WorkActivity.add_work_activity(id, changes.to_json, current_user_id, activity_type: WorkActivity::FILE_CHANGES)	×
275	end	×
276
277	def activities	×
278	WorkActivity.activities_for_work(id, WorkActivity::MESSAGE_ACTIVITY_TYPES + WorkActivity::CHANGE_LOG_ACTIVITY_TYPES)	×
279	end	×
280
281	def new_notification_count_for_user(user_id)	×
282	WorkActivityNotification.joins(:work_activity)	×
283	.where(user_id:, read_at: nil)	×
284	.where(work_activity: { work_id: id })	×
285	.count	×
286	end	×
287
288	# Marks as read the notifications for the given user_id in this work.
289	# In practice, the user_id is the id of the current user and therefore this method marks the current's user
290	# notifications as read.
291	def mark_new_notifications_as_read(user_id)	×
292	activities.each do \|activity\|	×
293	unread_notifications = WorkActivityNotification.where(user_id:, work_activity_id: activity.id, read_at: nil)	×
294	unread_notifications.each do \|notification\|	×
295	notification.read_at = Time.now.utc	×
296	notification.save	×
297	end	×
298	end	×
299	end	×
300
301	def current_transition	×
302	aasm.current_event.to_s.humanize.delete("!")	×
303	end	×
304
305	# Retrieve the S3 file uploads associated with the Work
306	# @return [Array<S3File>]
307	def uploads	×
308	return post_curation_uploads if approved?	×
309
310	pre_curation_uploads	×
311	end	×
312
313	# Retrieve the S3 file uploads named "README"
314	# @return [Array<S3File>]
315	def readme_uploads	×
316	uploads.select { \|s3_file\| s3_file.filename.include?("README") }	×
317	end	×
318
319	# Retrieve the S3 file uploads which are research artifacts proper (not README or other files providing metadata/documentation)
320	# @return [Array<S3File>]
321	def artifact_uploads	×
322	uploads.reject { \|s3_file\| s3_file.filename.include?("README") }	×
323	end	×
324
325	# Returns the list of files for the work with some basic information about each of them.
326	# This method is much faster than `uploads` because it does not return the actual S3File
327	# objects to the client, instead it returns just a few selected data elements.
328	def file_list	×
329	s3_files = approved? ? post_curation_uploads : pre_curation_uploads	×
330	files_info = s3_files.map do \|s3_file\|	×
331	{	×
332	"safe_id": s3_file.safe_id,	×
333	"filename": s3_file.filename,	×
334	"filename_display": s3_file.filename_display,	×
335	"last_modified": s3_file.last_modified,	×
336	"last_modified_display": s3_file.last_modified_display,	×
337	"size": s3_file.size,	×
338	"display_size": s3_file.display_size,	×
339	"url": s3_file.url,	×
340	"is_folder": s3_file.is_folder	×
341	}	×
342	end	×
343	files_info	×
344	end	×
345
346	def total_file_size	×
347	@total_file_size \|\|= begin	×
348	total_size = 0	×
349	file_list.each do \|file\|	×
350	total_size += file[:size]	×
351	end	×
352	total_size	×
353	end	×
354	end	×
355
356	# Fetches the data from S3 directly bypassing ActiveStorage
357	def pre_curation_uploads	×
358	s3_query_service.client_s3_files.sort_by(&:filename)	×
359	end	×
360
361	# Accesses post-curation S3 Bucket Objects
362	def post_curation_s3_resources	×
363	if approved?	×
364	s3_resources	×
365	else	×
366	[]	×
367	end	×
368	end	×
369
370	# Returns the files in post-curation for the work
371	def post_curation_uploads(force_post_curation: false)	×
372	if force_post_curation	×
373	# Always use the post-curation data regardless of the work's status
374	post_curation_s3_query_service = S3QueryService.new(self, "postcuration")	×
375	post_curation_s3_query_service.data_profile.fetch(:objects, [])	×
376	else	×
377	# Return the list based of files honoring the work status
378	post_curation_s3_resources	×
379	end	×
380	end	×
381
382	def s3_files	×
383	pre_curation_uploads	×
384	end	×
385
386	def s3_client	×
387	s3_query_service.client	×
388	end	×
389
390	delegate :bucket_name, :prefix, to: :s3_query_service	×
391	delegate :doi_attribute_url, :curator_or_current_uid, to: :datacite_service	×
392
393	# Generates the S3 Object key
394	# @return [String]
395	def s3_object_key	×
396	"#{doi}/#{id}"	×
397	end	×
398
399	# Transmit a HEAD request for the S3 Bucket directory for this Work
400	# @param bucket_name location to be checked to be found
401	# @return [Aws::S3::Types::HeadObjectOutput]
402	def find_post_curation_s3_dir(bucket_name:)	×
403	# TODO: Directories really do not exists in S3
404	# if we really need this check then we need to do something else to check the bucket
405	s3_client.head_object({	×
406	bucket: bucket_name,	×
407	key: s3_object_key	×
408	})	×
409	true	×
410	rescue Aws::S3::Errors::NotFound	×
411	nil	×
412	end	×
413
414	# Generates the JSON serialized expression of the Work
415	# @param args [Array<Hash>]
416	# @option args [Boolean] :force_post_curation Force the request of AWS S3
417	# Resources, clearing the in-memory cache
418	# @return [String]
419	def as_json(*args)	×
420	files = files_as_json(*args)	×
421
422	# to_json returns a string of serialized JSON.
423	# as_json returns the corresponding hash.
424	{	×
425	"resource" => resource.as_json,	×
426	"files" => files,	×
427	"group" => group.as_json.except("id"),	×
428	"embargo_date" => embargo_date_as_json,	×
429	"created_at" => format_date_for_solr(created_at),	×
430	"updated_at" => format_date_for_solr(updated_at)	×
431	}	×
432	end	×
433
434	# Format the date for Apache Solr
435	# @param date [ActiveSupport::TimeWithZone]
436	# @return [String]
437	def format_date_for_solr(date)	×
438	date.strftime("%Y-%m-%dT%H:%M:%SZ")	×
439	end	×
440
441	def pre_curation_uploads_count	×
442	s3_query_service.file_count	×
443	end	×
444
445	delegate :ark, :doi, :resource_type, :resource_type=, :resource_type_general, :resource_type_general=,	×
446	:to_xml, to: :resource	×
447
448	# S3QueryService object associated with this Work
449	# @return [S3QueryService]
450	def s3_query_service	×
451	mode = approved? ? "postcuration" : "precuration"	×
452	@s3_query_service \|\|= S3QueryService.new(self, mode)	×
453	end	×
454
455	def past_snapshots	×
456	UploadSnapshot.where(work: self)	×
457	end	×
458
459	# Build or find persisted UploadSnapshot models for this Work
460	# @param [integer] user_id optional user to assign the snapshot to
461	# @return [UploadSnapshot]
462	def reload_snapshots(user_id: nil)	×
463	work_changes = []	×
464	s3_files = pre_curation_uploads	×
465	s3_filenames = s3_files.map(&:filename)	×
466
467	upload_snapshot = latest_snapshot	×
468
469	upload_snapshot.snapshot_deletions(work_changes, s3_filenames)	×
470
471	upload_snapshot.snapshot_modifications(work_changes, s3_files)	×
472
473	# Create WorkActivity models with the set of changes
474	unless work_changes.empty?	×
475	new_snapshot = UploadSnapshot.new(work: self, url: s3_query_service.prefix)	×
476	new_snapshot.store_files(s3_files)	×
477	new_snapshot.save!	×
478	WorkActivity.add_work_activity(id, work_changes.to_json, user_id, activity_type: WorkActivity::FILE_CHANGES)	×
479	end	×
480	end	×
481
482	def self.presenter_class	×
483	WorkPresenter	×
484	end	×
485
486	def presenter	×
487	self.class.presenter_class.new(work: self)	×
488	end	×
489
490	def changes	×
491	@changes \|\|= []	×
492	end	×
493
494	def track_change(action, filename)	×
495	changes << { action:, filename: }	×
496	end	×
497
498	# rubocop:disable Naming/PredicateName
499	def has_rights?(rights_id)	×
500	resource.rights_many.index { \|rights\| rights.identifier == rights_id } != nil	×
501	end	×
502	# rubocop:enable Naming/PredicateName
503
504	# This is the solr id / work show page in PDC Discovery
505	def pdc_discovery_url	×
506	"https://datacommons.princeton.edu/discovery/catalog/doi-#{doi.tr('/', '-').tr('.', '-')}"	×
507	end	×
508
509	# Determine whether or not the Work is under active embargo
510	# @return [Boolean]
511	def embargoed?	×
512	return false if embargo_date.blank?	×
513
514	current_date = Time.zone.now	×
515	embargo_date >= current_date	×
516	end	×
517
518	def upload_count	×
519	@upload_count \|\|= s3_query_service.count_objects	×
520	end	×
521
522	protected	×
523
524	def work_validator	×
525	@work_validator \|\|= WorkValidator.new(self)	×
526	end	×
527
528	# This must be protected, NOT private for ActiveRecord to work properly with this attribute.
529	# Protected will still keep others from setting the metatdata, but allows ActiveRecord the access it needs
530	def metadata=(metadata)	×
531	super	×
532	@resource = PDCMetadata::Resource.new_from_jsonb(metadata)	×
533	end	×
534
535	private	×
536
537	def publish(user)	×
538	datacite_service.publish_doi(user)	×
539	update_ark_information	×
540	publish_precurated_files(user)	×
541	save!	×
542	end	×
543
544	# Update EZID (our provider of ARKs) with the new information for this work.
545	def update_ark_information	×
546	# We only want to update the ark url under certain conditions.
547	# Set this value in config/update_ark_url.yml
548	if Rails.configuration.update_ark_url	×
549	if ark.present?	×
550	Ark.update(ark, datacite_service.doi_attribute_url)	×
551	end	×
552	end	×
553	end	×
554
555	def track_state_change(user, state = aasm.to_state)	×
556	uw = UserWork.new(user_id: user.id, work_id: id, state:)	×
557	uw.save!	×
558	WorkActivity.add_work_activity(id, "marked as #{state.to_s.titleize}", user.id, activity_type: WorkActivity::SYSTEM)	×
559	WorkStateTransitionNotification.new(self, user.id).send	×
560	end	×
561
562	# Request S3 Bucket Objects associated with this Work
563	# @return [Array<S3File>]
564	def s3_resources	×
565	data_profile = s3_query_service.data_profile	×
566	data_profile.fetch(:objects, [])	×
567	end	×
568	alias pre_curation_s3_resources s3_resources	×
569
570	def s3_object_persisted?(s3_file)	×
571	uploads_keys = uploads.map(&:key)	×
572	uploads_keys.include?(s3_file.key)	×
573	end	×
574
575	def publish_precurated_files(user)	×
576	# We need to explicitly check the to post-curation bucket here.
577	s3_post_curation_query_service = S3QueryService.new(self, "postcuration")	×
578
579	s3_dir = find_post_curation_s3_dir(bucket_name: s3_post_curation_query_service.bucket_name)	×
580	raise(StandardError, "Attempting to publish a Work with an existing S3 Bucket directory for: #{s3_object_key}") unless s3_dir.nil?	×
581
582	# Copy the pre-curation S3 Objects to the post-curation S3 Bucket...
583	s3_query_service.publish_files(user)	×
584	end	×
585
586	def latest_snapshot	×
587	return upload_snapshots.first unless upload_snapshots.empty?	×
588
589	UploadSnapshot.new(work: self, files: [])	×
590	end	×
591
592	def datacite_service	×
593	@datacite_service \|\|= PULDatacite.new(self)	×
594	end	×
595
596	def files_as_json(*args)	×
597	return [] if embargoed?	×
598
599	force_post_curation = args.any? { \|arg\| arg[:force_post_curation] == true }	×
600
601	# Pre-curation files are not accessible externally,
602	# so we are not interested in listing them in JSON.
603	post_curation_uploads(force_post_curation:).map do \|upload\|	×
604	{	×
605	"filename": upload.filename,	×
606	"size": upload.size,	×
607	"display_size": upload.display_size,	×
608	"url": upload.globus_url	×
609	}	×
610	end	×
611	end	×
612
613	def embargo_date_as_json	×
614	if embargo_date.present?	×
615	embargo_datetime = embargo_date.to_datetime	×
616	embargo_date_iso8601 = embargo_datetime.iso8601	×
617	# Apache Solr timestamps require the following format:
618	# 1972-05-20T17:33:18Z
619	# https://solr.apache.org/guide/solr/latest/indexing-guide/date-formatting-math.html
620	embargo_date_iso8601.gsub(/\+.+$/, "Z")	×
621	end	×
622	end	×
623	end	×
624	# rubocop:enable Metrics/ClassLength

pulibrary / pdc_describe / 9dbcf7a4-1c56-4510-9614-74ad5a22cff6

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous