08c0f2d4-d9d2-405e-8877-48785d70651c

Committed 01 Aug 2025 05:24PM UTC coverage: 75.569% (-2.4%) from 77.946%

Build # 08c0f2d4-d9d2-405e-8877-48785d70651c

Build Type

Pull #1670

circleci

Committed by

kelynch

Commit Message

Added rake tasks for user privilege assignments and updated README

Pull Request Pull Request #1670: 1610 remove load users registration

Run Details

2 of 7 new or added lines in 1 file covered. (28.57%)

34 existing lines in 2 files now uncovered.

2224 of 2943 relevant lines covered (75.57%)

312.03 hits per line

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

77.14

/app/models/project.rb

# frozen_string_literal: true
class Project < ApplicationRecord

  class ProjectCreateError < StandardError; end

  validates_with ProjectValidator
  has_many :provenance_events, dependent: :destroy
  before_save do |project|
    # Ensure that the metadata JSONB postgres field is persisted properly
    project.metadata = project.metadata_model
  end

  # Valid project status described in ADR 7
  # See `architecture-decisions/0007-valid-project-statuses.md`
  PENDING_STATUS = "pending"
  APPROVED_STATUS = "approved"
  ACTIVE_STATUS = "active"

  delegate :to_json, to: :metadata_json # field in the database

  def create!(initial_metadata:, user:)
    self.metadata_model = initial_metadata
    if self.valid?
      if initial_metadata.project_id == ProjectMetadata::DOI_NOT_MINTED
        self.draft_doi(user: user)
        self.save!
        ProvenanceEvent.generate_submission_events(project: self, user: user)
      else
        self.save!
      end
      # return doi
      self.metadata_model.project_id
    else
      nil
    end
  end

  def approve!(current_user:)
    request = Mediaflux::ProjectCreateServiceRequest.new(session_token: current_user.mediaflux_session, project: self)
    request.resolve

    if request.mediaflux_id.to_i == 0
      raise ProjectCreateError.new("Error saving project #{self.id} to Mediaflux: #{request.response_error}. Debug output: #{request.debug_output}")
    end

    self.mediaflux_id = request.mediaflux_id
    self.metadata_model.status = Project::APPROVED_STATUS
    self.save!
    debug_output = request.debug_output
    # create provenance events:
    # - one for approving the project and
    # - another for changing the status of the project
    # - another with debug information from the create project service
    ProvenanceEvent.generate_approval_events(project: self, user: current_user, debug_output: debug_output)
  end

  def reload
    super
    @metadata_model = ProjectMetadata.new_from_hash(self.metadata)
    self
  end

  def draft_doi(user: nil)
    puldatacite = PULDatacite.new
    self.metadata_model.project_id = puldatacite.draft_doi
  end

  # Ideally this method should return a ProjectMetadata object (like `metadata_model` does)
  # but we'll keep them both while we are refactoring the code so that we don't break
  # everything at once since `metadata` is used everywhere.
  def metadata
    @metadata_hash = (metadata_json || {}).with_indifferent_access
  end

  def metadata_model
    @metadata_model ||= ProjectMetadata.new_from_hash(self.metadata)
  end

  def metadata_model=(new_metadata_model)
    @metadata_model = new_metadata_model
  end

  def metadata=(metadata_model)
    # Convert our metadata to a hash so it can be saved on our JSONB field
    metadata_hash = JSON.parse(metadata_model.to_json)
    self.metadata_json = metadata_hash
  end

  def title
    self.metadata_model.title
  end

  def departments
    unsorted = metadata_model.departments || []
    unsorted.sort
  end

  def project_directory
    metadata_model.project_directory || ""
  end

  def project_directory_parent_path
    # The tigerdata.project.create expectes every project to be under "tigerdata"
    Mediaflux::Connection.root
  end

  def project_directory_short
    project_directory
  end

  def status
    metadata_model.status
  end

  def pending?
    status == PENDING_STATUS
  end

  def in_mediaflux?
    mediaflux_id.present?
  end

  def self.users_projects(user)
    # See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
    # for information on the @> operator
    uid = user.uid
    query_ro = '{"data_user_read_only":["' + uid + '"]}'
    query_rw = '{"data_user_read_write":["' + uid + '"]}'
    query = "(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb)"
    args = [query_ro, query_rw]
    if user.eligible_sponsor?
      query += "OR (metadata_json->>'data_sponsor' = ?)"
      args << uid
    end
    if user.eligible_manager?
      query += "OR (metadata_json->>'data_manager' = ?)"
      args << uid
    end
    Project.where( query, *args)
  end

  def self.sponsored_projects(sponsor)
    Project.where("metadata_json->>'data_sponsor' = ?", sponsor)
  end

  def self.managed_projects(manager)
    Project.where("metadata_json->>'data_manager' = ?", manager)
  end

  def self.pending_projects
    Project.where("mediaflux_id IS NULL")
  end

  def self.approved_projects
    Project.where("mediaflux_id IS NOT NULL")
  end

  def self.data_user_projects(user)
    # See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
    # for information on the @> operator
    query_ro = '{"data_user_read_only":["' + user + '"]}'
    query_rw = '{"data_user_read_write":["' + user + '"]}'
    Project.where("(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb)", query_ro, query_rw)
  end

  def user_has_access?(user:)
    return true if user.eligible_sysadmin?
    metadata_model.data_sponsor == user.uid || metadata_model.data_manager == user.uid ||
    metadata_model.data_user_read_only.include?(user.uid) || metadata_model.data_user_read_write.include?(user.uid)
  end

  def save_in_mediaflux(user:)
    ProjectMediaflux.save(project: self, user: user)
  end

  def created_by_user
    User.find_by(uid: metadata_model.created_by)
  end

  def to_xml
    ProjectShowPresenter.new(self).to_xml
  end

  # @return [String] XML representation of the <meta> element
  def mediaflux_meta_xml(user:)
    doc = ProjectMediaflux.document(project: self, user: user)
    doc.xpath("/response/reply/result/asset/meta").to_s
  end

  def mediaflux_metadata(session_id:)
    @mediaflux_metadata ||= begin
      accum_req = Mediaflux::AssetMetadataRequest.new(session_token: session_id, id: mediaflux_id)
      accum_req.metadata
    end
    @mediaflux_metadata
  end

  def asset_count(session_id:)
    values = mediaflux_metadata(session_id:)
    values.fetch(:total_file_count, 0)
  end

  def self.default_storage_unit
    "KB"
  end

  def self.default_storage_usage
    "0 #{default_storage_unit}"
  end

  def storage_usage(session_id:)
    values = mediaflux_metadata(session_id:)
    values.fetch(:quota_used, self.class.default_storage_usage) # if the storage is empty use the default
  end

  def storage_usage_raw(session_id:)
    values = mediaflux_metadata(session_id:)
    values.fetch(:quota_used_raw, 0) # if the storage raw is empty use zero
  end

  def self.default_storage_capacity
    "0 GB"
  end

  def storage_capacity(session_id:)
    values = mediaflux_metadata(session_id:)
    quota_value = values.fetch(:quota_allocation, '') #if quota does not exist, set value to an empty string
    if quota_value.blank?
      return self.class.default_storage_capacity
    else
      return quota_value
    end
  end

  def storage_capacity_raw(session_id:)
    values = mediaflux_metadata(session_id:)
    quota_value = values.fetch(:quota_allocation_raw, 0) #if quota does not exist, set value to 0
    quota_value
  end

  # Fetches the first n files
  def file_list(session_id:, size: 10)
    return { files: [] } if mediaflux_id.nil?

    query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")
    iterator_id = query_req.result

    iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: size)
    results = iterator_req.result

    # Destroy _after_ fetching the first set of results from iterator_req.
    # This call is required since it possible that we have read less assets than
    # what the collection has but we are done with the iterator.
    Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve

    results
  end

  # Fetches the entire file list to a file
  def file_list_to_file(session_id:, filename:)
    return { files: [] } if mediaflux_id.nil?

    query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true,  aql_query: "type!='application/arc-asset-collection'")
    iterator_id = query_req.result

    start_time = Time.zone.now
    prefix = "file_list_to_file #{session_id[0..7]} #{self.metadata_model.project_id}"
    log_elapsed(start_time, prefix, "STARTED")

    File.open(filename, "w") do |file|
      page_number = 0
      # file header
      file.write("ID, PATH, NAME, COLLECTION?, LAST_MODIFIED, SIZE\r\n")
      loop do
        iterator_start_time = Time.zone.now
        page_number += 1
        iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: 1000)
        iterator_resp = iterator_req.result
        log_elapsed(iterator_start_time, prefix, "FETCHED page #{page_number} from iterator")
        lines = files_from_iterator(iterator_resp)
        file.write(lines.join("\r\n") + "\r\n")
        break if iterator_resp[:complete] || iterator_req.error?
      end
      log_elapsed(start_time, prefix, "ENDED")
    end

    # Destroy _after_ fetching the results from iterator_req
    # This call is technically not necessary since Mediaflux automatically deletes the iterator
    # once we have ran through it and by now we have. But it does not hurt either.
    Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve
  end


  private

    def files_from_iterator(iterator_resp)
      lines = []
      iterator_resp[:files].each do |asset|
        lines << "#{asset.id}, #{asset.path_only}, #{asset.name}, #{asset.collection}, #{asset.last_modified}, #{asset.size}"
      end
      lines
    end

    def project_directory_pathname
      # allow the directory to be modified by changes in the metadata_json
      @project_directory_pathname = nil if @original_directory.present? && @original_directory != metadata_model.project_directory

      @project_directory_pathname ||= begin
        @original_directory = metadata_model.project_directory
        Pathname.new(@original_directory)
      end
    end

    # Ensure that the project directory is a valid path
    def safe_directory(directory)
      Project.safe_directory(directory)
    end

    def log_elapsed(start_time, prefix, message)
      elapsed_time = Time.zone.now - start_time
      timing_info = "#{format('%.2f', elapsed_time)} s"
      Rails.logger.info "#{prefix}: #{message}, #{timing_info}"
    end
end

1	# frozen_string_literal: true
2	class Project < ApplicationRecord	1✔
3
4	class ProjectCreateError < StandardError; end	1✔
5
6	validates_with ProjectValidator	1✔
7	has_many :provenance_events, dependent: :destroy	1✔
8	before_save do \|project\|	1✔
9	# Ensure that the metadata JSONB postgres field is persisted properly
10	project.metadata = project.metadata_model	331✔
11	end
12
13	# Valid project status described in ADR 7
14	# See `architecture-decisions/0007-valid-project-statuses.md`
15	PENDING_STATUS = "pending"	1✔
16	APPROVED_STATUS = "approved"	1✔
17	ACTIVE_STATUS = "active"	1✔
18
19	delegate :to_json, to: :metadata_json # field in the database	1✔
20
21	def create!(initial_metadata:, user:)	1✔
22	self.metadata_model = initial_metadata	4✔
23	if self.valid?	4✔
24	if initial_metadata.project_id == ProjectMetadata::DOI_NOT_MINTED	3✔
25	self.draft_doi(user: user)	2✔
26	self.save!	2✔
27	ProvenanceEvent.generate_submission_events(project: self, user: user)	2✔
28	else
29	self.save!	1✔
30	end
31	# return doi
32	self.metadata_model.project_id	3✔
33	else
34	nil
35	end
36	end
37
38	def approve!(current_user:)	1✔
39	request = Mediaflux::ProjectCreateServiceRequest.new(session_token: current_user.mediaflux_session, project: self)	1✔
40	request.resolve	1✔
41
42	if request.mediaflux_id.to_i == 0	1✔
43	raise ProjectCreateError.new("Error saving project #{self.id} to Mediaflux: #{request.response_error}. Debug output: #{request.debug_output}")	1✔
44	end
45
46	self.mediaflux_id = request.mediaflux_id	×
47	self.metadata_model.status = Project::APPROVED_STATUS	×
48	self.save!	×
49	debug_output = request.debug_output	×
50	# create provenance events:
51	# - one for approving the project and
52	# - another for changing the status of the project
53	# - another with debug information from the create project service
54	ProvenanceEvent.generate_approval_events(project: self, user: current_user, debug_output: debug_output)	×
55	end
56
57	def reload	1✔
58	super	7✔
59	@metadata_model = ProjectMetadata.new_from_hash(self.metadata)	7✔
60	self	7✔
61	end
62
63	def draft_doi(user: nil)	1✔
64	puldatacite = PULDatacite.new	3✔
65	self.metadata_model.project_id = puldatacite.draft_doi	3✔
66	end
67
68	# Ideally this method should return a ProjectMetadata object (like `metadata_model` does)
69	# but we'll keep them both while we are refactoring the code so that we don't break
70	# everything at once since `metadata` is used everywhere.
71	def metadata	1✔
72	@metadata_hash = (metadata_json \|\| {}).with_indifferent_access	430✔
73	end
74
75	def metadata_model	1✔
76	@metadata_model \|\|= ProjectMetadata.new_from_hash(self.metadata)	3,855✔
77	end
78
79	def metadata_model=(new_metadata_model)	1✔
80	@metadata_model = new_metadata_model	311✔
81	end
82
83	def metadata=(metadata_model)	1✔
84	# Convert our metadata to a hash so it can be saved on our JSONB field
85	metadata_hash = JSON.parse(metadata_model.to_json)	706✔
86	self.metadata_json = metadata_hash	706✔
87	end
88
89	def title	1✔
90	self.metadata_model.title	263✔
91	end
92
93	def departments	1✔
94	unsorted = metadata_model.departments \|\| []	77✔
95	unsorted.sort	77✔
96	end
97
98	def project_directory	1✔
99	metadata_model.project_directory \|\| ""	173✔
100	end
101
102	def project_directory_parent_path	1✔
103	# The tigerdata.project.create expectes every project to be under "tigerdata"
104	Mediaflux::Connection.root	11✔
105	end
106
107	def project_directory_short	1✔
108	project_directory	15✔
109	end
110
111	def status	1✔
112	metadata_model.status	555✔
113	end
114
115	def pending?	1✔
116	status == PENDING_STATUS	269✔
117	end
118
119	def in_mediaflux?	1✔
120	mediaflux_id.present?	61✔
121	end
122
123	def self.users_projects(user)	1✔
124	# See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
125	# for information on the @> operator
126	uid = user.uid	62✔
127	query_ro = '{"data_user_read_only":["' + uid + '"]}'	62✔
128	query_rw = '{"data_user_read_write":["' + uid + '"]}'	62✔
129	query = "(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb)"	62✔
130	args = [query_ro, query_rw]	62✔
131	if user.eligible_sponsor?	62✔
132	query += "OR (metadata_json->>'data_sponsor' = ?)"	16✔
133	args << uid	16✔
134	end
135	if user.eligible_manager?	62✔
136	query += "OR (metadata_json->>'data_manager' = ?)"	12✔
137	args << uid	12✔
138	end
139	Project.where( query, *args)	62✔
140	end
141
142	def self.sponsored_projects(sponsor)	1✔
143	Project.where("metadata_json->>'data_sponsor' = ?", sponsor)	1✔
144	end
145
146	def self.managed_projects(manager)	1✔
147	Project.where("metadata_json->>'data_manager' = ?", manager)	1✔
148	end
149
150	def self.pending_projects	1✔
151	Project.where("mediaflux_id IS NULL")	58✔
152	end
153
154	def self.approved_projects	1✔
155	Project.where("mediaflux_id IS NOT NULL")	58✔
156	end
157
158	def self.data_user_projects(user)	1✔
159	# See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
160	# for information on the @> operator
161	query_ro = '{"data_user_read_only":["' + user + '"]}'	1✔
162	query_rw = '{"data_user_read_write":["' + user + '"]}'	1✔
163	Project.where("(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb)", query_ro, query_rw)	1✔
164	end
165
166	def user_has_access?(user:)	1✔
167	return true if user.eligible_sysadmin?	46✔
168	metadata_model.data_sponsor == user.uid \|\| metadata_model.data_manager == user.uid \|\|	36✔
169	metadata_model.data_user_read_only.include?(user.uid) \|\| metadata_model.data_user_read_write.include?(user.uid)
170	end
171
172	def save_in_mediaflux(user:)	1✔
UNCOV 173	ProjectMediaflux.save(project: self, user: user)	×
174	end
175
176	def created_by_user	1✔
177	User.find_by(uid: metadata_model.created_by)	12✔
178	end
179
180	def to_xml	1✔
181	ProjectShowPresenter.new(self).to_xml	4✔
182	end
183
184	# @return [String] XML representation of the <meta> element
185	def mediaflux_meta_xml(user:)	1✔
186	doc = ProjectMediaflux.document(project: self, user: user)	1✔
187	doc.xpath("/response/reply/result/asset/meta").to_s	1✔
188	end
189
190	def mediaflux_metadata(session_id:)	1✔
191	@mediaflux_metadata \|\|= begin	109✔
192	accum_req = Mediaflux::AssetMetadataRequest.new(session_token: session_id, id: mediaflux_id)	29✔
193	accum_req.metadata	29✔
194	end
195	@mediaflux_metadata	109✔
196	end
197
198	def asset_count(session_id:)	1✔
199	values = mediaflux_metadata(session_id:)	11✔
200	values.fetch(:total_file_count, 0)	11✔
201	end
202
203	def self.default_storage_unit	1✔
204	"KB"	26✔
205	end
206
207	def self.default_storage_usage	1✔
208	"0 #{default_storage_unit}"	25✔
209	end
210
211	def storage_usage(session_id:)	1✔
212	values = mediaflux_metadata(session_id:)	24✔
213	values.fetch(:quota_used, self.class.default_storage_usage) # if the storage is empty use the default	24✔
214	end
215
216	def storage_usage_raw(session_id:)	1✔
UNCOV 217	values = mediaflux_metadata(session_id:)	×
UNCOV 218	values.fetch(:quota_used_raw, 0) # if the storage raw is empty use zero	×
219	end
220
221	def self.default_storage_capacity	1✔
222	"0 GB"	24✔
223	end
224
225	def storage_capacity(session_id:)	1✔
226	values = mediaflux_metadata(session_id:)	24✔
227	quota_value = values.fetch(:quota_allocation, '') #if quota does not exist, set value to an empty string	24✔
228	if quota_value.blank?	24✔
229	return self.class.default_storage_capacity	24✔
230	else
UNCOV 231	return quota_value	×
232	end
233	end
234
235	def storage_capacity_raw(session_id:)	1✔
236	values = mediaflux_metadata(session_id:)	48✔
237	quota_value = values.fetch(:quota_allocation_raw, 0) #if quota does not exist, set value to 0	48✔
238	quota_value	48✔
239	end
240
241	# Fetches the first n files
242	def file_list(session_id:, size: 10)	1✔
243	return { files: [] } if mediaflux_id.nil?	11✔
244
245	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	4✔
246	iterator_id = query_req.result	4✔
247
248	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: size)	4✔
249	results = iterator_req.result	4✔
250
251	# Destroy _after_ fetching the first set of results from iterator_req.
252	# This call is required since it possible that we have read less assets than
253	# what the collection has but we are done with the iterator.
254	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	4✔
255
256	results	4✔
257	end
258
259	# Fetches the entire file list to a file
260	def file_list_to_file(session_id:, filename:)	1✔
261	return { files: [] } if mediaflux_id.nil?	1✔
262
UNCOV 263	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	×
264	iterator_id = query_req.result	×
265
266	start_time = Time.zone.now	×
UNCOV 267	prefix = "file_list_to_file #{session_id[0..7]} #{self.metadata_model.project_id}"	×
268	log_elapsed(start_time, prefix, "STARTED")	×
269
UNCOV 270	File.open(filename, "w") do \|file\|	×
271	page_number = 0	×
272	# file header
273	file.write("ID, PATH, NAME, COLLECTION?, LAST_MODIFIED, SIZE\r\n")	×
274	loop do	×
275	iterator_start_time = Time.zone.now	×
276	page_number += 1	×
277	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: 1000)	×
278	iterator_resp = iterator_req.result	×
279	log_elapsed(iterator_start_time, prefix, "FETCHED page #{page_number} from iterator")	×
280	lines = files_from_iterator(iterator_resp)	×
UNCOV 281	file.write(lines.join("\r\n") + "\r\n")	×
282	break if iterator_resp[:complete] \|\| iterator_req.error?	×
283	end
UNCOV 284	log_elapsed(start_time, prefix, "ENDED")	×
285	end
286
287	# Destroy _after_ fetching the results from iterator_req
288	# This call is technically not necessary since Mediaflux automatically deletes the iterator
289	# once we have ran through it and by now we have. But it does not hurt either.
UNCOV 290	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	×
291	end
292
293
294	private	1✔
295
296	def files_from_iterator(iterator_resp)	1✔
297	lines = []	×
UNCOV 298	iterator_resp[:files].each do \|asset\|	×
299	lines << "#{asset.id}, #{asset.path_only}, #{asset.name}, #{asset.collection}, #{asset.last_modified}, #{asset.size}"	×
300	end
UNCOV 301	lines	×
302	end
303
304	def project_directory_pathname	1✔
305	# allow the directory to be modified by changes in the metadata_json
306	@project_directory_pathname = nil if @original_directory.present? && @original_directory != metadata_model.project_directory	×
307
308	@project_directory_pathname \|\|= begin	×
UNCOV 309	@original_directory = metadata_model.project_directory	×
UNCOV 310	Pathname.new(@original_directory)	×
311	end
312	end
313
314	# Ensure that the project directory is a valid path
315	def safe_directory(directory)	1✔
UNCOV 316	Project.safe_directory(directory)	×
317	end
318
319	def log_elapsed(start_time, prefix, message)	1✔
320	elapsed_time = Time.zone.now - start_time	×
UNCOV 321	timing_info = "#{format('%.2f', elapsed_time)} s"	×
UNCOV 322	Rails.logger.info "#{prefix}: #{message}, #{timing_info}"	×
323	end
324	end

pulibrary / tigerdata-app / 08c0f2d4-d9d2-405e-8877-48785d70651c

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