80edfcb6-378a-466d-a97a-75160485a79a

Committed 17 Sep 2025 10:36PM UTC coverage: 88.64% (-0.2%) from 88.85%

Build # 80edfcb6-378a-466d-a97a-75160485a79a

Build Type

Pull #1864

circleci

Committed by

web-flow

Commit Message

Merge branch 'main' into 1707-project-approve

Pull Request Pull Request #1864: Updated tests to use the new request.approve() method

Run Details

2528 of 2852 relevant lines covered (88.64%)

325.43 hits per line

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

92.59

/app/models/project.rb

# frozen_string_literal: true
class Project < ApplicationRecord

  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`
  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

  # TODO: Remove this method https://github.com/pulibrary/tigerdata-app/issues/1707 has been completed
  def approve!(current_user:)
    # This code is duplicated with Request.approve() and it should
    # be removed. We keep it for now since we have way too many tests
    # wired to it already. The goal is that projects won't be approved,
    # instead Request are approved (and that process creates the project)
    create_project_operation = ProjectCreate.new
    result = create_project_operation.call(request: nil, approver: current_user, project: self)
    if result.success?
      self.mediaflux_id
    else
      raise ProjectCreate::ProjectCreateError, result.failure
    end
  end

  def activate(current_user:)
    raise StandardError.new("Only approved projects can be activated") if self.status != Project::APPROVED_STATUS
    metadata_request = Mediaflux::AssetMetadataRequest.new(session_token: current_user.mediaflux_session, id: self.mediaflux_id)
    metadata_request.resolve
    raise metadata_request.response_error if metadata_request.error?
    if self.title == metadata_request.metadata[:title]
      self.metadata_model.status = Project::ACTIVE_STATUS
      self.save!
    else
      raise StandardError.new("Title mismatch: #{title} != #{metadata_request.title}")
    end
  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 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) OR (metadata_json->>'data_sponsor' = ?) OR (metadata_json->>'data_manager' = ?)"
    args = [query_ro, query_rw, uid, uid]
    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.all_projects
    Project.all
  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 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	validates_with ProjectValidator	1✔
5	has_many :provenance_events, dependent: :destroy	1✔
6	before_save do \|project\|	1✔
7	# Ensure that the metadata JSONB postgres field is persisted properly
8	project.metadata = project.metadata_model	643✔
9	end
10
11	# Valid project status described in ADR 7
12	# See `architecture-decisions/0007-valid-project-statuses.md`
13	APPROVED_STATUS = "approved"	1✔
14	ACTIVE_STATUS = "active"	1✔
15
16	delegate :to_json, to: :metadata_json # field in the database	1✔
17
18	def create!(initial_metadata:, user:)	1✔
19	self.metadata_model = initial_metadata	5✔
20	if self.valid?	5✔
21	if initial_metadata.project_id == ProjectMetadata::DOI_NOT_MINTED	3✔
22	self.draft_doi(user: user)	2✔
23	self.save!	2✔
24	ProvenanceEvent.generate_submission_events(project: self, user: user)	2✔
25	else
26	self.save!	1✔
27	end
28	# return doi
29	self.metadata_model.project_id	3✔
30	else
31	nil
32	end
33	end
34
35	# TODO: Remove this method https://github.com/pulibrary/tigerdata-app/issues/1707 has been completed
36	def approve!(current_user:)	1✔
37	# This code is duplicated with Request.approve() and it should
38	# be removed. We keep it for now since we have way too many tests
39	# wired to it already. The goal is that projects won't be approved,
40	# instead Request are approved (and that process creates the project)
41	create_project_operation = ProjectCreate.new	38✔
42	result = create_project_operation.call(request: nil, approver: current_user, project: self)	38✔
43	if result.success?	38✔
44	self.mediaflux_id	38✔
45	else
46	raise ProjectCreate::ProjectCreateError, result.failure	×
47	end
48	end
49
50	def activate(current_user:)	1✔
51	raise StandardError.new("Only approved projects can be activated") if self.status != Project::APPROVED_STATUS	74✔
52	metadata_request = Mediaflux::AssetMetadataRequest.new(session_token: current_user.mediaflux_session, id: self.mediaflux_id)	74✔
53	metadata_request.resolve	74✔
54	raise metadata_request.response_error if metadata_request.error?	74✔
55	if self.title == metadata_request.metadata[:title]	74✔
56	self.metadata_model.status = Project::ACTIVE_STATUS	74✔
57	self.save!	74✔
58	else
59	raise StandardError.new("Title mismatch: #{title} != #{metadata_request.title}")	×
60	end
61	end
62
63	def reload	1✔
64	super	×
65	@metadata_model = ProjectMetadata.new_from_hash(self.metadata)	×
66	self	×
67	end
68
69	def draft_doi(user: nil)	1✔
70	puldatacite = PULDatacite.new	43✔
71	self.metadata_model.project_id = puldatacite.draft_doi	43✔
72	end
73
74	# Ideally this method should return a ProjectMetadata object (like `metadata_model` does)
75	# but we'll keep them both while we are refactoring the code so that we don't break
76	# everything at once since `metadata` is used everywhere.
77	def metadata	1✔
78	@metadata_hash = (metadata_json \|\| {}).with_indifferent_access	466✔
79	end
80
81	def metadata_model	1✔
82	@metadata_model \|\|= ProjectMetadata.new_from_hash(self.metadata)	7,275✔
83	end
84
85	def metadata_model=(new_metadata_model)	1✔
86	@metadata_model = new_metadata_model	339✔
87	end
88
89	def metadata=(metadata_model)	1✔
90	# Convert our metadata to a hash so it can be saved on our JSONB field
91	metadata_hash = JSON.parse(metadata_model.to_json)	1,520✔
92	self.metadata_json = metadata_hash	1,520✔
93	end
94
95	def title	1✔
96	self.metadata_model.title	463✔
97	end
98
99	def departments	1✔
100	unsorted = metadata_model.departments \|\| []	15✔
101	unsorted.sort	15✔
102	end
103
104	def project_directory	1✔
105	metadata_model.project_directory \|\| ""	181✔
106	end
107
108	def project_directory_parent_path	1✔
109	# The tigerdata.project.create expectes every project to be under "tigerdata"
110	Mediaflux::Connection.root	×
111	end
112
113	def project_directory_short	1✔
114	project_directory	10✔
115	end
116
117	def status	1✔
118	metadata_model.status	479✔
119	end
120
121	def in_mediaflux?	1✔
122	mediaflux_id.present?	26✔
123	end
124
125	def self.users_projects(user)	1✔
126	# See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
127	# for information on the @> operator
128	uid = user.uid	67✔
129	query_ro = '{"data_user_read_only":["' + uid + '"]}'	67✔
130	query_rw = '{"data_user_read_write":["' + uid + '"]}'	67✔
131	query = "(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb) OR (metadata_json->>'data_sponsor' = ?) OR (metadata_json->>'data_manager' = ?)"	67✔
132	args = [query_ro, query_rw, uid, uid]	67✔
133	Project.where( query, *args)	67✔
134	end
135
136	def self.sponsored_projects(sponsor)	1✔
137	Project.where("metadata_json->>'data_sponsor' = ?", sponsor)	1✔
138	end
139
140	def self.managed_projects(manager)	1✔
141	Project.where("metadata_json->>'data_manager' = ?", manager)	1✔
142	end
143
144	def self.all_projects	1✔
145	Project.all	4✔
146	end
147
148	def user_has_access?(user:)	1✔
149	return true if user.eligible_sysadmin?	55✔
150	metadata_model.data_sponsor == user.uid \|\| metadata_model.data_manager == user.uid \|\|	47✔
151	metadata_model.data_user_read_only.include?(user.uid) \|\| metadata_model.data_user_read_write.include?(user.uid)
152	end
153
154	def created_by_user	1✔
155	User.find_by(uid: metadata_model.created_by)	×
156	end
157
158	def to_xml	1✔
159	ProjectShowPresenter.new(self).to_xml	1✔
160	end
161
162	# @return [String] XML representation of the <meta> element
163	def mediaflux_meta_xml(user:)	1✔
164	doc = ProjectMediaflux.document(project: self, user: user)	3✔
165	doc.xpath("/response/reply/result/asset/meta").to_s	3✔
166	end
167
168	def mediaflux_metadata(session_id:)	1✔
169	@mediaflux_metadata \|\|= begin	764✔
170	accum_req = Mediaflux::AssetMetadataRequest.new(session_token: session_id, id: mediaflux_id)	138✔
171	accum_req.metadata	138✔
172	end
173	@mediaflux_metadata	760✔
174	end
175
176	def asset_count(session_id:)	1✔
177	values = mediaflux_metadata(session_id:)	30✔
178	values.fetch(:total_file_count, 0)	30✔
179	end
180
181	def self.default_storage_unit	1✔
182	"KB"	248✔
183	end
184
185	def self.default_storage_usage	1✔
186	"0 #{default_storage_unit}"	247✔
187	end
188
189	def storage_usage(session_id:)	1✔
190	values = mediaflux_metadata(session_id:)	250✔
191	values.fetch(:quota_used, self.class.default_storage_usage) # if the storage is empty use the default	246✔
192	end
193
194	def storage_usage_raw(session_id:)	1✔
195	values = mediaflux_metadata(session_id:)	43✔
196	values.fetch(:quota_used_raw, 0) # if the storage raw is empty use zero	43✔
197	end
198
199	def self.default_storage_capacity	1✔
200	"0 GB"	202✔
201	end
202
203	def storage_capacity(session_id:)	1✔
204	values = mediaflux_metadata(session_id:)	246✔
205	quota_value = values.fetch(:quota_allocation, '') #if quota does not exist, set value to an empty string	246✔
206	if quota_value.blank?	246✔
207	return self.class.default_storage_capacity	202✔
208	else
209	return quota_value	44✔
210	end
211	end
212
213	def storage_capacity_raw(session_id:)	1✔
214	values = mediaflux_metadata(session_id:)	193✔
215	quota_value = values.fetch(:quota_allocation_raw, 0) #if quota does not exist, set value to 0	193✔
216	quota_value	193✔
217	end
218
219	# Fetches the first n files
220	def file_list(session_id:, size: 10)	1✔
221	return { files: [] } if mediaflux_id.nil?	33✔
222
223	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	26✔
224	iterator_id = query_req.result	26✔
225
226	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: size)	26✔
227	results = iterator_req.result	26✔
228
229	# Destroy _after_ fetching the first set of results from iterator_req.
230	# This call is required since it possible that we have read less assets than
231	# what the collection has but we are done with the iterator.
232	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	26✔
233
234	results	26✔
235	end
236
237	# Fetches the entire file list to a file
238	def file_list_to_file(session_id:, filename:)	1✔
239	return { files: [] } if mediaflux_id.nil?	11✔
240
241	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	10✔
242	iterator_id = query_req.result	10✔
243
244	start_time = Time.zone.now	10✔
245	prefix = "file_list_to_file #{session_id[0..7]} #{self.metadata_model.project_id}"	10✔
246	log_elapsed(start_time, prefix, "STARTED")	10✔
247
248	File.open(filename, "w") do \|file\|	10✔
249	page_number = 0	10✔
250	# file header
251	file.write("ID, PATH, NAME, COLLECTION?, LAST_MODIFIED, SIZE\r\n")	10✔
252	loop do	10✔
253	iterator_start_time = Time.zone.now	10✔
254	page_number += 1	10✔
255	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: 1000)	10✔
256	iterator_resp = iterator_req.result	10✔
257	log_elapsed(iterator_start_time, prefix, "FETCHED page #{page_number} from iterator")	10✔
258	lines = files_from_iterator(iterator_resp)	10✔
259	file.write(lines.join("\r\n") + "\r\n")	10✔
260	break if iterator_resp[:complete] \|\| iterator_req.error?	10✔
261	end
262	log_elapsed(start_time, prefix, "ENDED")	10✔
263	end
264
265	# Destroy _after_ fetching the results from iterator_req
266	# This call is technically not necessary since Mediaflux automatically deletes the iterator
267	# once we have ran through it and by now we have. But it does not hurt either.
268	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	10✔
269	end
270
271
272	private	1✔
273
274	def files_from_iterator(iterator_resp)	1✔
275	lines = []	10✔
276	iterator_resp[:files].each do \|asset\|	10✔
277	lines << "#{asset.id}, #{asset.path_only}, #{asset.name}, #{asset.collection}, #{asset.last_modified}, #{asset.size}"	16✔
278	end
279	lines	10✔
280	end
281
282	def project_directory_pathname	1✔
283	# allow the directory to be modified by changes in the metadata_json
284	@project_directory_pathname = nil if @original_directory.present? && @original_directory != metadata_model.project_directory	×
285
286	@project_directory_pathname \|\|= begin	×
287	@original_directory = metadata_model.project_directory	×
288	Pathname.new(@original_directory)	×
289	end
290	end
291
292	# Ensure that the project directory is a valid path
293	def safe_directory(directory)	1✔
294	Project.safe_directory(directory)	×
295	end
296
297	def log_elapsed(start_time, prefix, message)	1✔
298	elapsed_time = Time.zone.now - start_time	30✔
299	timing_info = "#{format('%.2f', elapsed_time)} s"	30✔
300	Rails.logger.info "#{prefix}: #{message}, #{timing_info}"	30✔
301	end
302	end

pulibrary / tigerdata-app / 80edfcb6-378a-466d-a97a-75160485a79a

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