bb6442c4-eb43-430c-9683-6ffb5848882a

Committed 21 Oct 2025 08:43PM UTC coverage: 87.325% (+0.4%) from 86.877%

Build # bb6442c4-eb43-430c-9683-6ffb5848882a

Build Type

Pull #2066

circleci

Committed by

bess

Commit Message

Stop mediaflux when we stop everything else

Pull Request Pull Request #2066: Stop mediaflux when we stop everything else

Run Details

2625 of 3006 relevant lines covered (87.33%)

307.22 hits per line

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

89.33

/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

  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: #{self.title} != #{metadata_request.metadata[:title]}")
    end
  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_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	4✔
3
4	validates_with ProjectValidator	4✔
5	has_many :provenance_events, dependent: :destroy	4✔
6	before_save do \|project\|	4✔
7	# Ensure that the metadata JSONB postgres field is persisted properly
8	project.metadata = project.metadata_model	621✔
9	end
10
11	# Valid project status described in ADR 7
12	# See `architecture-decisions/0007-valid-project-statuses.md`
13	APPROVED_STATUS = "approved"	4✔
14	ACTIVE_STATUS = "active"	4✔
15
16	delegate :to_json, to: :metadata_json # field in the database	4✔
17
18	def create!(initial_metadata:, user:)	4✔
19	self.metadata_model = initial_metadata	1✔
20	if self.valid?	1✔
21	if initial_metadata.project_id == ProjectMetadata::DOI_NOT_MINTED	×
22	self.draft_doi(user: user)	×
23	self.save!	×
24	ProvenanceEvent.generate_submission_events(project: self, user: user)	×
25	else
26	self.save!	×
27	end
28	# return doi
29	self.metadata_model.project_id	×
30	else
31	nil
32	end
33	end
34
35	def activate(current_user:)	4✔
36	raise StandardError.new("Only approved projects can be activated") if self.status != Project::APPROVED_STATUS	95✔
37	metadata_request = Mediaflux::AssetMetadataRequest.new(session_token: current_user.mediaflux_session, id: self.mediaflux_id)	95✔
38	metadata_request.resolve	95✔
39	raise metadata_request.response_error if metadata_request.error?	95✔
40	if self.title == metadata_request.metadata[:title]	95✔
41	self.metadata_model.status = Project::ACTIVE_STATUS	95✔
42	self.save!	95✔
43	else
44	raise StandardError.new("Title mismatch: #{self.title} != #{metadata_request.metadata[:title]}")	×
45	end
46	end
47
48	def draft_doi(user: nil)	4✔
49	puldatacite = PULDatacite.new	101✔
50	self.metadata_model.project_id = puldatacite.draft_doi	101✔
51	end
52
53	# Ideally this method should return a ProjectMetadata object (like `metadata_model` does)
54	# but we'll keep them both while we are refactoring the code so that we don't break
55	# everything at once since `metadata` is used everywhere.
56	def metadata	4✔
57	@metadata_hash = (metadata_json \|\| {}).with_indifferent_access	388✔
58	end
59
60	def metadata_model	4✔
61	@metadata_model \|\|= ProjectMetadata.new_from_hash(self.metadata)	7,421✔
62	end
63
64	def metadata_model=(new_metadata_model)	4✔
65	@metadata_model = new_metadata_model	227✔
66	end
67
68	def metadata=(metadata_model)	4✔
69	# Convert our metadata to a hash so it can be saved on our JSONB field
70	metadata_hash = JSON.parse(metadata_model.to_json)	1,300✔
71	self.metadata_json = metadata_hash	1,300✔
72	end
73
74	def title	4✔
75	self.metadata_model.title	496✔
76	end
77
78	def departments	4✔
79	unsorted = metadata_model.departments \|\| []	11✔
80	unsorted.sort	11✔
81	end
82
83	def project_directory	4✔
84	metadata_model.project_directory \|\| ""	159✔
85	end
86
87	def project_directory_short	4✔
88	project_directory	10✔
89	end
90
91	def status	4✔
92	metadata_model.status	489✔
93	end
94
95	def in_mediaflux?	4✔
96	mediaflux_id.present?	20✔
97	end
98
99	def self.users_projects(user)	4✔
100	# See https://scalegrid.io/blog/using-jsonb-in-postgresql-how-to-effectively-store-index-json-data-in-postgresql/
101	# for information on the @> operator
102	uid = user.uid	76✔
103	query_ro = '{"data_user_read_only":["' + uid + '"]}'	76✔
104	query_rw = '{"data_user_read_write":["' + uid + '"]}'	76✔
105	query = "(metadata_json @> ? :: jsonb) OR (metadata_json @> ? :: jsonb) OR (metadata_json->>'data_sponsor' = ?) OR (metadata_json->>'data_manager' = ?)"	76✔
106	args = [query_ro, query_rw, uid, uid]	76✔
107	Project.where( query, *args)	76✔
108	end
109
110	def self.sponsored_projects(sponsor)	4✔
111	Project.where("metadata_json->>'data_sponsor' = ?", sponsor)	×
112	end
113
114	def self.managed_projects(manager)	4✔
115	Project.where("metadata_json->>'data_manager' = ?", manager)	×
116	end
117
118	def self.all_projects	4✔
119	Project.all	3✔
120	end
121
122	def user_has_access?(user:)	4✔
123	return true if user.eligible_sysadmin?	43✔
124	metadata_model.data_sponsor == user.uid \|\| metadata_model.data_manager == user.uid \|\|	35✔
125	metadata_model.data_user_read_only.include?(user.uid) \|\| metadata_model.data_user_read_write.include?(user.uid)
126	end
127
128	def created_by_user	4✔
129	User.find_by(uid: metadata_model.created_by)	×
130	end
131
132	def to_xml	4✔
133	ProjectShowPresenter.new(self).to_xml	×
134	end
135
136	# @return [String] XML representation of the <meta> element
137	def mediaflux_meta_xml(user:)	4✔
138	doc = ProjectMediaflux.document(project: self, user: user)	1✔
139	doc.xpath("/response/reply/result/asset/meta").to_s	1✔
140	end
141
142	def mediaflux_metadata(session_id:)	4✔
143	@mediaflux_metadata \|\|= begin	747✔
144	accum_req = Mediaflux::AssetMetadataRequest.new(session_token: session_id, id: mediaflux_id)	130✔
145	accum_req.metadata	130✔
146	end
147	@mediaflux_metadata	747✔
148	end
149
150	def asset_count(session_id:)	4✔
151	values = mediaflux_metadata(session_id:)	27✔
152	values.fetch(:total_file_count, 0)	27✔
153	end
154
155	def self.default_storage_unit	4✔
156	"KB"	244✔
157	end
158
159	def self.default_storage_usage	4✔
160	"0 #{default_storage_unit}"	244✔
161	end
162
163	def storage_usage(session_id:)	4✔
164	values = mediaflux_metadata(session_id:)	244✔
165	values.fetch(:quota_used, self.class.default_storage_usage) # if the storage is empty use the default	244✔
166	end
167
168	def storage_usage_raw(session_id:)	4✔
169	values = mediaflux_metadata(session_id:)	41✔
170	values.fetch(:quota_used_raw, 0) # if the storage raw is empty use zero	41✔
171	end
172
173	def self.default_storage_capacity	4✔
174	"0 GB"	203✔
175	end
176
177	def storage_capacity(session_id:)	4✔
178	values = mediaflux_metadata(session_id:)	244✔
179	quota_value = values.fetch(:quota_allocation, '') #if quota does not exist, set value to an empty string	244✔
180	if quota_value.blank?	244✔
181	return self.class.default_storage_capacity	203✔
182	else
183	return quota_value	41✔
184	end
185	end
186
187	def storage_capacity_raw(session_id:)	4✔
188	values = mediaflux_metadata(session_id:)	191✔
189	quota_value = values.fetch(:quota_allocation_raw, 0) #if quota does not exist, set value to 0	191✔
190	quota_value	191✔
191	end
192
193	# Fetches the first n files
194	def file_list(session_id:, size: 10)	4✔
195	return { files: [] } if mediaflux_id.nil?	29✔
196
197	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	24✔
198	iterator_id = query_req.result	24✔
199
200	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: size)	24✔
201	results = iterator_req.result	24✔
202
203	# Destroy _after_ fetching the first set of results from iterator_req.
204	# This call is required since it possible that we have read less assets than
205	# what the collection has but we are done with the iterator.
206	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	24✔
207
208	results	24✔
209	end
210
211	# Fetches the entire file list to a file
212	def file_list_to_file(session_id:, filename:)	4✔
213	return { files: [] } if mediaflux_id.nil?	12✔
214
215	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	11✔
216	iterator_id = query_req.result	11✔
217
218	start_time = Time.zone.now	11✔
219	prefix = "file_list_to_file #{session_id[0..7]} #{self.metadata_model.project_id}"	11✔
220	log_elapsed(start_time, prefix, "STARTED")	11✔
221
222	File.open(filename, "w") do \|file\|	11✔
223	page_number = 0	11✔
224	# file header
225	file.write("ID, PATH, NAME, COLLECTION?, LAST_MODIFIED, SIZE\r\n")	11✔
226	loop do	11✔
227	iterator_start_time = Time.zone.now	11✔
228	page_number += 1	11✔
229	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: 1000)	11✔
230	iterator_resp = iterator_req.result	11✔
231	log_elapsed(iterator_start_time, prefix, "FETCHED page #{page_number} from iterator")	11✔
232	lines = files_from_iterator(iterator_resp)	11✔
233	file.write(lines.join("\r\n") + "\r\n")	11✔
234	break if iterator_resp[:complete] \|\| iterator_req.error?	11✔
235	end
236	log_elapsed(start_time, prefix, "ENDED")	11✔
237	end
238
239	# Destroy _after_ fetching the results from iterator_req
240	# This call is technically not necessary since Mediaflux automatically deletes the iterator
241	# once we have ran through it and by now we have. But it does not hurt either.
242	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	11✔
243	end
244
245
246	private	4✔
247
248	def files_from_iterator(iterator_resp)	4✔
249	lines = []	11✔
250	iterator_resp[:files].each do \|asset\|	11✔
251	lines << "#{asset.id}, #{asset.path_only}, #{asset.name}, #{asset.collection}, #{asset.last_modified}, #{asset.size}"	16✔
252	end
253	lines	11✔
254	end
255
256	def project_directory_pathname	4✔
257	# allow the directory to be modified by changes in the metadata_json
258	@project_directory_pathname = nil if @original_directory.present? && @original_directory != metadata_model.project_directory	×
259
260	@project_directory_pathname \|\|= begin	×
261	@original_directory = metadata_model.project_directory	×
262	Pathname.new(@original_directory)	×
263	end
264	end
265
266	# Ensure that the project directory is a valid path
267	def safe_directory(directory)	4✔
268	Project.safe_directory(directory)	×
269	end
270
271	def log_elapsed(start_time, prefix, message)	4✔
272	elapsed_time = Time.zone.now - start_time	33✔
273	timing_info = "#{format('%.2f', elapsed_time)} s"	33✔
274	Rails.logger.info "#{prefix}: #{message}, #{timing_info}"	33✔
275	end
276	end

pulibrary / tigerdata-app / bb6442c4-eb43-430c-9683-6ffb5848882a

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