085a83f3-f8ba-4ace-b5ad-5cb6d3d55a35

Committed 24 Oct 2025 02:47PM UTC coverage: 65.148% (-26.1%) from 91.202%

Build # 085a83f3-f8ba-4ace-b5ad-5cb6d3d55a35

Build Type

Pull #2057

circleci

Committed by

hectorcorrea

Commit Message

Make test more forgiving

Pull Request Pull Request #2057: Dashboard displays projects straights from Mediaflux

Run Details

11 of 87 new or added lines in 4 files covered. (12.64%)

625 existing lines in 35 files now uncovered.

2088 of 3205 relevant lines covered (65.15%)

181.75 hits per line

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

69.18

/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)
    request = Mediaflux::ProjectListRequest.new(session_token: user.mediaflux_session, aql_query: "xpath(tigerdata:project/ProjectID) has value")
    request.results
  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	2✔
3
4	validates_with ProjectValidator	2✔
5	has_many :provenance_events, dependent: :destroy	2✔
6	before_save do \|project\|	2✔
7	# Ensure that the metadata JSONB postgres field is persisted properly
8	project.metadata = project.metadata_model	414✔
9	end
10
11	# Valid project status described in ADR 7
12	# See `architecture-decisions/0007-valid-project-statuses.md`
13	APPROVED_STATUS = "approved"	2✔
14	ACTIVE_STATUS = "active"	2✔
15
16	delegate :to_json, to: :metadata_json # field in the database	2✔
17
18	def create!(initial_metadata:, user:)	2✔
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	def activate(current_user:)	2✔
36	raise StandardError.new("Only approved projects can be activated") if self.status != Project::APPROVED_STATUS	76✔
37	metadata_request = Mediaflux::AssetMetadataRequest.new(session_token: current_user.mediaflux_session, id: self.mediaflux_id)	76✔
38	metadata_request.resolve	76✔
39	raise metadata_request.response_error if metadata_request.error?	76✔
40	if self.title == metadata_request.metadata[:title]	76✔
41	self.metadata_model.status = Project::ACTIVE_STATUS	76✔
42	self.save!	76✔
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)	2✔
49	puldatacite = PULDatacite.new	80✔
50	self.metadata_model.project_id = puldatacite.draft_doi	80✔
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	2✔
57	@metadata_hash = (metadata_json \|\| {}).with_indifferent_access	127✔
58	end
59
60	def metadata_model	2✔
61	@metadata_model \|\|= ProjectMetadata.new_from_hash(self.metadata)	4,702✔
62	end
63
64	def metadata_model=(new_metadata_model)	2✔
65	@metadata_model = new_metadata_model	92✔
66	end
67
68	def metadata=(metadata_model)	2✔
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)	873✔
71	self.metadata_json = metadata_hash	873✔
72	end
73
74	def title	2✔
75	self.metadata_model.title	126✔
76	end
77
78	def departments	2✔
UNCOV 79	unsorted = metadata_model.departments \|\| []	×
UNCOV 80	unsorted.sort	×
81	end
82
83	def project_directory	2✔
84	metadata_model.project_directory \|\| ""	131✔
85	end
86
87	def project_directory_short	2✔
88	project_directory	4✔
89	end
90
91	def status	2✔
92	metadata_model.status	114✔
93	end
94
95	def in_mediaflux?	2✔
UNCOV 96	mediaflux_id.present?	×
97	end
98
99	def self.users_projects(user)	2✔
NEW 100	request = Mediaflux::ProjectListRequest.new(session_token: user.mediaflux_session, aql_query: "xpath(tigerdata:project/ProjectID) has value")	3✔
NEW 101	request.results	3✔
102	end
103
104	def self.sponsored_projects(sponsor)	2✔
105	Project.where("metadata_json->>'data_sponsor' = ?", sponsor)	1✔
106	end
107
108	def self.managed_projects(manager)	2✔
109	Project.where("metadata_json->>'data_manager' = ?", manager)	1✔
110	end
111
112	def self.all_projects	2✔
113	Project.all	×
114	end
115
116	def user_has_access?(user:)	2✔
117	return true if user.eligible_sysadmin?	2✔
118	metadata_model.data_sponsor == user.uid \|\| metadata_model.data_manager == user.uid \|\|	2✔
119	metadata_model.data_user_read_only.include?(user.uid) \|\| metadata_model.data_user_read_write.include?(user.uid)
120	end
121
122	def created_by_user	2✔
123	User.find_by(uid: metadata_model.created_by)	×
124	end
125
126	def to_xml	2✔
UNCOV 127	ProjectShowPresenter.new(self).to_xml	×
128	end
129
130	# @return [String] XML representation of the <meta> element
131	def mediaflux_meta_xml(user:)	2✔
132	doc = ProjectMediaflux.document(project: self, user: user)	2✔
133	doc.xpath("/response/reply/result/asset/meta").to_s	2✔
134	end
135
136	def mediaflux_metadata(session_id:)	2✔
137	@mediaflux_metadata \|\|= begin	8✔
138	accum_req = Mediaflux::AssetMetadataRequest.new(session_token: session_id, id: mediaflux_id)	3✔
139	accum_req.metadata	3✔
140	end
141	@mediaflux_metadata	8✔
142	end
143
144	def asset_count(session_id:)	2✔
145	values = mediaflux_metadata(session_id:)	2✔
146	values.fetch(:total_file_count, 0)	2✔
147	end
148
149	def self.default_storage_unit	2✔
150	"KB"	4✔
151	end
152
153	def self.default_storage_usage	2✔
154	"0 #{default_storage_unit}"	3✔
155	end
156
157	def storage_usage(session_id:)	2✔
158	values = mediaflux_metadata(session_id:)	2✔
159	values.fetch(:quota_used, self.class.default_storage_usage) # if the storage is empty use the default	2✔
160	end
161
162	def storage_usage_raw(session_id:)	2✔
163	values = mediaflux_metadata(session_id:)	×
164	values.fetch(:quota_used_raw, 0) # if the storage raw is empty use zero	×
165	end
166
167	def self.default_storage_capacity	2✔
168	"0 GB"	2✔
169	end
170
171	def storage_capacity(session_id:)	2✔
172	values = mediaflux_metadata(session_id:)	2✔
173	quota_value = values.fetch(:quota_allocation, '') #if quota does not exist, set value to an empty string	2✔
174	if quota_value.blank?	2✔
175	return self.class.default_storage_capacity	2✔
176	else
UNCOV 177	return quota_value	×
178	end
179	end
180
181	def storage_capacity_raw(session_id:)	2✔
UNCOV 182	values = mediaflux_metadata(session_id:)	×
UNCOV 183	quota_value = values.fetch(:quota_allocation_raw, 0) #if quota does not exist, set value to 0	×
UNCOV 184	quota_value	×
185	end
186
187	# Fetches the first n files
188	def file_list(session_id:, size: 10)	2✔
189	return { files: [] } if mediaflux_id.nil?	3✔
190
UNCOV 191	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	1✔
UNCOV 192	iterator_id = query_req.result	1✔
193
UNCOV 194	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: size)	1✔
UNCOV 195	results = iterator_req.result	1✔
196
197	# Destroy _after_ fetching the first set of results from iterator_req.
198	# This call is required since it possible that we have read less assets than
199	# what the collection has but we are done with the iterator.
UNCOV 200	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	1✔
201
UNCOV 202	results	1✔
203	end
204
205	# Fetches the entire file list to a file
206	def file_list_to_file(session_id:, filename:)	2✔
UNCOV 207	return { files: [] } if mediaflux_id.nil?	×
208
UNCOV 209	query_req = Mediaflux::QueryRequest.new(session_token: session_id, collection: mediaflux_id, deep_search: true, aql_query: "type!='application/arc-asset-collection'")	×
UNCOV 210	iterator_id = query_req.result	×
211
UNCOV 212	start_time = Time.zone.now	×
UNCOV 213	prefix = "file_list_to_file #{session_id[0..7]} #{self.metadata_model.project_id}"	×
UNCOV 214	log_elapsed(start_time, prefix, "STARTED")	×
215
UNCOV 216	File.open(filename, "w") do \|file\|	×
UNCOV 217	page_number = 0	×
218	# file header
UNCOV 219	file.write("ID, PATH, NAME, COLLECTION?, LAST_MODIFIED, SIZE\r\n")	×
UNCOV 220	loop do	×
UNCOV 221	iterator_start_time = Time.zone.now	×
UNCOV 222	page_number += 1	×
UNCOV 223	iterator_req = Mediaflux::IteratorRequest.new(session_token: session_id, iterator: iterator_id, size: 1000)	×
UNCOV 224	iterator_resp = iterator_req.result	×
UNCOV 225	log_elapsed(iterator_start_time, prefix, "FETCHED page #{page_number} from iterator")	×
UNCOV 226	lines = files_from_iterator(iterator_resp)	×
UNCOV 227	file.write(lines.join("\r\n") + "\r\n")	×
UNCOV 228	break if iterator_resp[:complete] \|\| iterator_req.error?	×
229	end
UNCOV 230	log_elapsed(start_time, prefix, "ENDED")	×
231	end
232
233	# Destroy _after_ fetching the results from iterator_req
234	# This call is technically not necessary since Mediaflux automatically deletes the iterator
235	# once we have ran through it and by now we have. But it does not hurt either.
UNCOV 236	Mediaflux::IteratorDestroyRequest.new(session_token: session_id, iterator: iterator_id).resolve	×
237	end
238
239
240	private	2✔
241
242	def files_from_iterator(iterator_resp)	2✔
UNCOV 243	lines = []	×
UNCOV 244	iterator_resp[:files].each do \|asset\|	×
245	lines << "#{asset.id}, #{asset.path_only}, #{asset.name}, #{asset.collection}, #{asset.last_modified}, #{asset.size}"	×
246	end
UNCOV 247	lines	×
248	end
249
250	def project_directory_pathname	2✔
251	# allow the directory to be modified by changes in the metadata_json
252	@project_directory_pathname = nil if @original_directory.present? && @original_directory != metadata_model.project_directory	×
253
254	@project_directory_pathname \|\|= begin	×
255	@original_directory = metadata_model.project_directory	×
256	Pathname.new(@original_directory)	×
257	end
258	end
259
260	# Ensure that the project directory is a valid path
261	def safe_directory(directory)	2✔
262	Project.safe_directory(directory)	×
263	end
264
265	def log_elapsed(start_time, prefix, message)	2✔
UNCOV 266	elapsed_time = Time.zone.now - start_time	×
UNCOV 267	timing_info = "#{format('%.2f', elapsed_time)} s"	×
UNCOV 268	Rails.logger.info "#{prefix}: #{message}, #{timing_info}"	×
269	end
270	end

pulibrary / tigerdata-app / 085a83f3-f8ba-4ace-b5ad-5cb6d3d55a35

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