Class: Project

Inherits:
ApplicationRecord show all
Defined in:
app/models/project.rb

Constant Summary collapse

PENDING_STATUS =

Valid project status described in ADR 7 See ‘architecture-decisions/0007-valid-project-statuses.md`

"pending"
APPROVED_STATUS = 
"approved"
ACTIVE_STATUS = 
"active"

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.approved_projects ⇒ Object 



148
149
150
 
# File 'app/models/project.rb', line 148

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

.data_user_projects(user) ⇒ Object 



152
153
154
155
156
157
158
 
# File 'app/models/project.rb', line 152

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

.default_storage_capacity ⇒ Object 



211
212
213
 
# File 'app/models/project.rb', line 211

def self.default_storage_capacity
  "0 GB"
end

.default_storage_unit ⇒ Object 



193
194
195
 
# File 'app/models/project.rb', line 193

def self.default_storage_unit
  "KB"
end

.default_storage_usage ⇒ Object 



197
198
199
 
# File 'app/models/project.rb', line 197

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

.managed_projects(manager) ⇒ Object 



140
141
142
 
# File 'app/models/project.rb', line 140

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

.pending_projects ⇒ Object 



144
145
146
 
# File 'app/models/project.rb', line 144

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


136
137
138
 
# File 'app/models/project.rb', line 136

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

.users_projects(user) ⇒ Object 



117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
 
# File 'app/models/project.rb', line 117

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

Instance Method Details

#approve!(current_user:) ⇒ Object

TODO: Remove this method github.com/pulibrary/tigerdata-app/issues/1707 has been completed



37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'app/models/project.rb', line 37

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

#asset_count(session_id:) ⇒ Object 



188
189
190
191
 
# File 'app/models/project.rb', line 188

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

#create!(initial_metadata:, user:) ⇒ Object 



19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
 
# File 'app/models/project.rb', line 19

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

#created_by_user ⇒ Object 



166
167
168
 
# File 'app/models/project.rb', line 166

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

#departments ⇒ Object 



87
88
89
90
 
# File 'app/models/project.rb', line 87

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

#draft_doi(user: nil) ⇒ Object 



57
58
59
60
 
# File 'app/models/project.rb', line 57

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

#file_list(session_id:, size: 10) ⇒ Object

Fetches the first n files



232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
 
# File 'app/models/project.rb', line 232

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

#file_list_to_file(session_id:, filename:) ⇒ Object

Fetches the entire file list to a file



250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
 
# File 'app/models/project.rb', line 250

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

#in_mediaflux? ⇒ Boolean

Returns:

  • (Boolean) 


113
114
115
 
# File 'app/models/project.rb', line 113

def in_mediaflux?
  mediaflux_id.present?
end

#mediaflux_meta_xml(user:) ⇒ String

Returns XML representation of the <meta> element.

Returns:

  • (String) —

    XML representation of the <meta> element



175
176
177
178
 
# File 'app/models/project.rb', line 175

def mediaflux_meta_xml(user:)
  doc = ProjectMediaflux.document(project: self, user: user)
  doc.xpath("/response/reply/result/asset/meta").to_s
end

#mediaflux_metadata(session_id:) ⇒ Object 



180
181
182
183
184
185
186
 
# File 'app/models/project.rb', line 180

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

#metadata ⇒ Object

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.



65
66
67
 
# File 'app/models/project.rb', line 65

def metadata
  @metadata_hash = (metadata_json || {}).with_indifferent_access
end

#metadata=(metadata_model) ⇒ Object 



77
78
79
80
81
 
# File 'app/models/project.rb', line 77

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

#metadata_model ⇒ Object 



69
70
71
 
# File 'app/models/project.rb', line 69

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

#metadata_model=(new_metadata_model) ⇒ Object 



73
74
75
 
# File 'app/models/project.rb', line 73

def metadata_model=(new_metadata_model)
  @metadata_model = new_metadata_model
end

#pending? ⇒ Boolean

Returns:

  • (Boolean) 


109
110
111
 
# File 'app/models/project.rb', line 109

def pending?
  status == PENDING_STATUS
end

#project_directory ⇒ Object 



92
93
94
 
# File 'app/models/project.rb', line 92

def project_directory
  metadata_model.project_directory || ""
end

#project_directory_parent_path ⇒ Object 



96
97
98
99
 
# File 'app/models/project.rb', line 96

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

#project_directory_short ⇒ Object 



101
102
103
 
# File 'app/models/project.rb', line 101

def project_directory_short
  project_directory
end

#reload ⇒ Object 



51
52
53
54
55
 
# File 'app/models/project.rb', line 51

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

#status ⇒ Object 



105
106
107
 
# File 'app/models/project.rb', line 105

def status
  metadata_model.status
end

#storage_capacity(session_id:) ⇒ Object 



215
216
217
218
219
220
221
222
223
 
# File 'app/models/project.rb', line 215

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

#storage_capacity_raw(session_id:) ⇒ Object 



225
226
227
228
229
 
# File 'app/models/project.rb', line 225

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

#storage_usage(session_id:) ⇒ Object 



201
202
203
204
 
# File 'app/models/project.rb', line 201

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

#storage_usage_raw(session_id:) ⇒ Object 



206
207
208
209
 
# File 'app/models/project.rb', line 206

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

#title ⇒ Object 



83
84
85
 
# File 'app/models/project.rb', line 83

def title
  self.metadata_model.title
end

#to_xml ⇒ Object 



170
171
172
 
# File 'app/models/project.rb', line 170

def to_xml
  ProjectShowPresenter.new(self).to_xml
end

#user_has_access?(user:) ⇒ Boolean

Returns:

  • (Boolean) 


160
161
162
163
164
 
# File 'app/models/project.rb', line 160

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