pushmi_pullyu 1.0.6 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 20bc0ae42a4b328138102a38407b45cb98d5b40f
-  data.tar.gz: fbf4b359d8a417a3346b40a6c96ab24fb9c9144d
+SHA256:
+  metadata.gz: 8596ed87a287d7708b92c6a3538dcff1f976ddd1b2dc0171b30011bcfe62276b
+  data.tar.gz: 9babcb9576246ee00e581056361731b324e27d02c49814af9b26346fbd269120
 SHA512:
-  metadata.gz: a1dc673525ae368200ee27bed31fa356378715fc00b65ee65026fd06b9875b072f32901f799e52819ad056f2f632c19186e01a552d59e3cf38ea2f23d693c45f
-  data.tar.gz: 5ed4ed3d6ff59bc804cd780b95908757aaafe44f18046e463c30a14cf4f15ce5848e78ff9ef8777c691bae24e2fc303b479d9d7a1566a589fe6226bf123d8183
+  metadata.gz: 0d031eef30ea6828c6fac6c094ab040e914fce8094298249f56ebf2b44d8820c22fd3de6ff98abc546e482f278351b6838419d9b09a90cdd9be6b50576896afc
+  data.tar.gz: 858a1d2f2a91a7ee9f9c0257391aa601316afa8fb07c09e883b817ec7b90879f5fdeda9eca34b375aeea5c1ed2c9b06b13c1212a03e30c538d2942a2e4976dca

data/.github/CODEOWNERS

@@ -0,0 +1,2 @@
+# https://help.github.com/articles/about-codeowners/
+* @ConnorSheremeta @lagoan @mbarnett @murny @pgwillia @piyapongch

data/.github/PULL_REQUEST_TEMPLATE

@@ -0,0 +1,11 @@
+## Context
+
+Please describe any relevant motivation and context for these changes.
+
+Related to # (issue)
+
+## What's New
+
+Please summarize what's new or changed in this PR.
+
+Remember to request a reviewer when you submit!

data/.github/workflows/ruby.yml

@@ -0,0 +1,31 @@
+name: CI
+
+on:
+  push:
+    branches: [ master ]
+  pull_request:
+    branches: [ master ]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    services:
+      redis:
+        image: redis:alpine
+        ports: ["6379:6379"]
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Ruby
+      uses: ruby/setup-ruby@v1
+      with:
+        ruby-version: 2.6.6
+    - name: Install dependencies
+      run: bundle install
+    - name: Lint with RuboCop
+      run: bundle exec rubocop --parallel
+    - name: Run Danger
+      env:
+        DANGER_GITHUB_API_TOKEN: ${{ secrets.DANGER_GITHUB_API_TOKEN }}
+      run: bundle exec danger
+    - name: Run tests
+      run: bundle exec rake spec

data/.rubocop.yml

@@ -10,8 +10,9 @@ AllCops:
   Exclude:
     - 'tmp/**/*'
     - 'vendor/**/*'
+    - 'Dangerfile'
   ExtraDetails: true
-  TargetRubyVersion: 2.3
+  TargetRubyVersion: 2.5
 
 # readability is Actually Good
 Layout/EmptyLinesAroundClassBody:
@@ -59,6 +60,15 @@ Style/ClassAndModuleChildren:
 Style/Documentation:
   Enabled: false
 
+Style/HashEachMethods:
+  Enabled: true
+
+Style/HashTransformKeys:
+  Enabled: true
+
+Style/HashTransformValues:
+  Enabled: true
+
 Naming/FileName:
   Exclude:
     - Dangerfile
@@ -88,3 +98,12 @@ RSpec/MultipleExpectations:
 
 RSpec/DescribedClass:
   EnforcedStyle: explicit
+
+Lint/RaiseException:
+  Enabled: true
+
+Lint/StructNewOverride:
+  Enabled: true
+
+RSpec/MultipleMemoizedHelpers:
+  Enabled: false

data/.travis.yml

@@ -9,8 +9,8 @@ before_install:
   - gem update bundler
 
 rvm:
-  - 2.3.3
-  - 2.4.0
+  - 2.5.0
+  - 2.6.0
 
 before_script:
   - bundle exec danger

data/CHANGELOG.md

@@ -0,0 +1,16 @@
+# Changelog
+All notable changes to PushmiPullyu project will be documented in this file. 
+
+PushmiPullyu is a Ruby application, whose primary job is to manage the flow of content from [Jupiter](https://github.com/ualbertalib/jupiter/) into Swift for preservation.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and releases in PushmiPullyu adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [2.0.0] - 2020-12-14
+
+### Removed
+- Data output for original_file information
+
+## [1.0.6] - 2018-11-29

data/Dangerfile

@@ -13,9 +13,13 @@ end
 # just leaving a title
 warn('Please add a detailed summary in the description.') if github.pr_body.length < 5
 
-# Ensure a clean commits history
-if git.commits.any? { |c| c.message =~ /^Merge branch '#{github.branch_for_base}'/ }
-  warn 'Please rebase to get rid of the merge commits in this PR'
+# Let people say that this isn't worth a CHANGELOG entry in the PR if they choose
+declared_trivial = (github.pr_title + github.pr_body).include?('#trivial') || !has_app_changes
+
+if !git.modified_files.include?('CHANGELOG.md') && !declared_trivial
+  error_message = "Please include a CHANGELOG entry. \nYou can find it at " \
+                  '[CHANGELOG.md](https://github.com/ualbertalib/pushmi_pullyu/blob/master/CHANGELOG.md).'
+  fail(error_message, sticky: false)
 end
 
 # Warn when there is a big PR

data/README.md

@@ -5,22 +5,21 @@
 </p>
 
 [![Gem Version](https://badge.fury.io/rb/pushmi_pullyu.svg)](https://rubygems.org/gems/pushmi_pullyu)
-[![Dependency Status](https://gemnasium.com/badges/github.com/ualbertalib/pushmi_pullyu.svg)](https://gemnasium.com/github.com/ualbertalib/pushmi_pullyu)
-[![Build Status](https://travis-ci.org/ualbertalib/pushmi_pullyu.svg?branch=master)](https://travis-ci.org/ualbertalib/pushmi_pullyu)
+[![Github Build Status](https://github.com/ualbertalib/pushmi_pullyu/workflows/CI/badge.svg)](https://github.com/ualbertalib/pushmi_pullyu/actions)
 [![Coverage Status](https://coveralls.io/repos/github/ualbertalib/pushmi_pullyu/badge.svg?branch=master)](https://coveralls.io/github/ualbertalib/pushmi_pullyu?branch=master)
 
 PushmiPullyu is a Ruby application, running behind the firewall that protects our Swift environment.
 
-Its primary job is to manage the flow of content from Fedora into Swift for preservation.
+Its primary job is to manage the flow of content from Jupiter into Swift for preservation.
 
 ![System Infrastructure Diagram](docs/images/system-infrastructure-diagram.png)
 
 ## Workflow
 
-1.  Any save (create or update) on a Item/Thesis in ERA/Jupiter will trigger an after save callback that will push the item's unique identifier (UUID or NOID) into a Queue.
+1.  Any save (create or update) on a Item/Thesis in ERA/Jupiter will trigger an after save callback that will push the item's unique identifier (UUID) into a Queue.
 2. The queue (Redis) is setup to be a unique set (which only allows one item's UUID to be included in the queue at a single time), and ordered by priority from First In, First out (FIFO).
 3. PushmiPullyu will then monitor the queue. After a certain wait period has passed since an element has been on the queue, PushmiPullyu will then retrieve the elements off the queue and begin to process the preservation event.
-4. All the GenericFile information and data required for preservation are retrieved from Fedora using multiple REST calls. A database connection to the user database fetches (via ActiveRecord )owner emails and modifies the fetched documents, where applicable.
+4. All the Item/Thesis information and data required for preservation are retrieved from Jupiter using multiple REST calls to Jupiter's AIP API.
 5. An Archival Information Package (AIP) is created from the item's information. It is then bagged and tarred.
 6. The AIP tar is then uploaded to Swift via a REST call.
 7. On a successful Swift upload, a entry is added for this preservation event to the preservation event logs.

data/examples/pushmi_pullyu.yml

@@ -21,12 +21,6 @@ minimum_age: 0
 redis:
   url: redis://localhost:6379
 
-fedora:
-  url: http://localhost:8080/fcrepo/rest
-  user: fedoraAdmin
-  password: fedoraAdmin
-  base_path: /dev
-
 database:
   url: postgresql://jupiter:mysecretpassword@127.0.0.1/jupiter_development
 
@@ -44,3 +38,9 @@ rollbar:
   token: 'abc123xyz'
   proxy_host: 'your_proxy_host_url'
   proxy_port: '80'
+
+jupiter:
+  user: jupiter@ualberta.ca
+  api_key: 5042c4ad-6d22-486d-bc63-2b9e5b9a630a
+  jupiter_url: http://localhost:3000/
+  aip_api_path: aip/v1

data/lib/pushmi_pullyu.rb

@@ -1,22 +1,15 @@
-# require 'pushmi_pullyu/version' must be first as it declares the PushmiPullyu module.
-# (This fixes a weird NameError bug when using the nested compact syntax for
-# defining modules/classes like `module PushmiPullyu::Logging`)
-require 'pushmi_pullyu/version'
+# require 'pushmi_pullyu/version' must be first as it declares the PushmiPullyu
+# module. (This fixes a weird NameError bug when using the nested compact syntax
+# for defining modules/classes like `module PushmiPullyu::Logging`)
 
+require 'pushmi_pullyu/version'
 require 'pushmi_pullyu/logging'
-
 require 'pushmi_pullyu/aip'
 require 'pushmi_pullyu/aip/creator'
 require 'pushmi_pullyu/aip/downloader'
-require 'pushmi_pullyu/aip/fedora_fetcher'
-require 'pushmi_pullyu/aip/file_list_creator'
-require 'pushmi_pullyu/aip/owner_email_editor'
-require 'active_record'
-require 'pushmi_pullyu/aip/user'
 require 'pushmi_pullyu/cli'
 require 'pushmi_pullyu/preservation_queue'
 require 'pushmi_pullyu/swift_depositer'
-
 require 'active_support'
 require 'active_support/core_ext'
 
@@ -36,12 +29,6 @@ module PushmiPullyu
     redis: {
       url: 'redis://localhost:6379'
     },
-    fedora: {
-      url: 'http://localhost:8080/fcrepo/rest',
-      user: 'fedoraAdmin',
-      password: 'fedoraAdmin',
-      base_path: '/dev'
-    },
     swift: {
       tenant: 'tester',
       username: 'test:tester',
@@ -58,6 +45,12 @@ module PushmiPullyu
       pool: ENV['RAILS_MAX_THREADS'] || 5,
       url: ENV['DATABASE_URL'] || ENV['JUPITER_DATABASE_URL'] || 'postgresql://jupiter:mysecretpassword@127.0.0.1',
       database: 'jupiter_development'
+    },
+    jupiter: {
+      user: ENV['JUPITER_USER'],
+      api_key: ENV['JUPITER_API_KEY'],
+      jupiter_url: ENV['JUPITER_URL'] || 'http://localhost:3000/',
+      aip_api_path: ENV['JUPITER_AIP_API_PATH'] || 'aip/v1'
     }
   }.freeze
 

data/lib/pushmi_pullyu/aip.rb

@@ -1,17 +1,20 @@
 require 'fileutils'
+require 'uuid'
 
 module PushmiPullyu::AIP
-  class NoidInvalid < StandardError; end
+  class EntityInvalid < StandardError; end
   module_function
 
-  def create(noid)
-    raise NoidInvalid if noid.blank? || noid.include?('/')
+  def create(entity)
+    raise EntityInvalid if entity.nil? ||
+                           UUID.validate(entity[:uuid]) != true ||
+                           entity[:type].blank?
 
-    aip_directory = "#{PushmiPullyu.options[:workdir]}/#{noid}"
+    aip_directory = "#{PushmiPullyu.options[:workdir]}/#{entity[:uuid]}"
     aip_filename = "#{aip_directory}.tar"
 
-    PushmiPullyu::AIP::Downloader.new(noid, aip_directory).run
-    PushmiPullyu::AIP::Creator.new(noid, aip_directory, aip_filename).run
+    PushmiPullyu::AIP::Downloader.new(entity, aip_directory).run
+    PushmiPullyu::AIP::Creator.new(entity[:uuid], aip_directory, aip_filename).run
 
     yield aip_filename, aip_directory
 

data/lib/pushmi_pullyu/aip/downloader.rb

@@ -2,261 +2,213 @@ require 'fileutils'
 require 'ostruct'
 require 'rdf'
 require 'rdf/n3'
+require 'net/http'
+require 'uri'
+require 'digest'
 
-# Download all of the metadata/datastreams and associated data
-# related to an object
+# Download all of the metadata/datastreams and associated data related to an object
 class PushmiPullyu::AIP::Downloader
 
   PREDICATE_URIS = {
     filename: 'http://purl.org/dc/terms/title',
     member_files: 'http://pcdm.org/models#hasFile',
     member_file_sets: 'http://pcdm.org/models#hasMember',
-    original_file: 'http://pcdm.org/use#OriginalFile',
     type: 'http://www.w3.org/1999/02/22-rdf-syntax-ns#type'
   }.freeze
 
-  class NoFileSets < StandardError; end
-  class NoMemberFiles < StandardError; end
-  class NoContentFilename < StandardError; end
-  class NoOriginalFile < StandardError; end
+  class JupiterDownloadError < StandardError; end
+  class JupiterCopyError < StandardError; end
+  class JupiterAuthenticationError < StandardError; end
 
-  def initialize(noid, aip_directory)
-    @noid = noid
+  def initialize(entity, aip_directory)
+    @entity = entity
+    @entity_identifier = "[#{entity[:type]} - #{entity[:uuid]}]".freeze
     @aip_directory = aip_directory
   end
 
   def run
-    make_directories
+    PushmiPullyu.logger.info("#{@entity_identifier}: Retreiving data from Jupiter ...")
 
-    PushmiPullyu.logger.info("#{@noid}: Retreiving data from Fedora ...")
+    authenticate_http_calls
+    make_directories
 
     # Main object metadata
-    object_downloader = PushmiPullyu::AIP::FedoraFetcher.new(@noid)
-    download_and_log(object_aip_paths[:main_object], object_downloader)
-
-    # Construct the file ordering file
-    list_source_uri = object_downloader.object_url + object_aip_paths.list_source.remote
-    create_and_log_file_order_list(list_source_uri)
-
-    member_file_set_uuids.each do |file_set_uuid|
-      make_file_set_directories(file_set_uuid)
-
-      # FileSet metadata
-      file_set_downloader = PushmiPullyu::AIP::FedoraFetcher.new(file_set_uuid)
-      path_spec = file_set_aip_paths(file_set_uuid)[:main_object]
-      download_and_log(path_spec, file_set_downloader)
-
-      # Find the original file by looping through the files in the file_set
-      original_file_remote_base = nil
-      member_files(file_set_uuid).each do |file_path|
-        path_spec = OpenStruct.new(
-          remote: "/files/#{file_path}/fcr:metadata",
-          # Note: local file gets clobbered on each download until it finds the right one
-          local: "#{file_set_dirs(file_set_uuid).metadata}/original_file_metadata.n3",
-          optional: true
-        )
-        download_and_log(path_spec, file_set_downloader)
-        if original_file?(path_spec.local)
-          original_file_remote_base = "/files/#{file_path}"
-          break
-        end
-      end
-
-      raise NoOriginalFile unless original_file_remote_base.present?
-
-      [:content, :fixity].each do |item|
-        path_spec = file_aip_paths(file_set_uuid, original_file_remote_base)[item]
-        download_and_log(path_spec, file_set_downloader)
-      end
+    download_and_log(object_aip_paths[:main_object_remote],
+                     object_aip_paths[:main_object_local])
+    download_and_log(object_aip_paths[:file_sets_remote],
+                     object_aip_paths[:file_sets_local])
+
+    # Get file paths for processing
+    file_paths = get_file_paths(object_aip_paths[:file_paths_remote])
+
+    file_paths[:files].each do |file_path|
+      file_uuid = file_path[:file_uuid]
+      make_file_set_directories(file_uuid)
+      copy_and_log(file_uuid, file_path)
+      file_aip_path = file_aip_paths(file_uuid)
+      download_and_log(file_aip_path[:fixity_remote],
+                       file_aip_path[:fixity_local])
+      download_and_log(file_aip_path[:file_set_remote],
+                       file_aip_path[:file_set_local])
     end
   end
 
   private
 
-  def download_and_log(path_spec, fedora_fetcher)
-    output_file = path_spec.local
-
-    log_fetching(fedora_fetcher.object_url(path_spec.remote), output_file)
+  def copy_and_log(file_uuid, file_path)
+    remote = file_path[:file_path]
+    remote_checksum = file_path[:file_checksum]
+    files_path = file_set_dirs(file_uuid)[:files]
+    output_file = "#{files_path}/#{file_path[:file_name]}"
+    log_downloading(remote, output_file)
+    FileUtils.copy_file(remote, output_file)
 
-    is_rdf = (output_file =~ /\.n3$/)
-    should_add_user_email = path_spec.to_h.fetch(:should_add_user_email, false)
+    is_success = File.exist?(output_file) &&
+                 File.size(remote) == File.size(output_file) &&
+                 compare_md5(output_file, remote_checksum)
 
-    is_success = fedora_fetcher.download_object(output_file,
-                                                url_extra: path_spec.remote,
-                                                optional: path_spec.optional,
-                                                is_rdf: is_rdf,
-                                                should_add_user_email: should_add_user_email)
     log_saved(is_success, output_file)
+
+    raise JupiterCopyError unless is_success
+  end
+
+  def compare_md5(local, remote_checksum)
+    local_md5 = Digest::MD5.file local
+    local_md5.base64digest == remote_checksum
+  end
+
+  def authenticate_http_calls
+    @uri = URI.parse(PushmiPullyu.options[:jupiter][:jupiter_url])
+    @http = Net::HTTP.new(@uri.host, @uri.port)
+    request = Net::HTTP::Post.new(@uri.request_uri + 'auth/system')
+    request.set_form_data(
+      email: PushmiPullyu.options[:jupiter][:user],
+      api_key: PushmiPullyu.options[:jupiter][:api_key]
+    )
+    response = @http.request(request)
+    # If we cannot find the set-cookie header then the session was not set
+    raise JupiterAuthenticationError if response.response['set-cookie'].nil?
+
+    @cookies = response.response['set-cookie']
+  end
+
+  def download_and_log(remote, local)
+    log_downloading(remote, local)
+
+    @uri = URI.parse(PushmiPullyu.options[:jupiter][:jupiter_url])
+    request = Net::HTTP::Get.new(@uri.request_uri + remote)
+    # add previously stored cookies
+    request['Cookie'] = @cookies
+
+    response = @http.request(request)
+    is_success = if response.is_a?(Net::HTTPSuccess)
+                   File.open(local, 'wb') do |file|
+                     file.write(response.body)
+                   end
+                   # Response was a success and the file was saved to local
+                   File.exist? local
+                 end
+
+    log_saved(is_success, local)
+    raise JupiterDownloadError unless is_success
+  end
+
+  def get_file_paths(url)
+    request = Net::HTTP::Get.new(@uri.request_uri + url)
+    # add previously stored cookies
+    request['Cookie'] = @cookies
+
+    response = @http.request(request)
+
+    JSON.parse(response.body, symbolize_names: true)
   end
 
-  def create_and_log_file_order_list(url)
-    output_file = object_aip_paths.file_ordering.local
-    PushmiPullyu::Logging.log_aip_activity(@aip_directory,
-                                           "#{@noid}: #{output_file} -- creating from #{url} ...")
-    PushmiPullyu::AIP::FileListCreator.new(url, output_file, member_file_set_uuids).run
-    PushmiPullyu::Logging.log_aip_activity(@aip_directory,
-                                           "#{@noid}: #{output_file} -- created")
+  def object_uri
+    aip_api_url = PushmiPullyu.options[:jupiter][:aip_api_path]
+    @object_uri ||= "#{aip_api_url}/#{@entity[:type]}/#{@entity[:uuid]}"
   end
 
   ### Logging
 
-  def log_fetching(url, output_file)
-    message = "#{@noid}: #{output_file} -- fetching from #{url} ..."
+  def log_downloading(url, output_file)
+    message = "#{@entity_identifier}: #{output_file} -- Downloading from #{url} ..."
     PushmiPullyu::Logging.log_aip_activity(@aip_directory, message)
   end
 
   def log_saved(is_success, output_file)
-    message = "#{@noid}: #{output_file} -- #{is_success ? 'saved' : 'not_found'}"
+    message = "#{@entity_identifier}: #{output_file} -- #{is_success ? 'Saved' : 'Failed'}"
     PushmiPullyu::Logging.log_aip_activity(@aip_directory, message)
   end
 
   ### Directories
 
   def aip_dirs
-    @aip_dirs ||= OpenStruct.new(
+    @aip_dirs ||= {
       objects: "#{@aip_directory}/data/objects",
       metadata: "#{@aip_directory}/data/objects/metadata",
       files: "#{@aip_directory}/data/objects/files",
       files_metadata: "#{@aip_directory}/data/objects/metadata/files_metadata",
       logs: "#{@aip_directory}/data/logs",
       file_logs: "#{@aip_directory}/data/logs/files_logs"
-    )
+    }
   end
 
   def file_set_dirs(file_set_uuid)
     @file_set_dirs ||= {}
-    @file_set_dirs[file_set_uuid] ||= OpenStruct.new(
-      metadata: "#{aip_dirs.files_metadata}/#{file_set_uuid}",
-      files: "#{aip_dirs.files}/#{file_set_uuid}",
-      logs: "#{aip_dirs.file_logs}/#{file_set_uuid}"
-    )
+    @file_set_dirs[file_set_uuid] ||= {
+      metadata: "#{aip_dirs[:files_metadata]}/#{file_set_uuid}",
+      files: "#{aip_dirs[:files]}/#{file_set_uuid}",
+      logs: "#{aip_dirs[:file_logs]}/#{file_set_uuid}"
+    }
   end
 
   def make_directories
+    PushmiPullyu.logger.debug("#{@entity_identifier}: Creating directories ...")
     clean_directories
-    PushmiPullyu.logger.debug("#{@noid}: Creating directories ...")
-    aip_dirs.to_h.each_value do |path|
+    aip_dirs.each_value do |path|
       FileUtils.mkdir_p(path)
     end
-    PushmiPullyu.logger.debug("#{@noid}: Creating directories done")
+    PushmiPullyu.logger.debug("#{@entity_identifier}: Creating directories done")
   end
 
   def make_file_set_directories(file_set_uuid)
-    PushmiPullyu.logger.debug("#{@noid}: Creating file set #{file_set_uuid} directories ...")
-    file_set_dirs(file_set_uuid).to_h.each_value do |path|
+    PushmiPullyu.logger.debug("#{@entity_identifier}: Creating file set #{file_set_uuid} directories ...")
+    file_set_dirs(file_set_uuid).each_value do |path|
       FileUtils.mkdir_p(path)
     end
-    PushmiPullyu.logger.debug("#{@noid}: Creating file set #{file_set_uuid} directories done")
+    PushmiPullyu.logger.debug("#{@entity_identifier}: Creating file set #{file_set_uuid} directories done")
   end
 
   def clean_directories
     return unless File.exist?(@aip_directory)
 
-    PushmiPullyu.logger.debug("#{@noid}: Nuking directories ...")
+    PushmiPullyu.logger.debug("#{@entity_identifier}: Nuking directories ...")
     FileUtils.rm_rf(@aip_directory)
   end
 
   ### Files
 
   def object_aip_paths
-    @object_aip_paths ||= OpenStruct.new(
-      main_object: OpenStruct.new(
-        remote: nil, # Base path
-        local: "#{aip_dirs.metadata}/object_metadata.n3",
-        should_add_user_email: true,
-        optional: false
-      ),
-      list_source: OpenStruct.new(
-        # This is downloaded, but not saved
-        remote: '/list_source'
-      ),
-      # This is constructed, not downloaded
-      file_ordering: OpenStruct.new(
-        local: "#{aip_dirs.files_metadata}/file_order.xml"
-      )
-    ).freeze
-  end
-
-  def file_set_aip_paths(file_set_uuid)
-    @file_set_aip_paths ||= {}
-    @file_set_aip_paths[file_set_uuid] ||= OpenStruct.new(
-      main_object: OpenStruct.new(
-        remote: nil, # Base file_set path
-        local: "#{file_set_dirs(file_set_uuid).metadata}/file_set_metadata.n3",
-        should_add_user_email: true,
-        optional: false
-      )
-    ).freeze
-  end
-
-  def file_aip_paths(file_set_uuid, original_file_remote_base)
+    @object_aip_paths ||= {
+      # Base path
+      main_object_remote: object_uri,
+      main_object_local: "#{aip_dirs[:metadata]}/object_metadata.n3",
+      file_sets_remote: "#{object_uri}/filesets",
+      file_sets_local: "#{aip_dirs[:files_metadata]}/file_order.xml",
+      # This is downloaded for processing but not saved
+      file_paths_remote: "#{object_uri}/file_paths"
+    }.freeze
+  end
+
+  def file_aip_paths(file_set_uuid)
+    file_set_paths = file_set_dirs(file_set_uuid)
     @file_aip_paths ||= {}
-    @file_aip_paths[file_set_uuid] ||= OpenStruct.new(
-      content: OpenStruct.new(
-        remote: original_file_remote_base,
-        local: file_set_filename(file_set_uuid),
-        optional: false
-      ),
-      fixity: OpenStruct.new(
-        remote: "#{original_file_remote_base}/fcr:fixity",
-        local: "#{file_set_dirs(file_set_uuid)[:logs]}/content_fixity_report.n3",
-        optional: false
-      )
-    ).freeze
-  end
-
-  def member_file_set_uuids
-    @member_file_set_uuids ||= []
-    return @member_file_set_uuids unless @member_file_set_uuids.empty?
-
-    member_file_set_predicate = RDF::URI(PREDICATE_URIS[:member_file_sets])
-
-    graph = RDF::Graph.load(object_aip_paths.main_object.local)
-
-    graph.query(predicate: member_file_set_predicate) do |results|
-      # Get uuid from end of fedora path
-      @member_file_set_uuids << results.object.to_s.split('/').last
-    end
-    return @member_file_set_uuids unless @member_file_set_uuids.empty?
-
-    raise NoFileSets
-  end
-
-  def file_set_filename(file_set_uuid)
-    filename_predicate = RDF::URI(PREDICATE_URIS[:filename])
-
-    graph = RDF::Graph.load(file_set_aip_paths(file_set_uuid).main_object.local)
-
-    graph.query(predicate: filename_predicate) do |results|
-      return "#{file_set_dirs(file_set_uuid).files}/#{results.object}"
-    end
-
-    raise NoContentFilename
-  end
-
-  def member_files(file_set_uuid)
-    member_file_predicate = RDF::URI(PREDICATE_URIS[:member_files])
-
-    graph = RDF::Graph.load(file_set_aip_paths(file_set_uuid).main_object.local)
-
-    member_files = []
-    graph.query(predicate: member_file_predicate) do |results|
-      # Get uuid from end of fedora path
-      member_files << results.object.to_s.split('/').last
-    end
-    return member_files if member_files.present?
-
-    raise NoMemberFiles
-  end
-
-  def original_file?(metadata_filename)
-    type_predicate = RDF::URI(PREDICATE_URIS[:type])
-    original_file_uri = RDF::URI(PREDICATE_URIS[:original_file])
-    graph = RDF::Graph.load(metadata_filename)
-    graph.query(predicate: type_predicate) do |results|
-      return true if results.object == original_file_uri
-    end
-    false
+    @file_aip_paths[file_set_uuid] ||= {
+      fixity_remote: "#{object_uri}/filesets/#{file_set_uuid}/fixity",
+      fixity_local: "#{file_set_paths[:logs]}/content_fixity_report.n3",
+      file_set_remote: "#{object_uri}/filesets/#{file_set_uuid}",
+      file_set_local: "#{file_set_paths[:metadata]}/file_set_metadata.n3"
+    }.freeze
   end
 
 end

data/lib/pushmi_pullyu/cli.rb

@@ -4,6 +4,7 @@ require 'optparse'
 require 'rollbar'
 require 'singleton'
 require 'yaml'
+require 'json'
 
 # CLI runner
 class PushmiPullyu::CLI
@@ -64,8 +65,8 @@ class PushmiPullyu::CLI
       # add a filter after Rollbar has built the error payload but before it is delivered to the API,
       # in order to strip sensitive information out of certain error messages
       exception_message_transformer = proc do |payload|
-        clean_message = payload[:exception][:message].sub(/http:\/\/.+:.+@(.+)\/fedora\/rest\/prod\/(.*)/,
-                                                          "http://\1/fedora/rest/prod/\2")
+        clean_message = payload[:exception][:message].sub(/http:\/\/.+:.+@(.+)\/aip\/v1\/(.*)/,
+                                                          "http://\1/aip/v1/\2")
         payload[:exception][:message] = clean_message
         payload[:message] = clean_message
       end
@@ -181,26 +182,29 @@ class PushmiPullyu::CLI
   end
 
   def run_preservation_cycle
-    item = queue.wait_next_item
-    return unless item.present?
+    entity_json = JSON.parse(queue.wait_next_item)
+    entity = {
+      type: entity_json['type'],
+      uuid: entity_json['uuid']
+    }
+    return unless entity[:type].present? && entity[:uuid].present?
 
     # add additional information about the error context to errors that occur while processing this item.
-    Rollbar.scoped(noid: item) do
-      begin
-        # Download AIP from Fedora, bag and tar AIP directory and cleanup after block code
-        PushmiPullyu::AIP.create(item) do |aip_filename, aip_directory|
-          # Push tarred AIP to swift API
-          deposited_file = swift.deposit_file(aip_filename, options[:swift][:container])
-          # Log successful preservation event to the log files
-          PushmiPullyu::Logging.log_preservation_event(deposited_file, aip_directory)
-        end
-      # rubocop:disable Lint/RescueException
-      rescue Exception => e
-        Rollbar.error(e)
-        logger.error(e)
-        # TODO: we could re-raise here and let the daemon die on any preservation error, or just log the issue and
-        # move on to the next item.
+    Rollbar.scoped(entity_uuid: entity[:uuid]) do
+      # Download AIP from Jupiter, bag and tar AIP directory and cleanup after
+      # block code
+      PushmiPullyu::AIP.create(entity) do |aip_filename, aip_directory|
+        # Push tarred AIP to swift API
+        deposited_file = swift.deposit_file(aip_filename, options[:swift][:container])
+        # Log successful preservation event to the log files
+        PushmiPullyu::Logging.log_preservation_event(deposited_file, aip_directory)
       end
+      # rubocop:disable Lint/RescueException
+    rescue Exception => e
+      Rollbar.error(e)
+      logger.error(e)
+      # TODO: we could re-raise here and let the daemon die on any preservation error, or just log the issue and
+      # move on to the next item.
       # rubocop:enable Lint/RescueException
     end
   end

data/lib/pushmi_pullyu/logging.rb

@@ -48,7 +48,7 @@ module PushmiPullyu::Logging
 
       message = "#{deposited_file.name} was successfully deposited into Swift Storage!\n"\
       "Here are the details of this preservation event:\n"\
-      "\tNOID: '#{deposited_file.name}'\n"\
+      "\tUUID: '#{deposited_file.name}'\n"\
       "\tTimestamp of Completion: '#{deposited_file.last_modified}'\n"\
       "\tAIP Checksum: '#{deposited_file.etag}'\n"\
       "\tMetadata: #{deposited_file.metadata}\n"\

data/lib/pushmi_pullyu/version.rb

@@ -1,3 +1,3 @@
 module PushmiPullyu
-  VERSION = '1.0.6'.freeze
+  VERSION = '2.0.0'.freeze
 end

data/pushmi_pullyu.gemspec

@@ -5,10 +5,10 @@ require 'pushmi_pullyu/version'
 Gem::Specification.new do |spec|
   spec.name          = 'pushmi_pullyu'
   spec.version       = PushmiPullyu::VERSION
-  spec.authors       = ['Shane Murnaghan']
-  spec.email         = ['murnagha@ualberta.ca']
+  spec.authors       = ['Shane Murnaghan', 'Omar Rodriguez-Arenas']
+  spec.email         = ['murnagha@ualberta.ca', 'orodrigu@ualberta.ca']
 
-  spec.summary       = 'Ruby application to manage flow of content from Fedora into Swift for preservation'
+  spec.summary       = 'Ruby application to manage flow of content from Jupiter into Swift for preservation'
   spec.homepage      = 'https://github.com/ualbertalib/pushmi_pullyu'
   spec.license       = 'MIT'
 
@@ -19,10 +19,9 @@ Gem::Specification.new do |spec|
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ['lib']
 
-  spec.required_ruby_version = '>= 2.3.1'
+  spec.required_ruby_version = '>= 2.5'
 
-  spec.add_runtime_dependency 'activerecord', '>= 5.1.5', '< 5.3.0'
-  spec.add_runtime_dependency 'activesupport', '~> 5.0'
+  spec.add_runtime_dependency 'activesupport', '>= 5', '< 7'
   spec.add_runtime_dependency 'bagit', '~> 0.4'
   spec.add_runtime_dependency 'connection_pool', '~> 2.2'
   spec.add_runtime_dependency 'daemons', '~> 1.2', '>= 1.2.4'
@@ -33,18 +32,19 @@ Gem::Specification.new do |spec|
   spec.add_runtime_dependency 'rdf-n3', '>= 1.99', '< 4.0'
   spec.add_runtime_dependency 'redis', '>= 3.3', '< 5.0'
   spec.add_runtime_dependency 'rest-client', '>= 1.8', '< 3.0'
-  spec.add_runtime_dependency 'rollbar', '~> 2.18'
+  spec.add_runtime_dependency 'rollbar', '>= 2.18', '< 4.0'
 
-  spec.add_development_dependency 'bundler', '~> 1.16'
+  spec.add_development_dependency 'bundler', '~> 2.0'
   spec.add_development_dependency 'coveralls', '~> 0.8'
-  spec.add_development_dependency 'danger', '~> 5.2'
+  spec.add_development_dependency 'danger', '~> 8.0'
   spec.add_development_dependency 'pry', '~> 0.10', '>= 0.10.4'
   spec.add_development_dependency 'pry-byebug', '~> 3.6'
-  spec.add_development_dependency 'rake', '~> 12.0'
+  spec.add_development_dependency 'rake', '~> 13.0'
   spec.add_development_dependency 'rspec', '~> 3.0'
   spec.add_development_dependency 'rubocop', '~> 0.51'
   spec.add_development_dependency 'rubocop-rspec', '~> 1.10'
   spec.add_development_dependency 'timecop', '~> 0.8'
-  spec.add_development_dependency 'vcr', '~> 4.0'
+  spec.add_development_dependency 'uuid', '~> 2.3.9'
+  spec.add_development_dependency 'vcr', '~> 5.0'
   spec.add_development_dependency 'webmock', '~> 3.3'
 end

metadata

@@ -1,49 +1,36 @@
 --- !ruby/object:Gem::Specification
 name: pushmi_pullyu
 version: !ruby/object:Gem::Version
-  version: 1.0.6
+  version: 2.0.0
 platform: ruby
 authors:
 - Shane Murnaghan
+- Omar Rodriguez-Arenas
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-11-29 00:00:00.000000000 Z
+date: 2021-01-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: activerecord
+  name: activesupport
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 5.1.5
+        version: '5'
     - - "<"
       - !ruby/object:Gem::Version
-        version: 5.3.0
+        version: '7'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 5.1.5
+        version: '5'
     - - "<"
       - !ruby/object:Gem::Version
-        version: 5.3.0
-- !ruby/object:Gem::Dependency
-  name: activesupport
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '5.0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '5.0'
+        version: '7'
 - !ruby/object:Gem::Dependency
   name: bagit
   requirement: !ruby/object:Gem::Requirement
@@ -230,30 +217,36 @@ dependencies:
   name: rollbar
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '2.18'
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '4.0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '2.18'
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '4.0'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.16'
+        version: '2.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.16'
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: coveralls
   requirement: !ruby/object:Gem::Requirement
@@ -274,14 +267,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '5.2'
+        version: '8.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '5.2'
+        version: '8.0'
 - !ruby/object:Gem::Dependency
   name: pry
   requirement: !ruby/object:Gem::Requirement
@@ -322,14 +315,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '12.0'
+        version: '13.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '12.0'
+        version: '13.0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -386,20 +379,34 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '0.8'
+- !ruby/object:Gem::Dependency
+  name: uuid
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.3.9
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.3.9
 - !ruby/object:Gem::Dependency
   name: vcr
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.0'
+        version: '5.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.0'
+        version: '5.0'
 - !ruby/object:Gem::Dependency
   name: webmock
   requirement: !ruby/object:Gem::Requirement
@@ -417,6 +424,7 @@ dependencies:
 description: 
 email:
 - murnagha@ualberta.ca
+- orodrigu@ualberta.ca
 executables:
 - pushmi_pullyu
 extensions: []
@@ -424,11 +432,15 @@ extra_rdoc_files: []
 files:
 - ".coveralls.yml"
 - ".editorconfig"
+- ".github/CODEOWNERS"
+- ".github/PULL_REQUEST_TEMPLATE"
+- ".github/workflows/ruby.yml"
 - ".gitignore"
 - ".hound.yml"
 - ".rspec"
 - ".rubocop.yml"
 - ".travis.yml"
+- CHANGELOG.md
 - Dangerfile
 - Gemfile
 - LICENSE.txt
@@ -445,10 +457,6 @@ files:
 - lib/pushmi_pullyu/aip.rb
 - lib/pushmi_pullyu/aip/creator.rb
 - lib/pushmi_pullyu/aip/downloader.rb
-- lib/pushmi_pullyu/aip/fedora_fetcher.rb
-- lib/pushmi_pullyu/aip/file_list_creator.rb
-- lib/pushmi_pullyu/aip/owner_email_editor.rb
-- lib/pushmi_pullyu/aip/user.rb
 - lib/pushmi_pullyu/cli.rb
 - lib/pushmi_pullyu/logging.rb
 - lib/pushmi_pullyu/preservation_queue.rb
@@ -470,16 +478,15 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.3.1
+      version: '2.5'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.6.11
+rubygems_version: 3.0.3
 signing_key: 
 specification_version: 4
-summary: Ruby application to manage flow of content from Fedora into Swift for preservation
+summary: Ruby application to manage flow of content from Jupiter into Swift for preservation
 test_files: []

data/lib/pushmi_pullyu/aip/fedora_fetcher.rb

@@ -1,66 +0,0 @@
-require 'net/http'
-
-class PushmiPullyu::AIP::FedoraFetcher
-
-  class FedoraFetchError < StandardError; end
-
-  RDF_FORMAT = 'text/rdf+n3'.freeze
-
-  def initialize(noid)
-    @noid = noid
-  end
-
-  def object_url(url_extra = nil)
-    url = "#{PushmiPullyu.options[:fedora][:url]}#{base_path}/#{pairtree}"
-    url += url_extra if url_extra
-    url
-  end
-
-  # Return true on success, raise an error otherwise
-  # (or use 'optional' to return false on 404)
-  def download_object(download_path, url_extra: nil,
-                      optional: false, is_rdf: false,
-                      should_add_user_email: false)
-
-    uri = URI(object_url(url_extra))
-
-    request = Net::HTTP::Get.new(uri)
-    request.basic_auth(PushmiPullyu.options[:fedora][:user],
-                       PushmiPullyu.options[:fedora][:password])
-
-    request['Accept'] = RDF_FORMAT if is_rdf
-
-    response = Net::HTTP.start(uri.hostname, uri.port) do |http|
-      http.request(request)
-    end
-
-    if response.is_a?(Net::HTTPSuccess)
-      body = if should_add_user_email
-               PushmiPullyu::AIP::OwnerEmailEditor.new(response.body).run
-             else
-               response.body
-             end
-      file = File.open(download_path, 'wb')
-      file.write(body)
-      file.close
-      return true
-    elsif response.is_a?(Net::HTTPNotFound)
-      raise FedoraFetchError unless optional
-
-      return false
-    else
-      raise FedoraFetchError
-    end
-  end
-
-  private
-
-  def pairtree
-    "#{@noid[0..1]}/#{@noid[2..3]}/#{@noid[4..5]}/#{@noid[6..7]}/#{@noid}"
-  end
-
-  def base_path
-    PushmiPullyu.options[:fedora][:base_path]
-  end
-
-end

data/lib/pushmi_pullyu/aip/file_list_creator.rb

@@ -1,121 +0,0 @@
-require 'rdf'
-require 'rdf/n3'
-require 'rest-client'
-
-class PushmiPullyu::AIP::FileListCreator
-
-  IANA = 'http://www.iana.org/assignments/relation/'.freeze
-  PREDICATES = {
-    proxy_for: RDF::URI('http://www.openarchives.org/ore/terms/proxyFor'),
-    first: RDF::URI(IANA + 'first'),
-    last: RDF::URI(IANA + 'last'),
-    prev: RDF::URI(IANA + 'prev'),
-    next: RDF::URI(IANA + 'next'),
-    has_part: RDF::URI('http://purl.org/dc/terms/hasPart')
-  }.freeze
-
-  class NoProxyURIFound < StandardError; end
-  class NoFirstProxyFound < StandardError; end
-  class FirstProxyHasPrev < StandardError; end
-  class ListSourceFileSetMismatch < StandardError; end
-
-  def initialize(list_source_uri, output_xml_file, file_set_uuids)
-    @uri = RDF::URI(list_source_uri)
-    @auth_uri = RDF::URI(list_source_uri)
-    @auth_uri.user = PushmiPullyu.options[:fedora][:user]
-    @auth_uri.password = PushmiPullyu.options[:fedora][:password]
-    @output_file = output_xml_file
-
-    # These are the known fileset uuids, used for validation
-    @file_set_uuids = file_set_uuids
-  end
-
-  def run
-    extract_list_source_uuids
-    raise ListSourceFileSetMismatch, @uri.to_s if @list_source_uuids.sort != @file_set_uuids.sort
-
-    write_output_file
-  end
-
-  def extract_list_source_uuids
-    # Note: raises IOError if can't find
-    #       raises RDF::ReaderError if can't parse
-    @graph = RDF::Graph.load(@auth_uri, validate: true)
-    @list_source_uuids = []
-
-    # Fetch first FileSet in list source
-    this_proxy = find_first_proxy
-
-    while @list_source_uuids.count <= num_proxies
-      @list_source_uuids << uuid_from_proxy(this_proxy)
-      next_proxy = find_next_proxy(this_proxy)
-
-      break if next_proxy.nil?
-
-      raise NextPreviousProxyMismatch if this_proxy != find_prev_proxy(next_proxy)
-
-      this_proxy = next_proxy
-    end
-
-    raise ProxyCountIncorrect if @list_source_uuids.count != num_proxies
-    raise LastProxyFailsValidation if this_proxy != find_last_proxy
-  end
-
-  def num_proxies
-    @num_proxies ||= @graph.query(subject: @uri, predicate: PREDICATES[:has_part]).count
-  end
-
-  def uuid_from_proxy(proxy_uri)
-    @graph.query(subject: proxy_uri, predicate: PREDICATES[:proxy_for]) do |statement|
-      return statement.object.to_s.split('/').last
-    end
-    raise NoProxyURIFound, proxy_uri.to_s
-  end
-
-  def find_first_proxy
-    @graph.query(subject: @uri, predicate: PREDICATES[:first]) do |statement|
-      first_uri = statement.object
-      # Validate that the first proxy doesn't have a previous one
-      raise FirstProxyHasPrev, @uri.to_s if find_prev_proxy(first_uri)
-
-      return first_uri
-    end
-    raise NoFirstProxyFound, @uri.to_s
-  end
-
-  def find_last_proxy
-    @graph.query(subject: @uri, predicate: PREDICATES[:last]) do |statement|
-      last_uri = statement.object
-      # Validate that the last proxy doesn't have a next one
-      raise LastProxyHasNext, @uri.to_s if find_next_proxy(last_uri)
-
-      return last_uri
-    end
-    raise LastProxyFound, @uri.to_s
-  end
-
-  def find_next_proxy(proxy_uri)
-    @graph.query(subject: proxy_uri, predicate: PREDICATES[:next]) do |statement|
-      return statement.object
-    end
-    nil
-  end
-
-  def find_prev_proxy(proxy_uri)
-    @graph.query(subject: proxy_uri, predicate: PREDICATES[:prev]) do |statement|
-      return statement.object
-    end
-    nil
-  end
-
-  def write_output_file
-    File.open(@output_file, 'w') do |file|
-      file.write("<file_order>\n")
-      @list_source_uuids.each do |uuid|
-        file.write("  <uuid>#{uuid}</uuid>\n")
-      end
-      file.write("</file_order>\n")
-    end
-  end
-
-end

data/lib/pushmi_pullyu/aip/owner_email_editor.rb

@@ -1,50 +0,0 @@
-require 'net/http'
-
-class PushmiPullyu::AIP::OwnerEmailEditor
-
-  OWNER_PREDICATE = RDF::URI('http://purl.org/ontology/bibo/owner').freeze
-
-  class NoOwnerPredicate < StandardError; end
-
-  def initialize(rdf_string)
-    @document = rdf_string
-  end
-
-  def run
-    setup_db_connection
-    is_modified = false
-    prefixes = nil
-    # Read once to load prefixes (the @things at the top of an n3 file)
-    RDF::N3::Reader.new(input = @document) do |reader|
-      reader.each_statement { |_statement| }
-      prefixes = reader.prefixes
-    end
-    new_body = RDF::N3::Writer.buffer(prefixes: prefixes) do |writer|
-      RDF::N3::Reader.new(input = @document) do |reader|
-        reader.each_statement do |statement|
-          if statement.predicate == OWNER_PREDICATE
-            user = PushmiPullyu::AIP::User.find(statement.object.to_i)
-            writer << [statement.subject, statement.predicate, user.email]
-            is_modified = true
-          else
-            writer << statement
-          end
-        end
-      end
-    end
-    return new_body if is_modified
-
-    raise NoOwnerPredicate
-  end
-
-  private
-
-  def setup_db_connection
-    ActiveRecord::Base.establish_connection(database_configuration)
-  end
-
-  def database_configuration
-    PushmiPullyu.options[:database][:url]
-  end
-
-end

data/lib/pushmi_pullyu/aip/user.rb

@@ -1,2 +0,0 @@
-class PushmiPullyu::AIP::User < ActiveRecord::Base
-end