pushmi_pullyu 1.0.2 → 2.0.0

data/lib/pushmi_pullyu/aip/fedora_fetcher.rb

@@ -1,65 +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

@@ -1,118 +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

@@ -1,62 +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
-    ensure_database_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 ensure_database_connection
-    return if ActiveRecord::Base.connected?
-    ActiveRecord::Base.establish_connection(database_configuration)
-  end
-
-  def database_configuration
-    # Config either from URL, or with more granular options (the later taking precedence)
-    config = {}
-    uri = URI.parse(PushmiPullyu.options[:database][:url])
-    config[:adapter] = PushmiPullyu.options[:database][:adaptor] || uri.scheme
-    config[:host] = PushmiPullyu.options[:database][:host] || uri.host
-    config[:database] = PushmiPullyu.options[:database][:database] || uri.path.split('/')[1].to_s
-    config[:username] = PushmiPullyu.options[:database][:username] || uri.user
-    config[:password] = PushmiPullyu.options[:database][:password] || uri.password
-    params = CGI.parse(uri.query || '')
-    config[:encoding] = PushmiPullyu.options[:database][:encoding] || params['encoding'].to_a.first
-    config[:pool] = PushmiPullyu.options[:database][:pool] || params['pool'].to_a.first
-    config
-  end
-
-end

data/lib/pushmi_pullyu/aip/user.rb

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