Class: Dor::ContentMetadataDS
- Inherits:
-
ActiveFedora::OmDatastream
- Object
- ActiveFedora::OmDatastream
- Dor::ContentMetadataDS
- Defined in:
- lib/dor/datastreams/content_metadata_ds.rb
Instance Method Summary collapse
-
#add_virtual_resource(child_druid, child_resource) ⇒ Nokogiri::XML::Element
Copies the child’s resource into the parent (self) as a virtual resource.
-
#generate_also_available_as_node(objectId) ⇒ Nokogiri::XML::Element
Generates the XML tree for virtual resource relationship reference.
-
#generate_external_file_node(objectId, resourceId, fileId, mimetype) ⇒ Nokogiri::XML::Element
Generates the XML tree for externalFile references.
-
#move_resource(resource_name, new_position) ⇒ Nokogiri::XML::Element
You just had to have ordered lists in XML, didn’t you? Re-enumerate the sequence numbers affected.
-
#prefix ⇒ Object
maintain AF < 8 indexing behavior.
-
#rename_file(old_name, new_name) ⇒ Nokogiri::XML::Element
The file node.
-
#set_content_type(old_type, old_resource_type, new_type, new_resource_type) ⇒ Object
Set the content type (e.g. “book”) and the resource type (e.g. “book”) for all resources.
-
#singular_node(xpath) ⇒ Nokogiri::XML::Element
Only use this when you want the behavior of raising an exception if anything besides exactly one matching node is found.
-
#to_solr(solr_doc = {}, *_args) ⇒ Object
Terminology-based solrization is going to be painfully slow for large contentMetadata streams.
-
#unshelve_and_unpublish ⇒ Object
END: READ ONLY METHODS DATSTREAM WRITING METHODS.
- #update_attributes(file_name, publish, shelve, preserve, attributes = {}) ⇒ Object
- #update_file(file, old_file_id) ⇒ Object
-
#update_resource_label(resource_name, new_label) ⇒ Nokogiri::XML::Element
Updates old label OR creates a new one if necessary.
- #update_resource_type(resource_name, new_type) ⇒ Object
Instance Method Details
#add_virtual_resource(child_druid, child_resource) ⇒ Nokogiri::XML::Element
Copies the child’s resource into the parent (self) as a virtual resource. Assumes the resource isn’t a duplicate of an existing virtual or real resource.
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 143 def add_virtual_resource(child_druid, child_resource) # create a virtual resource element with attributes linked to the child and omit label ng_xml_will_change! sequence_max = ng_xml.search('//resource').map { |node| node[:sequence].to_i }.max || 0 resource = Nokogiri::XML::Element.new('resource', ng_xml) resource[:sequence] = sequence_max + 1 resource[:id] = "#{pid.gsub(/^druid:/, '')}_#{resource[:sequence]}" resource[:type] = child_resource[:type] # iterate over all the published files and link to them child_resource.search('file[@publish=\'yes\']').each do |file| resource << generate_external_file_node(child_druid, child_resource[:id], file[:id], file[:mimetype]) end resource << generate_also_available_as_node(child_druid) # attach the virtual resource as a sibling and return ng_xml.root << resource resource end |
#generate_also_available_as_node(objectId) ⇒ Nokogiri::XML::Element
Generates the XML tree for virtual resource relationship reference. For example:
<relationship type="alsoAvailableAs" objectId="druid:mn123pq4567" />
73 74 75 76 77 78 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 73 def generate_also_available_as_node(objectId) relationship = ng_xml.create_element 'relationship' relationship[:type] = 'alsoAvailableAs' relationship[:objectId] = objectId relationship end |
#generate_external_file_node(objectId, resourceId, fileId, mimetype) ⇒ Nokogiri::XML::Element
Generates the XML tree for externalFile references. For example:
<externalFile objectId="druid:mn123pq4567" resourceId="Image01" fileId="image_01.jp2000" mimetype="image/jp2" />
60 61 62 63 64 65 66 67 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 60 def generate_external_file_node(objectId, resourceId, fileId, mimetype) externalFile = ng_xml.create_element 'externalFile' externalFile[:objectId] = objectId externalFile[:resourceId] = resourceId externalFile[:fileId] = fileId externalFile[:mimetype] = mimetype externalFile end |
#move_resource(resource_name, new_position) ⇒ Nokogiri::XML::Element
You just had to have ordered lists in XML, didn’t you? Re-enumerate the sequence numbers affected
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 241 def move_resource(resource_name, new_position) ng_xml_will_change! node = singular_node('//resource[@id=\'' + resource_name + '\']') position = node['sequence'].to_i new_position = new_position.to_i # tolerate strings as a Legacy behavior return node if position == new_position # otherwise, is the resource being moved earlier in the sequence or later? up = new_position > position others = new_position..(up ? position - 1 : position + 1) # a range others.each do |i| item = ng_xml.at_xpath('/resource[@sequence=\'' + i.to_s + '\']') item['sequence'] = (up ? i - 1 : i + 1).to_s # if you're going up, everything else comes down and vice versa end node['sequence'] = new_position.to_s # set the node we already had last, so we don't hit it twice! node end |
#prefix ⇒ Object
maintain AF < 8 indexing behavior
275 276 277 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 275 def prefix '' end |
#rename_file(old_name, new_name) ⇒ Nokogiri::XML::Element
Returns the file node.
204 205 206 207 208 209 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 204 def rename_file(old_name, new_name) ng_xml_will_change! file_node = ng_xml.search('//file[@id=\'' + old_name + '\']').first file_node['id'] = new_name file_node end |
#set_content_type(old_type, old_resource_type, new_type, new_resource_type) ⇒ Object
Set the content type (e.g. “book”) and the resource type (e.g. “book”) for all resources
264 265 266 267 268 269 270 271 272 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 264 def set_content_type(old_type, old_resource_type, new_type, new_resource_type) ng_xml_will_change! ng_xml.search('/contentMetadata[@type=\'' + old_type + '\']').each do |node| node['type'] = new_type ng_xml.search('//resource[@type=\'' + old_resource_type + '\']').each do |resource| resource['type'] = new_resource_type end end end |
#singular_node(xpath) ⇒ Nokogiri::XML::Element
Only use this when you want the behavior of raising an exception if anything besides exactly one matching node is found. Otherwise just use .xpath, .at_xpath or .search.
44 45 46 47 48 49 50 51 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 44 def singular_node(xpath) node = ng_xml.search(xpath) len = node.length raise "#{xpath} not found" if len < 1 raise "#{xpath} duplicated: #{len} found" if len != 1 node.first end |
#to_solr(solr_doc = {}, *_args) ⇒ Object
Terminology-based solrization is going to be painfully slow for large contentMetadata streams. Just select the relevant elements instead. TODO: Call super()?
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 83 def to_solr(solr_doc = {}, *_args) doc = ng_xml return solr_doc unless doc.root['type'] preserved_size = 0 shelved_size = 0 counts = Hash.new(0) # default count is zero resource_type_counts = Hash.new(0) # default count is zero file_roles = ::Set.new mime_types = ::Set.new first_shelved_image = nil doc.xpath('contentMetadata/resource').sort { |a, b| a['sequence'].to_i <=> b['sequence'].to_i }.each do |resource| counts['resource'] += 1 resource_type_counts[resource['type']] += 1 if resource['type'] resource.xpath('file').each do |file| counts['content_file'] += 1 preserved_size += file['size'].to_i if file['preserve'] == 'yes' shelved_size += file['size'].to_i if file['shelve'] == 'yes' if file['shelve'] == 'yes' counts['shelved_file'] += 1 first_shelved_image ||= file['id'] if file['id'] =~ /jp2$/ end mime_types << file['mimetype'] file_roles << file['role'] if file['role'] end end solr_doc['content_type_ssim'] = doc.root['type'] solr_doc['content_file_mimetypes_ssim'] = mime_types.to_a solr_doc['content_file_count_itsi'] = counts['content_file'] solr_doc['shelved_content_file_count_itsi'] = counts['shelved_file'] solr_doc['resource_count_itsi'] = counts['resource'] solr_doc['preserved_size_dbtsi'] = preserved_size # double (trie) to support very large sizes solr_doc['shelved_size_dbtsi'] = shelved_size # double (trie) to support very large sizes solr_doc['resource_types_ssim'] = resource_type_counts.keys if resource_type_counts.size > 0 solr_doc['content_file_roles_ssim'] = file_roles.to_a if file_roles.size > 0 resource_type_counts.each do |key, count| solr_doc["#{key}_resource_count_itsi"] = count end # first_shelved_image is neither indexed nor multiple solr_doc['first_shelved_image_ss'] = first_shelved_image unless first_shelved_image.nil? solr_doc end |
#unshelve_and_unpublish ⇒ Object
END: READ ONLY METHODS DATSTREAM WRITING METHODS
130 131 132 133 134 135 136 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 130 def unshelve_and_unpublish ng_xml.xpath('/contentMetadata/resource//file').each_with_index do |file_node, index| ng_xml_will_change! if index == 0 file_node['publish'] = 'no' file_node['shelve'] = 'no' end end |
#update_attributes(file_name, publish, shelve, preserve, attributes = {}) ⇒ Object
167 168 169 170 171 172 173 174 175 176 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 167 def update_attributes(file_name, publish, shelve, preserve, attributes = {}) ng_xml_will_change! file_node = ng_xml.search('//file[@id=\'' + file_name + '\']').first file_node['publish'] = publish file_node['shelve'] = shelve file_node['preserve'] = preserve attributes.each do |key, value| file_node[key] = value end end |
#update_file(file, old_file_id) ⇒ Object
180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 180 def update_file(file, old_file_id) ng_xml_will_change! file_node = ng_xml.search('//file[@id=\'' + old_file_id + '\']').first file_node['id'] = file[:name] %i[md5 sha1].each do |algo| next if file[algo].nil? checksum_node = ng_xml.search('//file[@id=\'' + old_file_id + '\']/checksum[@type=\'' + algo.to_s + '\']').first if checksum_node.nil? checksum_node = Nokogiri::XML::Node.new('checksum', ng_xml) file_node.add_child(checksum_node) end checksum_node['type'] = algo.to_s checksum_node.content = file[algo] end %i[size shelve preserve publish role].each do |x| file_node[x.to_s] = file[x] if file[x] end end |
#update_resource_label(resource_name, new_label) ⇒ Nokogiri::XML::Element
Updates old label OR creates a new one if necessary
215 216 217 218 219 220 221 222 223 224 225 226 227 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 215 def update_resource_label(resource_name, new_label) ng_xml_will_change! node = singular_node('//resource[@id=\'' + resource_name + '\']') labels = node.xpath('./label') if labels.length == 0 label_node = Nokogiri::XML::Node.new('label', ng_xml) # create a label label_node.content = new_label node.add_child(label_node) else labels.first.content = new_label end node end |
#update_resource_type(resource_name, new_type) ⇒ Object
231 232 233 234 |
# File 'lib/dor/datastreams/content_metadata_ds.rb', line 231 def update_resource_type(resource_name, new_type) ng_xml_will_change! singular_node('//resource[@id=\'' + resource_name + '\']')['type'] = new_type end |