Class: RDF::Vocabulary

Inherits:
Object
  • Object
show all
Defined in:
lib/rdf/vocab/extensions.rb

Defined Under Namespace

Modules: VocabFormatExtensions

Class Method Summary collapse

Class Method Details

._orig_each {|klass| ... } ⇒ Enumerator

Enumerates known RDF vocabulary classes.

Yields:

  • (klass)

Yield Parameters:

  • klass (Class)

Returns:

  • (Enumerator)

17
# File 'lib/rdf/vocab/extensions.rb', line 17

alias_method :_orig_each, :each

._orig_from_symObject


43
# File 'lib/rdf/vocab/extensions.rb', line 43

alias_method :_orig_from_sym, :from_sym

.each(&block) ⇒ Object


18
19
20
21
22
23
24
25
26
27
28
29
30
# File 'lib/rdf/vocab/extensions.rb', line 18

def each(&block)
  if self.equal?(Vocabulary)
    # This is needed since all vocabulary classes are defined using
    # Ruby's autoloading facility, meaning that `@@subclasses` will be
    # empty until each subclass has been touched or require'd.
    RDF::Vocab::VOCABS.each do |n, params|
      clsname = params[:class_name].to_sym
      RDF::Vocab.const_get(clsname) # Forces class to load
    end unless instance_variable_defined?(:@classes_loaded) && @classes_loaded
    @classes_loaded = true
  end
  _orig_each(&block)
end

.from_sym(sym) ⇒ RDF::Vocabulary

Return the vocabulary based on it's class_name symbol

Parameters:

  • sym (Symbol)

Returns:


49
50
51
52
# File 'lib/rdf/vocab/extensions.rb', line 49

def from_sym(sym)
  RDF::Vocab.const_defined?(sym.to_sym) ?
    RDF::Vocab.const_get(sym.to_sym) : _orig_from_sym(sym)
end

.limit_vocabs(*vocabs) ⇒ Array<RDF::Vocabulary>

Limits iteration over vocabularies to just those selected

Examples:

limit to set of vocabularies by symbol

RDF::Vocabulary.limit_vocabs(:rdf, :rdfs, :schema)
RDF::Vocabulary.find_term('http://schema.org/CreativeWork').pname
# => 'schema:CreativeWork'

limit to set of vocabularies by class name

RDF::Vocabulary.limit_vocabs(RDF::RDFV, RDF::RDFS, RDF::Vocab::SCHEMA)
RDF::Vocabulary.find_term('http://schema.org/CreativeWork').pname
# => 'schema:CreativeWork'

Parameters:

  • vocabs (Array<symbol, RDF::Vocabulary>)

    A list of vocabularies (symbols or classes) which may be returned by each. Also limits vocabularies that will be inspeced for other methods. Set to nil, or an empty array to reset.

Returns:


73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
# File 'lib/rdf/vocab/extensions.rb', line 73

def limit_vocabs(*vocabs)
  @vocabs = if Array(vocabs).empty?
    nil
  else
    @classes_loaded = true
    vocabs.map do |vocab|
      if vocab == :rdf || vocab == :rdfv
        RDF::RDFV
      elsif vocab.is_a?(Symbol) && RDF::Vocab::VOCABS.key?(vocab)
        RDF::Vocab.const_get(RDF::Vocab::VOCABS[vocab][:class_name].to_sym)
      else
        vocab
      end
    end.compact
  end
end

.to_html(graph: nil, prefixes: nil, jsonld: nil, template: nil) ⇒ String

Generate HTML+RDFa representation, specific to vocabularies. This uses generated JSON-LD and a Haml template.

Parameters:

  • graph (RDF::Queryable) (defaults to: nil)

    Optional graph, otherwise uses statements from vocabulary.

  • prefixes (Hash{#to_sym => String}) (defaults to: nil)

    to add to output

  • jsonld (String, Hash) (defaults to: nil)

    If not provided, the `to_jsonld` method is used to generate it.

  • template (String) (defaults to: nil)

    The path to a Haml or ERB template used to generate the output using the JSON-LD serialization

Returns:

  • (String)

329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
# File 'lib/rdf/vocab/extensions.rb', line 329

def to_html(graph: nil, prefixes: nil, jsonld: nil, template: nil)
  # Find namespaces used in the vocabulary
  graph = RDF::Graph.new {|g| each_statement {|s| g << s}} if graph.nil? || graph.empty?

  # Get JSON as an object
  json = case jsonld
  when String then ::JSON.parse(File.read jsonld)
  when Hash   then jsonld
  else
    ::JSON.parse(to_jsonld(graph: graph, prefixes: prefixes))
  end
  raise "Expected JSON-LD data within the '@graph' key" unless json.has_key?('@graph')

  template ||= File.expand_path("../../../../etc/template.erb", __FILE__)

  prefixes = vocab_prefixes(graph).merge(prefixes || {})
  prefixes[:owl] = RDF::OWL.to_uri.to_s

  # Make sure ontology is typed
  json['@graph']['@type'] ||= ['owl:Ontology']

  jld_context = ::JSON::LD::Context.new.parse([json['@context'], json['@graph']['@context']])

  # Expand the JSON-LD to normalize accesses
  expanded = ::JSON::LD::API.expand(json).first
  expanded.delete('@reverse')

  # Re-compact keys
  expanded = expanded.inject({}) do |memo, (k, v)|
    term = RDF::Vocabulary.find_term(k)
    k = term.pname if term
    memo.merge(k => v)
  end

  # Normalize label accessors
  expanded['rdfs:label'] ||= %w(dc:title dc11:title skos:prefLabel).inject(nil) do |memo, key|
    memo || expanded[key]
  end || [{'@value' => json['@graph']['@id']}]
  %w(rdfs_classes rdfs_properties rdfs_datatypes rdfs_instances).each do |section|
    next unless json['@graph'][section]
    json['@graph'][section].each do |node|
      node['rdfs:label'] ||= %w(dc:title dc11:title skos:prefLabel).inject do |memo, key|
        memo || node[key]
      end || [{'@value' => node['@id']}]
    end
  end

  # Expand each part separately, as well.
  %w(rdfs_classes rdfs_properties rdfs_datatypes rdfs_instances).each do |section|
    next unless json['@graph'][section]
    expanded_section = ::JSON::LD::API.expand(json['@graph'][section], expandContext: jld_context)
    # Re-compact keys
    expanded[section] = expanded_section.map do |node|
      node.inject({}) do |memo, (k, v)|
        term = RDF::Vocabulary.find_term(k)
        k = term.pname if term
        memo.merge(k => v)
      end
    end
  end

  # Template invoked with expanded JSON-LD with outer object including `rdfs_classes`, `rdfs_properties`, and `rdf_instances` sections.
  case template
  when /.haml$/
    require 'haml'
    haml = Haml::Engine.new(File.read(template))
    haml.render(self, ont: expanded, context: json['@context'], prefixes: prefixes)
  when /.erb$/
    require 'erubis'
    eruby = Erubis::FastEruby.new(File.read(template))
    eruby.evaluate(binding: self, ont: expanded, context: json['@context'], prefixes: prefixes)
  else
    raise "Unknown template type #{template}. Should have '.erb' or '.haml' extension"
  end
end

.to_jsonld(graph: nil, prefixes: nil) ⇒ String

Generate JSON-LD representation, specific to vocabularies

Parameters:

  • graph (RDF::Queryable) (defaults to: nil)

    Optional graph, otherwise uses statements from vocabulary.

  • prefixes (Hash{#to_sym => String}) (defaults to: nil)

    to add to output

Returns:

  • (String)

176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
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
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
# File 'lib/rdf/vocab/extensions.rb', line 176

def to_jsonld(graph: nil, prefixes: nil)
  require 'json/ld'

  context = {}
  rdfs_context = ::JSON.parse %({
    "dc:title": {"@container": "@language"},
    "dc:description": {"@container": "@language"},
    "dc:date": {"@type": "xsd:date"},
    "rdfs:comment": {"@container": "@language"},
    "rdfs:domain": {"@type": "@vocab"},
    "rdfs:label": {"@container": "@language"},
    "rdfs:range": {"@type": "@vocab"},
    "rdfs:seeAlso": {"@type": "@id"},
    "rdfs:subClassOf": {"@type": "@vocab"},
    "rdfs:subPropertyOf": {"@type": "@vocab"},
    "schema:domainIncludes": {"@type": "@vocab"},
    "schema:rangeIncludes": {"@type": "@vocab"},
    "owl:equivalentClass": {"@type": "@vocab"},
    "owl:equivalentProperty": {"@type": "@vocab"},
    "owl:oneOf": {"@container": "@list", "@type": "@vocab"},
    "owl:imports": {"@type": "@id"},
    "owl:versionInfo": {"@type": "@id"},
    "owl:inverseOf": {"@type": "@vocab"},
    "owl:unionOf": {"@type": "@vocab", "@container": "@list"},
    "rdfs_classes": {"@reverse": "rdfs:isDefinedBy", "@type": "@id"},
    "rdfs_properties": {"@reverse": "rdfs:isDefinedBy", "@type": "@id"},
    "rdfs_datatypes": {"@reverse": "rdfs:isDefinedBy", "@type": "@id"},
    "rdfs_instances": {"@reverse": "rdfs:isDefinedBy", "@type": "@id"}
  })
  rdfs_classes, rdfs_properties, rdfs_datatypes, rdfs_instances = [], [], [], [], []

  ontology = {
    "@context" => rdfs_context,
    "@id" => to_uri.to_s
  }

  # Find namespaces used in the vocabulary
  graph = RDF::Graph.new {|g| each_statement {|s| g << s}} if graph.nil? || graph.empty?

  prefixes = vocab_prefixes(graph).merge(prefixes || {})
  prefixes.each do |pfx, uri|
    context[pfx.to_s] = uri.to_s unless pfx.to_s.empty?
  end

  # Determine the category for each subject in the vocabulary graph
  cats = subject_categories(graph)

  # Generate term definitions from graph subjects
  cats.values.flatten.each do |term|
    next unless Array(term.qname).length == 2
    context[term.qname.last.to_s] = term.to_uri.to_s
  end

  # Parse the two contexts so we know what terms are in scope
  jld_context = ::JSON::LD::Context.new.parse([context, rdfs_context])

  {
    ont: {
      heading:  "# #{__name__.split('::').last} Vocabulary definition\n",
      bucket:   ontology,
    },
    classes: {
      heading:  "# Class definitions\n",
      bucket:   rdfs_classes,
      rev_prop: "rdfs_classes"
    },
    properties: {
      heading:  "# Property definitions\n",
      bucket:   rdfs_properties,
      rev_prop: "rdfs_properties"
    },
    datatypes: {
      heading:  "# Datatype definitions\n",
      bucket:   rdfs_datatypes,
      rev_prop: "rdfs_datatypes"
    },
    other: {
      heading:  "# Other definitions\n",
      bucket:   rdfs_instances,
      rev_prop: "rdfs_instances"
    }
  }.each do |key, hash|
    next unless cats[key]

    cats[key].each do |subject|
      node = {"@id" => subject.pname}
      po = {}

      # Group predicates with their values
      graph.query({subject: subject}) do |statement|
        # Sanity check this, as these are set to an empty string if not defined.
        next if [RDF::RDFS.label, RDF::RDFS.comment].include?(statement.predicate) && statement.object.to_s.empty?
        po[statement.predicate] ||= []
        po[statement.predicate] << statement.object
      end

      next if po.empty?

      node['@type'] = po.delete(RDF.type).map {|t| jld_context.compact_iri(t, vocab: true)}

      po.each do |predicate, objects|
        term = jld_context.compact_iri(predicate, vocab: true)
        node[term] = if jld_context.container(term) == '@language'
          lang_map = objects.inject({}) do |memo, o|
            raise "Language-mapped term #{term} with non plain-literal #{o.inspect}" unless o.literal? && o.plain?
            memo.merge(o.language.to_s => o.value)
          end
          # Don't use language map if there's only one entry with no language
          lang_map = lang_map[""] if lang_map.keys == [""]
          [lang_map]
        else
          objects.map do |o|
            expanded_value = jld_context.expand_value(term, o)
            jld_context.compact_value(term, expanded_value)
          end
        end
      end

      node.each do |property, values|
        case values.length
        when 0 then node.delete(property)
        when 1 then node[property] = values.first
        end
      end

      # Either set bucket from node, or append node to bucket
      if hash[:bucket].is_a?(Hash)
        hash[:bucket].merge!(node)
      else
        ontology[hash[:rev_prop]] ||= hash[:bucket]
        hash[:bucket] << node
      end
    end
  end

  # Serialize result
  {
    "@context" => context,
    "@graph" => ontology
  }.to_json(::JSON::LD::JSON_STATE)
rescue LoadError
  # No JSON-LD serialization unless gem loaded
end

.to_ttl(graph: nil, prefixes: nil) ⇒ String

Generate Turtle representation, specific to vocabularies

Parameters:

  • graph (RDF::Queryable) (defaults to: nil)

    Optional graph, otherwise uses statements from vocabulary.

  • prefixes (Hash{#to_sym => String}) (defaults to: nil)

    to add to output

Returns:

  • (String)

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
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
# File 'lib/rdf/vocab/extensions.rb', line 96

def to_ttl(graph: nil, prefixes: nil)
  require 'rdf/turtle'
  output = []

  # Find namespaces used in the vocabulary
  graph = RDF::Graph.new {|g| each_statement {|s| g << s}} if graph.nil? || graph.empty?

  prefixes = vocab_prefixes(graph).merge(prefixes || {})
  pfx_width = prefixes.keys.map(&:to_s).map(&:length).max
  prefixes.each do |pfx, uri|
    output << "@prefix %*s: <%s> .\n" % [pfx_width, pfx, uri]
  end

  # Determine the category for each subject in the vocabulary graph
  cats = subject_categories(graph)

  writer = RDF::Turtle::Writer.new(StringIO.new, prefixes: prefixes)

  {
    ont: {
      heading:  "# #{__name__.split('::').last} Vocabulary definition\n"
    },
    classes: {
      heading:  "# Class definitions\n"
    },
    properties: {
      heading:  "# Property definitions\n"
    },
    datatypes: {
      heading:  "# Datatype definitions\n"
    },
    other: {
      heading:  "# Other definitions\n"
    }
  }.each do |key, hash|
    next unless cats[key]

    output << "\n\n#{hash[:heading]}"

    cats[key].each do |subject|
      po = {}

      # Group predicates with their values
      graph.query({subject: subject}) do |statement|
        # Sanity check this, as these are set to an empty string if not defined.
        next if [RDF::RDFS.label, RDF::RDFS.comment].include?(statement.predicate) && statement.object.to_s.empty?
        po[statement.predicate] ||= []
        po[statement.predicate] << statement.object
      end

      next if po.empty?

      po_list = []
      unless (types = po.delete(RDF.type)).empty?
        po_list << 'a ' + types.map {|o| writer.format_term(o)}.join(", ")
      end

      # Serialize other predicate/objects
      po.each do |predicate, objects|
        resource = predicate.qname ? predicate.pname : "<#{predicate}>"
        po_list << resource + ' ' + objects.map {|o| writer.format_term(o)}.join(", ")
      end

      # Output statements for this subject
      subj = subject.qname ? subject.pname : "<#{subject}>"
      output << "#{subj} " + po_list.join(";\n  ") + "\n  .\n"
    end
  end

  output.join("")
rescue LoadError
  # No Turtle serialization unless gem loaded
end

.value_to_html(property, value, tag) ⇒ Object

Create HTML for values (Helper method, needs to be public)


407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
# File 'lib/rdf/vocab/extensions.rb', line 407

def value_to_html(property, value, tag)
  value.map do |v|
    %(<#{tag} property="#{property}") +
    if v['@value']
      (v['@language'] ? %( lang="#{v['@language']}") : "") +
      (v['@type'] ? %( datatype="#{RDF::URI(v['@type']).pname}") : "") +
      %(>#{v['@value']})
    elsif v['@id']
      %( resource="#{RDF::URI(v['@id']).pname}">#{RDF::URI(v['@id']).pname})
    else
      raise "Unknown value type: #{v.inspect}, #{property}"
    end +
    %(</#{tag}>)
  end.join("\n")
end

.vocab_mapHash{Symbol => Hash{Symbol => String}}

A hash of all vocabularies by prefix showing relevant URI and associated vocabulary Class Name

alias_method :_orig_vocab_map, :vocab_map

Returns:

  • (Hash{Symbol => Hash{Symbol => String}})

38
39
40
41
# File 'lib/rdf/vocab/extensions.rb', line 38

def vocab_map
  @vocab_map ||= RDF::VOCABS.transform_values(&:freeze).merge(
    RDF::Vocab::VOCABS.transform_values(&:freeze))
end