Class: RDF::SAK::Context

Inherits:
Object
  • Object
show all
Includes:
Util, XML::Mixup
Defined in:
lib/rdf/sak.rb

Defined Under Namespace

Classes: Document

Constant Summary collapse

SKOS_HIER = 
[
  {
    element: :subject,
    pattern: -> c, p { [nil, p, c] },
    preds: [RDF::Vocab::SKOS.broader,  RDF::Vocab::SKOS.broaderTransitive],
  },
  {
    element: :object,
    pattern: -> c, p { [c, p, nil] },
    preds: [RDF::Vocab::SKOS.narrower, RDF::Vocab::SKOS.narrowerTransitive],
  }
]
AUTHOR_SPEC = 
[
  ['By:', [RDF::Vocab::BIBO.authorList, RDF::Vocab::DC.creator]],
  ['With:', [RDF::Vocab::BIBO.contributorList, RDF::Vocab::DC.contributor]],
  ['Edited by:', [RDF::Vocab::BIBO.editorList, RDF::Vocab::BIBO.editor]],
  ['Translated by:', [RDF::Vocab::BIBO.translator]],
].freeze
CONCEPTS =

generate skos concept schemes

Util.all_related(RDF::Vocab::SKOS.Concept).to_set
CSV_PRED = 
{
  audience:    RDF::Vocab::DC.audience,
  nonaudience: CI['non-audience'],
  subject:     RDF::Vocab::DC.subject,
  assumes:     CI.assumes,
  introduces:  CI.introduces,
  mentions:    CI.mentions,
}
DSD_SEQ = 
%i[characters words blocks sections
min low-quartile median high-quartile max mean sd].freeze
TH_SEQ = 
%w[Document Abstract Created Modified Characters Words Blocks 
Sections Min Q1 Median Q3 Max Mean SD].map { |t| { [t] => :th } }

Constants included from Util

Util::SCHEME_RANK, Util::XHTMLNS, Util::XHV, Util::XPATHNS

Instance Attribute Summary collapse

Instance Method Summary collapse

Methods included from Util

#all_related, asserted_types, base_for, canonical_uri, canonical_uuid, cmp_label, #cmp_resource, #coerce_node_spec, dates_for, #dehydrate, #get_base, #get_prefixes, #invert_struct, label_for, #modernize, #node_matches?, objects_for, #predicate_set, #prefix_subset, #prepare_collation, published?, #rehydrate, #reindent, replacements_for, #resolve_curie, #smush_struct, #split_pp, #split_pp2, #split_qp, struct_for, #subject_for, subjects_for, #subtree, #terminal_slug, #title_tag, traverse_links, #type_strata, #uri_pp

Constructor Details

#initialize(graph: nil, base: nil, config: nil, type: nil) ⇒ RDF::SAK::Context

Initialize a context.

Parameters:

  • graph (defaults to: nil)
  • base (defaults to: nil)
  • config (defaults to: nil)
  • type (defaults to: nil) 


229
230
231
232
233
234
235
236
237
238
239
240
241
 
# File 'lib/rdf/sak.rb', line 229

def initialize graph: nil, base: nil, config: nil, type: nil
  # RDF::Reasoner.apply(:rdfs, :owl)

  @config = coerce_config config

  graph ||= @config[:graph] if @config[:graph]
  base  ||= @config[:base]  if @config[:base]
  
  @graph  = coerce_graph graph, type: type
  @base   = RDF::URI.new base.to_s if base
  @ucache = RDF::Util::Cache.new(-1)
  @scache = {} # wtf rdf util cache doesn't like booleans
end

Instance Attribute Details

#base ⇒ Object (readonly)

Returns the value of attribute base.



218
219
220
 
# File 'lib/rdf/sak.rb', line 218

def base
  @base
end

#config ⇒ Object (readonly)

Returns the value of attribute config.



218
219
220
 
# File 'lib/rdf/sak.rb', line 218

def config
  @config
end

#graph ⇒ Object (readonly)

Returns the value of attribute graph.



218
219
220
 
# File 'lib/rdf/sak.rb', line 218

def graph
  @graph
end

Instance Method Details

#abbreviate(term, prefixes: , vocab: nil, noop: true, sort: true) ⇒ String

Parameters:

  • term (RDF::Term)
  • prefixes (Hash) (defaults to: )

Returns:

  • (String) 


260
261
262
263
 
# File 'lib/rdf/sak.rb', line 260

def abbreviate term, prefixes: @config[:prefixes],
    vocab: nil, noop: true, sort: true
  super term, prefixes: prefixes || {}, vocab: vocab, noop: noop, sort: sort
end

#all_internal_docs(published: true, exclude: []) ⇒ Object 



1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
 
# File 'lib/rdf/sak.rb', line 1173

def all_internal_docs published: true, exclude: []
  # find all UUIDs that are documents
  docs = all_of_type(RDF::Vocab::FOAF.Document,
    exclude: exclude).select { |x| x =~ /^urn:uuid:/ }

  # prune out all but the published documents if specified
  if published
    p = RDF::Vocab::BIBO.status
    o = RDF::Vocabulary.find_term(
      'http://purl.org/ontology/bibo/status/published')
    docs = docs.select do |s|
      @graph.has_statement? RDF::Statement(s, p, o)
    end
  end
  
  docs
end

#all_of_type(rdftype, exclude: []) ⇒ Array

Obtain every subject that is rdf:type the given type or its subtypes.

Parameters:

  • rdftype (RDF::Term)

Returns:

  • (Array) 


295
296
297
298
299
300
301
302
303
304
 
# File 'lib/rdf/sak.rb', line 295

def all_of_type rdftype, exclude: []
  exclude = term_list exclude
  t = RDF::Vocabulary.find_term(rdftype) or raise "No type #{rdftype.to_s}"
  out = []
  (all_types & all_related(t) - exclude).each do |type|
    out += @graph.query([nil, RDF.type, type]).subjects
  end
  
  out.uniq
end

#all_types ⇒ Array

Obtain everything in the graph that is an ‘rdf:type` of something.

Returns:

  • (Array) 


285
286
287
 
# File 'lib/rdf/sak.rb', line 285

def all_types
  @graph.query([nil, RDF.type, nil]).objects.uniq
end

#asserted_types(subject, type = nil) ⇒ Array

Obtain all and only the rdf:types directly asserted on the subject.

Parameters:

  • subject (RDF::Resource)
  • type (RDF::Term, :to_a) (defaults to: nil)

Returns:

  • (Array) 


313
314
315
 
# File 'lib/rdf/sak.rb', line 313

def asserted_types subject, type = nil
  Util.asserted_types @graph, subject, type
end

#audiences_for(uuid, proximate: false, invert: false) ⇒ Object 



511
512
513
514
515
516
517
518
519
520
521
 
# File 'lib/rdf/sak.rb', line 511

def audiences_for uuid, proximate: false, invert: false
  p = invert ? CI['non-audience'] : RDF::Vocab::DC.audience
  return @graph.query([uuid, p, nil]).objects if proximate

  out = []
  @graph.query([uuid, p, nil]).objects.each do |o|
    out += sub_concepts o
  end

  out
end

#authors_for(subject, unique: false, contrib: false) ⇒ RDF::Value, Array

Assuming the subject is a thing that has authors, return the list of authors. Try bibo:authorList first for an explicit ordering, then continue to the various other predicates.

Parameters:

  • subject (RDF::Resource)
  • unique (false, true) (defaults to: false) —

    only return the first author

  • contrib (false, true) (defaults to: false) —

    return contributors instead of authors

Returns:

  • (RDF::Value, Array) 


423
424
425
 
# File 'lib/rdf/sak.rb', line 423

def authors_for subject, unique: false, contrib: false
  Util.authors_for @graph, subject, unique: unique, contrib: contrib
end

#canonical_uri(subject, unique: true, rdf: true, slugs: false, fragment: false) ⇒ RDF::URI, ...

Obtain the “best” dereferenceable URI for the subject. Optionally returns all candidates.

Parameters:

  • subject (RDF::Resource)
  • unique (true, false) (defaults to: true) —

    flag for unique return value

  • rdf (true, false) (defaults to: true) —

    flag to specify RDF::URI vs URI

  • slugs (true, false) (defaults to: false) —

    flag to include slugs

  • fragment (true, false) (defaults to: false) —

    flag to include fragment URIs

Returns:

  • (RDF::URI, URI, Array) 


341
342
343
344
345
 
# File 'lib/rdf/sak.rb', line 341

def canonical_uri subject,
    unique: true, rdf: true, slugs: false, fragment: false
  Util.canonical_uri @graph, subject, base: @base,
    unique: unique, rdf: rdf, slugs: slugs, fragment: fragment
end

#canonical_uuid(uri, unique: true, published: false) ⇒ RDF::URI, Array

Obtain the canonical UUID for the given URI

Parameters:

  • uri (RDF::URI, URI, to_s) —

    the subject of the inquiry

  • unique (true, false) (defaults to: true) —

    return a single resource/nil or an array

  • published (true, false) (defaults to: false) —

    whether to restrict to published docs

Returns:

  • (RDF::URI, Array) 


325
326
327
328
 
# File 'lib/rdf/sak.rb', line 325

def canonical_uuid uri, unique: true, published: false
  Util.canonical_uuid @graph, uri, unique: unique,
    published: published, scache: @scache, ucache: @ucache, base: @base
end

#dates_for(subject, predicate: RDF::Vocab::DC.date, datatype: [RDF::XSD.date, RDF::XSD.dateTime]) ⇒ Array

Obtain dates for the subject as instances of Date(Time). This is just shorthand for a common application of ‘objects_for`.

Parameters:

  • subject
  • predicate (defaults to: RDF::Vocab::DC.date)
  • datatype (defaults to: [RDF::XSD.date, RDF::XSD.dateTime])

Returns:

  • (Array) —

    of dates



394
395
396
397
 
# File 'lib/rdf/sak.rb', line 394

def dates_for subject, predicate: RDF::Vocab::DC.date,
    datatype: [RDF::XSD.date, RDF::XSD.dateTime]
  Util.dates_for @graph, subject, predicate: predicate, datatype: datatype
end

#formats_for(subject, predicate: RDF::Vocab::DC.format, datatype: [RDF::XSD.token]) ⇒ Array

Obtain any specified MIME types for the subject. Just shorthand for a common application of ‘objects_for`.

Parameters:

  • subject
  • predicate (defaults to: RDF::Vocab::DC.format)
  • datatype (defaults to: [RDF::XSD.token])

Returns:

  • (Array) —

    of internet media types



408
409
410
411
 
# File 'lib/rdf/sak.rb', line 408

def formats_for subject, predicate: RDF::Vocab::DC.format,
    datatype: [RDF::XSD.token]
  Util.objects_for @graph, subject, predicate: predicate, datatype: datatype
end

#generate_atom_feed(id, published: true, related: []) ⇒ Object 



1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
 
# File 'lib/rdf/sak.rb', line 1191

def generate_atom_feed id, published: true, related: []
  raise 'ID must be a resource' unless id.is_a? RDF::Resource

  # prepare relateds
  raise 'related must be an array' unless related.is_a? Array
  related -= [id]

  # feed = struct_for id

  faudy = audiences_for id
  faudn = audiences_for id, invert: true
  faudy -= faudn

  docs = all_internal_docs published: published

  # now we create a hash keyed by uuid containing the metadata
  authors = {}
  titles  = {}
  dates   = {}
  entries = {}
  latest  = nil
  docs.each do |uu|
    # basically make a jsonld-like structure
    #rsrc = struct_for uu

    indexed = objects_for uu, RDF::SAK::CI.indexed, only: :literal
    next if !indexed.empty? and indexed.any? { |f| f == false }

    # get id (got it already duh)
    
    # get audiences
    audy = audiences_for uu, proximate: true
    audn = audiences_for uu, proximate: true, invert: true

    #warn "#{faudy.to_s} & #{faud"

    skip = false
    if audy.empty?
      # an unspecified audience implies "everybody", but if the
      # feed's audience *is* specified, then it's not for everybody
      skip = true unless faudy.empty?
    else
      # if document audience matches feed non-audience, disqualify
      skip = true unless (faudn & audy).empty?

      # absence of an explicit feed audience implies "everybody"
      if faudy.empty?
        # if document audience minus feed non-audience has
        # members, re-qualify
        skip = false unless (audy - faudn).empty?
      else
        # if document audience matches feed audience, re-qualify
        skip = false unless (faudy & audy).empty?
      end
    end

    # if document non-audience matches feed audience, re-disqualify
    skip = true if !(audn.empty? || faudy.empty?) && !(faudy & audn).empty?

    next if skip

    canon = URI.parse(canonical_uri(uu).to_s)
    
    xml = { '#entry' => [
      { '#link' => nil, rel: :alternate, href: canon, type: 'text/html' },
      { '#id' => uu.to_s }
    ] }

    # get published date first
    published = (objects_for uu,
                 [RDF::Vocab::DC.issued, RDF::Vocab::DC.created],
                 datatype: RDF::XSD.dateTime)[0]
    
    # get latest updated date
    updated = (objects_for uu, RDF::Vocab::DC.modified,
               datatype: RDF::XSD.dateTime).sort[-1]
    updated ||= published || RDF::Literal::DateTime.new(DateTime.now)
    updated = Time.parse(updated.to_s).utc
    latest = updated if !latest or latest < updated

    xml['#entry'].push({ '#updated' => updated.iso8601 })

    if published
      published = Time.parse(published.to_s).utc
      xml['#entry'].push({ '#published' => published.iso8601 })
      dates[uu] = [published, updated]
    else
      dates[uu] = [updated, updated]
    end

    # get author(s)
    al = []
    authors_for(uu).each do |a|
      unless authors[a]
        n = label_for a
        x = authors[a] = { '#author' => [{ '#name' => n[1].to_s }] }

        hp = @graph.first_object [a, RDF::Vocab::FOAF.homepage, nil]
        hp ||= canonical_uri a
        
        x['#author'].push({ '#uri' => hp.to_s }) if hp
      end

      al.push authors[a]
    end

    xml['#entry'] += al unless al.empty?

    # get title (note unshift)
    if (t = label_for uu)
      titles[uu] = t[1].to_s
      xml['#entry'].unshift({ '#title' => t[1].to_s })
    else
      titles[uu] = uu.to_s
    end
    
    # get abstract
    if (d = label_for uu, desc: true)
      xml['#entry'].push({ '#summary' => d[1].to_s })
    end
    
    entries[uu] = xml
  end

  # note we overwrite the entries hash here with a sorted array
  entrycmp = -> a, b {
    # first we sort by published date
    p = dates[a][0] <=> dates[b][0]
    # if the published dates are the same, sort by updated date
    u = dates[a][1] <=> dates[b][1]
    # to break any ties, finally sort by title
    p == 0 ? u == 0 ? titles[a] <=> titles[b] : u : p }
  entries = entries.values_at(
    *entries.keys.sort { |a, b| entrycmp.call(a, b) })
  # ugggh god forgot the asterisk and lost an hour

  # now we punt out the doc

  preamble = [
    { '#id' => id.to_s },
    { '#updated' => latest.iso8601 },
    { '#generator' => 'RDF::SAK', version: RDF::SAK::VERSION,
      uri: "https://github.com/doriantaylor/rb-rdf-sak" },
    { nil => :link, rel: :self, type: 'application/atom+xml',
      href: canonical_uri(id) },
    { nil => :link, rel: :alternate, type: 'text/html',
      href: @base },
  ] + related.map do |r|
    { nil => :link, rel: :related, type: 'application/atom+xml',
     href: canonical_uri(r) }
  end

  if (t = label_for id)
    preamble.unshift({ '#title' => t[1].to_s })
  end

  if (r = @graph.first_literal [id, RDF::Vocab::DC.rights, nil])
    rh = { '#rights' => r.to_s, type: :text }
    rh['xml:lang'] = r.language if r.has_language?
    preamble.push rh
  end

  markup(spec: { '#feed' => preamble + entries,
    xmlns: 'http://www.w3.org/2005/Atom' }).document
end

#generate_audience_csv(file = nil, published: true) ⇒ Object 



959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
 
# File 'lib/rdf/sak.rb', line 959

def generate_audience_csv file = nil, published: true
  require 'csv'
  file = coerce_to_path_or_io file if file
  lab = {}

  out = all_internal_docs(published: published,
                          exclude: RDF::Vocab::FOAF.Image).map do |s|
    u = canonical_uri s
    x = struct_for s
    c = x[RDF::Vocab::DC.created] ? x[RDF::Vocab::DC.created][0] : nil
    _, t = label_for s, candidates: x
    _, d = label_for s, candidates: x, desc: true

    # # audience(s)
    # a = objects_for(s, RDF::Vocab::DC.audience).map do |au|
    #   next lab[au] if lab[au]
    #   _, al = label_for au
    #   lab[au] = al
    # end.map(&:to_s).sort.join '; '

    # # explicit non-audience(s)
    # n = objects_for(s, RDF::SAK::CI['non-audience']).map do |au|
    #   next lab[au] if lab[au]
    #   _, al = label_for au
    #   lab[au] = al
    # end.map(&:to_s).sort.join '; '

    # audience and non-audience
    a, n = [RDF::Vocab::DC.audience, CI['non-audience']].map do |ap|
      objects_for(s, ap).map do |au|
        next lab[au] if lab[au]
        _, al = label_for au
        lab[au] = al
      end.map(&:to_s).sort.join '; '
    end

    # concepts???
    concepts = [RDF::Vocab::DC.subject, CI.introduces,
                CI.assumes, CI.mentions].map do |pred|
      objects_for(s, pred, only: :resource).map do |o|
        con = self.objects_for(o, RDF.type).to_set & CONCEPTS
        next if con.empty?
        next lab[o] if lab[o]
        _, ol = label_for o
        lab[o] = ol
      end.compact.map(&:to_s).sort.join '; '
    end

    [s, u, c, t, d, a, n].map(&:to_s) + concepts
  end.sort { |a, b| a[2] <=> b[2] }

  out.unshift ['ID', 'URL', 'Created', 'Title', 'Description', 'Audience',
    'Non-Audience', 'Subject', 'Introduces', 'Assumes', 'Mentions']

  if file
    # don't open until now
    file = file.expand_path.open('wb') unless file.is_a? IO

    csv = CSV.new file
    out.each { |x| csv << x }
    file.flush
  end

  out
end

#generate_backlinks(subject, published: true, ignore: nil) ⇒ Object 



678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
 
# File 'lib/rdf/sak.rb', line 678

def generate_backlinks subject, published: true, ignore: nil
  uri    = canonical_uri(subject, rdf: false) || URI(uri_pp subject)
  ignore ||= Set.new
  raise 'ignore must be amenable to a set' unless ignore.respond_to? :to_set
  ignore = ignore.to_set
  nodes  = {}
  labels = {}
  types  = {}
  @graph.query([nil, nil, subject]).each do |stmt|
    next if ignore.include?(sj = stmt.subject)
    preds = nodes[sj] ||= Set.new
    preds << (pr = stmt.predicate)
    types[sj]  ||= asserted_types sj
    labels[sj] ||= label_for sj
    labels[pr] ||= label_for pr
  end

  # prune out 
  nodes.select! { |k, _| published? k } if published
  
  return if nodes.empty?

  li = nodes.sort do |a, b|
    cmp_label a[0], b[0], labels: labels
  end.map do |rsrc, preds|
    cu  = canonical_uri(rsrc, rdf: false) or next
    lab = labels[rsrc] || [nil, rsrc]
    lp  = abbreviate(lab[0]) if lab[0]
    ty  = abbreviate(types[rsrc]) if types[rsrc]
    
    { [{ [{ [lab[1].to_s] => :span, property: lp }] => :a,
      href: uri.route_to(cu), typeof: ty, rev: abbreviate(preds) }] => :li }
  end.compact

  { [{ li => :ul }] => :nav }
end

#generate_bibliography(id, published: true) ⇒ Object 



760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
 
# File 'lib/rdf/sak.rb', line 760

def generate_bibliography id, published: true
  id  = canonical_uuid id
  uri = canonical_uri id
  struct = struct_for id
  nodes     = Set[id] + smush_struct(struct)
  bodynodes = Set.new
  parts     = {}
  referents = {}
  labels    = { id => label_for(id, candidates: struct) }
  canon     = {}

  # uggh put these somewhere
  preds = {
    hp:    predicate_set(RDF::Vocab::DC.hasPart),
    sa:    predicate_set(RDF::RDFS.seeAlso),
    canon: predicate_set([RDF::OWL.sameAs, CI.canonical]),
    ref:   predicate_set(RDF::Vocab::DC.references),
    al:    predicate_set(RDF::Vocab::BIBO.contributorList),
    cont:  predicate_set(RDF::Vocab::DC.contributor),
  }

  # collect up all the parts (as in dct:hasPart)
  objects_for(id, preds[:hp], entail: false, only: :resource).each do |part|
    bodynodes << part

    # gather up all the possible alias urls this thing can have
    sa = ([part] + objects_for(part,
      preds[:sa], only: :uri, entail: false)).map do |x|
      [x] + subjects_for(preds[:canon], x, only: :uri, entail: false)
    end.flatten.uniq

    # collect all the referents
    reftmp = {}
    sa.each do |u|
      subjects_for preds[:ref], u, only: :uri, entail: false do |s, *p|
        reftmp[s] ||= Set.new
        reftmp[s] += p[0].to_set
      end
    end

    # if we are producing a list of references identified by only
    # published resources, prune out all the unpublished referents
    reftmp.select! { |x, _| published? x } if published

    # unconditionally skip this item if nothing references it
    next if reftmp.empty?

    referents[part] = reftmp

    reftmp.each do |r, _|
      labels[r] ||= label_for r
      canon[r]  ||= canonical_uri r
    end

    # collect all the authors and author lists

    objects_for(part, preds[:al], only: :resource, entail: false) do |o|
      RDF::List.new(subject: o, graph: @graph).each do |a|
        labels[a] ||= label_for a
      end
    end

    objects_for(part, preds[:cont], only: :uri, entail: false) do |a|
      labels[a] ||= label_for a
    end

    ps = struct_for part
    labels[part] = label_for part, candidates: ps
    nodes |= smush_struct ps

    parts[part] = ps
  end

  bmap = prepare_collation struct
  pf = -> x { abbreviate bmap[x.literal? ? :literals : :resources][x] }

  body = []
  parts.sort { |a, b| cmp_label a[0], b[0], labels: labels }.each do |k, v|
    mapping = prepare_collation v
    p = -> x {
      abbreviate mapping[x.literal? ? :literals : :resources][x] }
    t = abbreviate mapping[:types]

    lp = label_for k, candidates: v
    h2c = [lp[1].to_s]
    h2  = { h2c => :h2 }
    cu  = canonical_uri k
    rel = nil
    unless cu.scheme.downcase.start_with? 'http'
      if sa = v[RDF::RDFS.seeAlso]
        rel = p.call sa[0]
        cu = canonical_uri sa[0]
      else
        cu = nil
      end
    end

    if cu
      h2c[0] = { [lp[1].to_s] => :a, rel: rel,
        property: p.call(lp[1]), href: cu.to_s }
    else
      h2[:property] = p.call(lp[1])  
    end

    # authors &c
    # authors contributors editors translators
    al = []
    AUTHOR_SPEC.each do |label, pl|
      dd = []
      seen = Set.new
      pl.each do |pred|
        # first check if the struct has the predicate
        next unless v[pred]
        li = []
        ul = { li => :ul, rel: abbreviate(pred) }
        v[pred].sort { |a, b| cmp_label a, b, labels: labels }.each do |o|
          # check if this is a list
          tl = RDF::List.new subject: o, graph: @graph
          if tl.empty? and !seen.include? o
            seen << o
            lab = labels[o] ? { [labels[o][1]] => :span,
              property: abbreviate(labels[o][0]) } : o
            li << { [lab] => :li, resource: o }
          else
            # XXX this will actually not be right if there are
            # multiple lists but FINE FOR NOW
            ul[:inlist] ||= ''
            tl.each do |a|
              seen << a 
              lab = labels[a] ? { [labels[a][1]] => :span,
                property: abbreviate(labels[a][0]) } : a
              li << { [lab] => :li, resource: a }
            end
          end
        end
        dd << ul unless li.empty?
      end
      al += [{ [label] => :dt }, { dd => :dd }] unless dd.empty?
    end

    # ref list
    rl = referents[k].sort do |a, b|
      cmp_label a[0], b[0], labels: labels
    end.map do |ref, pset|
      lab = labels[ref] ? { [labels[ref][1]] => :span,
        property: abbreviate(labels[ref][0]) } : ref
              
      { [{ [lab] => :a, rev: abbreviate(pset), href: canon[ref] }] => :li }
    end

    contents = [h2, {
      al + [{ ['Referenced in:'] => :dt },
        { [{ rl => :ul }] => :dd }] => :dl }]

    body << { contents => :section,
      rel: pf.call(k), resource: k.to_s, typeof: t }
  end

  # prepend abstract to body if it exists
  abs = label_for id, candidates: struct, desc: true
  if abs
    tag = { '#p' => abs[1], property: abbreviate(abs[0]) }
    body.unshift tag
  end

  # add labels to nodes
  nodes += smush_struct labels

  # get prefixes
  pfx = prefix_subset prefixes, nodes

  # get title tag
  title = title_tag labels[id][0], labels[id][1],
    prefixes: prefixes, lang: 'en'

  # get links
  link = head_links id,
    struct: struct, ignore: bodynodes, labels: labels, vocab: XHV

  # get metas
  mn = {}
  mn[abs[1]] = :description if abs
  mi = Set.new
  mi << labels[id][1] if labels[id]
  meta = head_meta id,
    struct: struct, lang: 'en', ignore: mi, meta_names: mn, vocab: XHV

  meta += generate_twitter_meta(id) || []

  xhtml_stub(base: uri, prefix: pfx, lang: 'en', title: title, vocab: XHV,
    link: link, meta: meta, transform: @config[:transform],
    body: { body => :body, about: '',
      typeof: abbreviate(struct[RDF::RDFV.type] || []) }).document
end

#generate_gone_map(published: false, docs: nil) ⇒ Object 



1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
 
# File 'lib/rdf/sak.rb', line 1483

def generate_gone_map published: false, docs: nil
  # published is a no-op for this one because these docs are by
  # definition not published
  docs ||= reachable published: false
  p    = RDF::Vocab::BIBO.status
  base = URI(@base.to_s)
  out  = {}
  docs.select { |s|
    @graph.has_statement? RDF::Statement(s, p, CI.retired) }.each do |doc|
    canon = canonical_uri doc, rdf: false
    next unless base.route_to(canon).relative?
    canon = canon.request_uri.delete_prefix '/'
    # value of the gone map doesn't matter
    out[canon] = canon
  end

  out
end

#generate_reading_list(subject, published: true) ⇒ Object 



1560
1561
1562
1563
1564
1565
1566
1567
1568
 
# File 'lib/rdf/sak.rb', line 1560

def generate_reading_list subject, published: true
  # struct = struct_for subject

  # find all the books, sort them by title

  # for each book, give title, authors, inbound references

  # punt out xhtml
end

#generate_redirect_map(published: false, docs: nil) ⇒ Object

you know what, it’s entirely possible that these ought never be called individually and the work to get one would duplicate the work of getting the other, so maybe just do ‘em both at once



1478
1479
1480
1481
 
# File 'lib/rdf/sak.rb', line 1478

def generate_redirect_map published: false, docs: nil
  generate_uuid_redirect_map(published: published, docs: docs).merge(
    generate_slug_redirect_map(published: published, docs: docs))
end

#generate_rewrite_map(published: false, docs: nil) ⇒ Object

generate rewrite map(s)



1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
 
# File 'lib/rdf/sak.rb', line 1372

def generate_rewrite_map published: false, docs: nil
  docs ||= reachable published: published
  base = URI(@base.to_s)
  rwm  = {}
  docs.each do |doc|
    tu = URI(doc.to_s)
    cu = canonical_uri doc, rdf: false
    next unless tu.respond_to?(:uuid) and cu.respond_to?(:request_uri)

    # skip external links obvs
    next unless base.route_to(cu).relative?

    # skip /uuid form
    cp = cu.request_uri.delete_prefix '/'
    next if cu.host == base.host and tu.uuid == cp
    
    rwm[cp] = tu.uuid
  end

  rwm
end

#generate_sitemap(published: true) ⇒ Object 



1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
 
# File 'lib/rdf/sak.rb', line 1126

def generate_sitemap published: true
  urls = {}

  # do feeds separately
  feeds = all_of_type RDF::Vocab::DCAT.Distribution
  #feeds.select! { |f| published? f } if published
  feeds.each do |f|
    uri = canonical_uri(f)
    f = generate_atom_feed f, published: published, related: feeds
    mt = f.at_xpath('/atom:feed/atom:updated[1]/text()',
      { atom: 'http://www.w3.org/2005/Atom' })
    urls[uri] = { [{ [uri.to_s] => :loc }, { [mt] => :lastmod }] => :url }
  end

  # build up hash of urls
  all_internal_docs(published: published).each do |doc|
    next if asserted_types(doc).include? RDF::Vocab::FOAF.Image
    uri  = canonical_uri(doc)
    next unless uri.authority && @base && uri.authority == base.authority
    mods = objects_for(doc, [RDF::Vocab::DC.created,
      RDF::Vocab::DC.modified, RDF::Vocab::DC.issued],
      datatype: RDF::XSD.dateTime).sort
    nodes = [{ [uri.to_s] => :loc }]
    nodes << { [mods[-1].to_s] => :lastmod } unless mods.empty?
    urls[uri] = { nodes => :url }
  end

  urls = urls.sort.map { |_, v| v }

  markup(spec: { urls => :urlset,
    xmlns: 'http://www.sitemaps.org/schemas/sitemap/0.9' }).document
end

#generate_slug_redirect_map(published: false, docs: nil) ⇒ Object

find all URIs/slugs that are not canonical, map them to slugs that are canonical



1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
 
# File 'lib/rdf/sak.rb', line 1423

def generate_slug_redirect_map published: false, docs: nil
  docs ||= reachable published: published
  base = URI(@base.to_s)

  # for redirects we collect all the docs, plus all their URIs,
  # separate canonical from the rest

  # actually an easy way to do this is just harvest all the
  # multi-addressed docs, remove the first one, then ask for the
  # canonical uuid back,

  fwd = {}
  rev = {}
  out = {}

  docs.each do |doc|
    uris  = canonical_uri doc, unique: false, rdf: false
    canon = uris.shift
    next unless canon.respond_to? :request_uri

    # cache the forward direction
    fwd[doc] = canon

    unless uris.empty?
      uris.each do |uri|
        next unless uri.respond_to? :request_uri
        next if canon == uri
        next unless base.route_to(uri).relative?

        # warn "#{canon} <=> #{uri}"

        requri = uri.request_uri.delete_prefix '/'
        next if requri == '' ||
          requri =~ /^[0-9a-f]{8}(?:-[0-9a-f]{4}){4}[0-9a-f]{8}$/

        # cache the reverse direction
        rev[uri] = requri
      end
    end
  end

  rev.each do |uri, requri|
    if (doc = canonical_uuid(uri, published: published)) and
        fwd[doc] and fwd[doc] != uri
      out[requri] = fwd[doc].to_s
    end
  end

  out
end

#generate_stats(published: true) ⇒ Object 



1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
 
# File 'lib/rdf/sak.rb', line 1585

def generate_stats published: true
  out = {}
  all_of_type(QB.DataSet).map do |s|
    base  = canonical_uri s, rdf: false
    types = abbreviate asserted_types(s)
    title = if t = label_for(s)
              [t[1].to_s, abbreviate(t[0])]
            end
    cache = {}
    subjects_for(QB.dataSet, s, only: :resource).each do |o|
      if d = objects_for(o, CI.document, only: :resource).first
        if !published or published?(d)
          # include a "sort" time that defaults to epoch zero
          c = cache[o] ||= {
            doc: d, stime: Time.at(0).getgm, struct: struct_for(o) }

          if t = label_for(d)
            c[:title] = t
          end
          if a = label_for(d, desc: true)
            c[:abstract] = a
          end
          if ct = objects_for(d,
            RDF::Vocab::DC.created, datatype: RDF::XSD.dateTime).first
            c[:stime] = c[:ctime] = ct.object.to_time.getgm
          end
          if mt = objects_for(d,
            RDF::Vocab::DC.modified, datatype:RDF::XSD.dateTime)
            c[:mtime] = mt.map { |m| m.object.to_time.getgm }.sort
            c[:stime] = c[:mtime].last unless mt.empty?
          end
        end
      end
    end

    # sort lambda closure
    sl = -> a, b do
      x = cache[b][:stime] <=> cache[a][:stime]
      return x unless x == 0
      x = cache[b][:ctime] <=> cache[a][:ctime]
      return x unless x == 0
      ta = cache[a][:title] || Array.new(2, cache[a][:uri])
      tb = cache[b][:title] || Array.new(2, cache[b][:uri])
      ta[1].to_s <=> tb[1].to_s
    end

    rows = []
    cache.keys.sort(&sl).each do |k|
      c = cache[k]
      href = base.route_to canonical_uri(c[:doc], rdf: false)
      dt = abbreviate asserted_types(c[:doc])
      uu = URI(k.to_s).uuid
      nc = UUID::NCName.to_ncname uu, version: 1
      tp, tt = c[:title] || []
      ab = if c[:abstract]
             { [c[:abstract][1].to_s] => :th, about: href,
              property: abbreviate(c[:abstract].first) }
           else
             { [] => :th }
           end
      
      td = [{ { { [tt.to_s] => :span, property: abbreviate(tp) } => :a,
        rel: 'ci:document', href: href } => :th },
        ab,
        { [c[:ctime].iso8601] => :th, property: 'dct:created',
         datatype: 'xsd:dateTime', about: href, typeof: dt },
        { c[:mtime].reverse.map { |m| { [m.iso8601] => :span,
           property: 'dct:modified', datatype: 'xsd:dateTime' } } => :th,
          about: href
        },
      ] + DSD_SEQ.map do |f|
        h = []
        x = { h => :td }
        p = CI[f]
        if y = c[:struct][p] and !y.empty?
          h << y = y.first
          x[:property] = abbreviate p
          x[:datatype] = abbreviate y.datatype if y.datatype?
        end
        x
      end
      rows << { td => :tr, id: nc, about: "##{nc}",
        typeof: 'qb:Observation' }
    end

    out[s] = xhtml_stub(base: base, title: title,
      transform: config[:transform], attr: { about: '', typeof: types },
      prefix: prefixes, content: {
        [{ [{ [{ ['About'] => :th, colspan: 4 },
            { ['Counts'] => :th, colspan: 4 },
            { ['Words per Block'] => :th, colspan: 7 }] => :tr },
          { TH_SEQ => :tr } ] => :thead },
         { rows => :tbody, rev: 'qb:dataSet' }] => :table }).document
  end

  out
end

#generate_twitter_meta(subject) ⇒ Object 



715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
 
# File 'lib/rdf/sak.rb', line 715

def generate_twitter_meta subject
  # get author
  author = authors_for(subject, unique: true) or return

  # get author's twitter account
  twitter = objects_for(author, RDF::Vocab::FOAF.,
    only: :resource).select { |t| t.to_s =~ /twitter\.com/
  }.sort.first or return
  twitter = URI(twitter.to_s).path.split(/\/+/)[1]
  twitter = ?@ + twitter unless twitter.start_with? ?@

  # get title
  title = label_for(subject) or return

  out = [
    { nil => :meta, name: 'twitter:card', content: :summary },
    { nil => :meta, name: 'twitter:site', content: twitter },
    { nil => :meta, name: 'twitter:title', content: title[1].to_s }
  ]

  # get abstract
  if desc = label_for(subject, desc: true)
    out.push({ nil => :meta, name: 'twitter:description',
      content: desc[1].to_s })
  end

  # get image (foaf:depiction)
  img = objects_for(subject, RDF::Vocab::FOAF.depiction, only: :resource)
  unless img.empty?
    img = img[0].to_s
    out.push({ nil => :meta, name: 'twitter:image', content: img })
    out[0][:content] = :summary_large_image
  end

  # return the appropriate xml-mixup structure
  out
end

#generate_uuid_redirect_map(published: false, docs: nil) ⇒ Object

give me all UUIDs of all documents, filter for published if applicable

find the “best” (relative) URL for the UUID and map the pair together



1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
 
# File 'lib/rdf/sak.rb', line 1399

def generate_uuid_redirect_map published: false, docs: nil
  docs ||= reachable published: published

  base = URI(@base.to_s)

  # keys are /uuid, values are 
  out = {}
  docs.each do |doc|
    tu = URI(doc.to_s)
    cu = canonical_uri doc, rdf: false
    next unless tu.respond_to?(:uuid) and cu.respond_to?(:request_uri)

    # skip /uuid form
    cp = cu.request_uri.delete_prefix '/'
    next if cu.host == base.host && tu.uuid == cp

    # all redirect links are absolute
    out[tu.uuid] = cu.to_s
  end
  out
end

#head_links(subject, struct: nil, nodes: nil, prefixes: {}, ignore: [], uris: {}, labels: {}, vocab: nil) ⇒ Object

generate indexes of books, not-books, and other external links



567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
 
# File 'lib/rdf/sak.rb', line 567

def head_links subject, struct: nil, nodes: nil, prefixes: {},
    ignore: [], uris: {}, labels: {}, vocab: nil

  raise 'ignore must be Array or Set' unless
    [Array, Set].any? { |c| ignore.is_a? c }

  struct ||= struct_for subject
  nodes  ||= invert_struct struct

  # make sure these are actually URI objects not RDF::URI
  uris = uris.transform_values { |v| URI(uri_pp v.to_s) }
  uri  = uris[subject] || canonical_uri(subject, rdf: false)

  ignore = ignore.to_set

  # output
  links = []

  nodes.reject { |n, _| ignore.include?(n) || !n.uri? }.each do |k, v|
    # first nuke rdf:type, that's never in there
    v = v.dup.delete RDF::RDFV.type
    next if v.empty?

    unless uris[k]
      cu = canonical_uri k
      uris[k] = cu || uri_pp(k.to_s)
    end

    # munge the url and make the tag
    rel = abbreviate v.to_a, vocab: vocab
    ru  = uri.route_to(uris[k])
    ln  = { nil => :link, rel: rel, href: ru.to_s }

    # add the title
    if lab = labels[k]
      ln[:title] = lab[1].to_s
    end

    # add type attribute
    unless (mts = formats_for k).empty?
      ln[:type] = mts.first.to_s

      if ln[:type] =~ /(java|ecma)script/i ||
          !(v.to_set & Set[RDF::Vocab::DC.requires]).empty?
        ln[:src] = ln.delete :href
        # make sure we pass in an empty string so there is a closing tag
        ln.delete nil
        ln[['']] = :script
      end
    end

    # finally add the link
    links.push ln
  end

  links.sort! do |a, b|
    # sort by rel, then by href
    # warn a.inspect, b.inspect
    s = 0
    [nil, :rel, :rev, :href, :title].each do |k|
      s = a.fetch(k, '').to_s <=> b.fetch(k, '').to_s
      break if s != 0
    end
    s
  end

  links
end

#head_meta(subject, struct: nil, nodes: nil, prefixes: {}, ignore: [], meta_names: {}, vocab: nil, lang: nil, xhtml: true) ⇒ Object 



636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
 
# File 'lib/rdf/sak.rb', line 636

def head_meta subject, struct: nil, nodes: nil, prefixes: {},
    ignore: [], meta_names: {}, vocab: nil, lang: nil, xhtml: true

  raise 'ignore must be Array or Set' unless
    [Array, Set].any? { |c| ignore.is_a? c }

  struct ||= struct_for subject
  nodes  ||= invert_struct struct

  ignore = ignore.to_set

  meta = []
  nodes.select { |n| n.literal? && !ignore.include?(n) }.each do |k, v|
    rel  = abbreviate v.to_a, vocab: vocab
    tag  = { nil => :meta, property: rel, content: k.to_s }

    lang = (k.language? && k.language != lang ? k.language : nil) ||
      (k.datatype == RDF::XSD.string && lang ? '' : nil)
    if lang
      tag['xml:lang'] = lang if xhtml
      tag[:lang] = lang
    end

    tag[:datatype] = abbreviate k.datatype, vocab: XHV if k.datatype?
    tag[:name] = meta_names[k] if meta_names[k]

    meta << tag
  end

  meta.sort! do |a, b|
    s = 0
    [:about, :property, :datatype, :content, :name].each do |k|
      # warn a.inspect, b.inspect
      s = a.fetch(k, '').to_s <=> b.fetch(k, '').to_s
      break if s != 0
    end
    s
  end

  meta
end

#ingest_csv(file) ⇒ Object 



1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
 
# File 'lib/rdf/sak.rb', line 1034

def ingest_csv file
  file = coerce_to_path_or_io file

  require 'csv'

  # key mapper
  km = { uuid: :id, url: :uri }
  kt = -> (k) { km[k] || k }

  # grab all the concepts and audiences

  audiences = {}
  all_of_type(CI.Audience).map do |c|
    s = struct_for c

    # homogenize the labels
    lab = [false, true].map do |b|
      label_for(c, candidates: s, unique: false, alt: b).map { |x| x[1] }
    end.flatten.map { |x| x.to_s.strip.downcase }

    # we want all the keys to share the same set
    set = nil
    lab.each { |t| set = audiences[t] ||= set || Set.new }
    set << c
  end

  concepts = {}
  all_of_type(RDF::Vocab::SKOS.Concept).map do |c|
    s = struct_for c

    # homogenize the labels
    lab = [false, true].map do |b|
      label_for(c, candidates: s, unique: false, alt: b).map { |x| x[1] }
    end.flatten.map { |x| x.to_s.strip.downcase }

    # we want all the keys to share the same set
    set = nil
    lab.each { |t| set = concepts[t] ||= set || Set.new }
    set << c
  end

  data = CSV.read(file, headers: true,
                  header_converters: :symbol).map do |o|
    o = o.to_h.transform_keys(&kt)
    s = canonical_uuid(o.delete :id) or next

    # LOLOL wtf

    # handle audience
    [:audience, :nonaudience].each do |a|
      if o[a]
        o[a] = o[a].strip.split(/\s*[;,]+\s*/, -1).map do |t|
          if t =~ /^[a-z+-]+:[^[:space:]]+$/
            u = RDF::URI(t)
            canonical_uuid(u) || u
          elsif audiences[t.downcase]
            audiences[t.downcase].to_a
          end
        end.flatten.compact.uniq
      else
        o[a] = []
      end
    end

    # handle concepts
    [:subject, :introduces, :assumes, :mentions].each do |a|
      if o[a]
        o[a] = o[a].strip.split(/\s*[;,]+\s*/, -1).map do |t|
          if t =~ /^[a-z+-]+:[^[:space:]]+$/
            u = RDF::URI(t)
            canonical_uuid(u) || u
          elsif concepts[t.downcase]
            concepts[t.downcase].to_a
          end
        end.flatten.compact.uniq
      else
        o[a] = []
      end

    end

    CSV_PRED.each do |sym, pred|
      o[sym].each do |obj|
        @graph << [s, pred, obj]
      end
    end

    [s, o]
  end.compact.to_h
  data
end

#label_for(subject, candidates: nil, unique: true, type: nil, lang: nil, desc: false, alt: false) ⇒ Array

Obtain the most appropriate label(s) for the subject’s type(s). Returns one or more (depending on the ‘unique` flag) predicate-object pairs in order of preference.

Parameters:

  • subject (RDF::Resource)
  • unique (true, false) (defaults to: true) —

    only return the first pair

  • type (RDF::Term, Array) (defaults to: nil) —

    supply asserted types if already retrieved

  • lang (nil) (defaults to: nil) —

    not currently implemented (will be conneg)

  • desc (false, true) (defaults to: false) —

    retrieve description instead of label

  • alt (false, true) (defaults to: false) —

    retrieve alternate instead of main

Returns:

  • (Array) —

    either a predicate-object pair or an array of pairs.



440
441
442
443
444
 
# File 'lib/rdf/sak.rb', line 440

def label_for subject, candidates: nil, unique: true, type: nil,
    lang: nil, desc: false, alt: false
  Util.label_for @graph, subject, candidates: candidates,
    unique: unique, type: type, lang: lang, desc: desc, alt: alt
end

#locate(uri) ⇒ Pathname

Locate the file in the source directory associated with the given URI.

Parameters:

  • the (RDF::URI, URI, :to_s) —

    URI requested

Returns:

  • (Pathname) —

    of the corresponding file or nil if no file was found.



1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
 
# File 'lib/rdf/sak.rb', line 1703

def locate uri
  uri = coerce_resource uri

  base = URI(@base.to_s)

  tu = URI(uri) # copy of uri for testing content
  unless tu.scheme == 'urn' and tu.nid == 'uuid'
    raise "could not find UUID for #{uri}" unless uuid = canonical_uuid(uri)
    tu = URI(uri = uuid)
  end

  # xxx bail if the uri isn't a subject in the graph

  candidates = [@config[:source] + tu.uuid]

  # try all canonical URIs
  (canonical_uri uri, unique: false, slugs: true).each do |u|
    u = URI(u.to_s)
    next unless u.hostname == base.hostname
    p = URI.unescape u.path[/^\/*(.*?)$/, 1]
    candidates.push(@config[:source] + p)
  end

  # warn candidates

  files = candidates.uniq.map do |c|
    Pathname.glob(c.to_s + '{,.*,/index{,.*}}')
  end.reduce(:+).reject do |x|
    x.directory? or RDF::SAK::MimeMagic.by_path(x).to_s !~ 
      /.*(?:markdown|(?:x?ht|x)ml).*/i
  end.uniq

  #warn files

  # XXX implement negotiation algorithm
  return files[0]

  # return the filename from the source
  # nil
end

#map_location(type) ⇒ Object

private?



1504
1505
1506
1507
1508
1509
1510
 
# File 'lib/rdf/sak.rb', line 1504

def map_location type
  # find file name in config
  fn = @config[:maps][type] or return

  # concatenate to target directory
  @config[:target] + fn
end

#objects_for(subject, predicate, entail: true, only: [], datatype: nil) ⇒ RDF::Term

Returns objects from the graph with entailment.

Parameters:

  • subject
  • predicate
  • entail (defaults to: true)
  • only (defaults to: [])
  • datatype (defaults to: nil)

Returns:

  • (RDF::Term) 


370
371
372
373
 
# File 'lib/rdf/sak.rb', line 370

def objects_for subject, predicate, entail: true, only: [], datatype: nil
  Util.objects_for @graph, subject, predicate,
    entail: entail, only: only, datatype: datatype
end

#prefixes ⇒ Hash

Get the prefix mappings from the configuration.

Returns:

  • (Hash) 


247
248
249
 
# File 'lib/rdf/sak.rb', line 247

def prefixes
  @config[:prefixes] || {}
end

#published?(uri, circulated: false) ⇒ true, false

Determine whether the URI represents a published document.

Parameters:

  • uri

Returns:

  • (true, false) 


1808
1809
1810
1811
 
# File 'lib/rdf/sak.rb', line 1808

def published? uri, circulated: false
  RDF::SAK::Util.published? @graph, uri,
    circulated: circulated, base: @base
end

#reachable(published: false) ⇒ Object

Get all “reachable” UUID-identified entities (subjects which are also objects)



525
526
527
528
529
530
531
 
# File 'lib/rdf/sak.rb', line 525

def reachable published: false
  p = published ? -> x { published?(x) } : -> x { true }
  # now get the subjects which are also objects
  @graph.subjects.select do |s|
    s.uri? && s =~ /^urn:uuid:/ && @graph.has_object?(s) && p.call(s)
  end
end

#reading_lists(published: true) ⇒ Object

whoops lol we forgot the book list



1554
1555
1556
1557
1558
 
# File 'lib/rdf/sak.rb', line 1554

def reading_lists published: true
  out = all_of_type RDF::Vocab::SiocTypes.ReadingList
  return out unless published
  out.select { |r| published? r }
end

#replacements_for(subject, published: true) ⇒ Set

Find the terminal replacements for the given subject, if any exist.

Parameters:

  • subject
  • published (defaults to: true) —

    indicate the context is published

Returns:

  • (Set) 


382
383
384
 
# File 'lib/rdf/sak.rb', line 382

def replacements_for subject, published: true
  Util.replacements_for @graph, subject, published: published
end

#resolve_documents ⇒ Object

resolve documents from source



1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
 
# File 'lib/rdf/sak.rb', line 1758

def resolve_documents
  src = @config[:source]
  out = []
  src.find do |f|
    Find.prune if f.basename.to_s[0] == ?.
    next if f.directory?
    out << f
  end

  out
end

#resolve_file(path) ⇒ Object 



1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
 
# File 'lib/rdf/sak.rb', line 1770

def resolve_file path
  return unless path.file?
  path = Pathname('/') + path.relative_path_from(@config[:source])
  base = URI(@base.to_s)
  uri  = base + path.to_s

  #warn "trying #{uri}"

  until (out = canonical_uuid uri)
    # iteratively strip off
    break if uri.path.end_with? '/'

    dn = path.dirname
    bn = path.basename '.*'

    # try index first
    if bn.to_s == 'index'
      p = dn.to_s
      p << '/' unless p.end_with? '/'
      uri = base + p
    elsif bn == path.basename
      break
    else
      path = dn + bn
      uri = base + path.to_s
    end

    # warn "trying #{uri}"
  end

  out
end

#struct_for(subject, rev: false, only: [], uuids: false, canon: false) ⇒ Hash

Obtain a key-value structure for the given subject, optionally constraining the result by node type (:resource, :uri/:iri, :blank/:bnode, :literal)

Parameters:

  • subject —

    of the inquiry

  • rev (defaults to: false) —

    map in reverse

  • only (defaults to: []) —

    one or more node types

  • uuids (defaults to: false) —

    coerce resources to if possible

Returns:

  • (Hash) 


276
277
278
279
 
# File 'lib/rdf/sak.rb', line 276

def struct_for subject, rev: false, only: [], uuids: false, canon: false
  Util.struct_for @graph, subject,
    rev: rev, only: only, uuids: uuids, canon: canon
end

#sub_concepts(concept, extra: []) ⇒ Object 



470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
 
# File 'lib/rdf/sak.rb', line 470

def sub_concepts concept, extra: []
  raise 'Concept must be exactly one concept' unless
    concept.is_a? RDF::Resource
  extra = term_list extra

  # we need an array for a queue, and a set to accumulate the
  # output as well as a separate 'seen' set
  queue = [concept]
  seen  = Set.new queue.dup
  out   = seen.dup

  # it turns out that the main SKOS hierarchy terms, while not
  # being transitive themselves, are subproperties of transitive
  # relations which means they are as good as being transitive.

  while c = queue.shift
    SKOS_HIER.each do |struct|
      elem, pat, preds = struct.values_at(:element, :pattern, :preds)
      preds.each do |p|
        @graph.query(pat.call c, p).each do |stmt|
          # obtain hierarchical element
          hierc = stmt.send elem

          # skip any further processing if we have seen this concept
          next if seen.include? hierc
          seen << hierc

          next if !extra.empty? and !extra.any? do |t|
            @graph.has_statement? RDF::Statement.new(hierc, RDF.type, t)
          end

          queue << hierc
          out   << hierc
        end
      end
    end
  end

  out.to_a.sort
end

#subjects_for(predicate, object, entail: true, only: []) ⇒ RDF::Resource

Returns subjects from the graph with entailment.

Parameters:

  • predicate
  • object
  • entail (defaults to: true)
  • only (defaults to: [])

Returns:

  • (RDF::Resource) 


356
357
358
 
# File 'lib/rdf/sak.rb', line 356

def subjects_for predicate, object, entail: true, only: []
  Util.subjects_for @graph, predicate, object, entail: entail, only: only
end

#target_for(uri, published: false) ⇒ Pathname

Find a destination pathname for the document

Parameters:

  • uri
  • published (defaults to: false)

Returns:

  • (Pathname) 


1819
1820
1821
1822
1823
1824
1825
1826
 
# File 'lib/rdf/sak.rb', line 1819

def target_for uri, published: false
  uri = coerce_resource uri
  uri = canonical_uuid uri
  target = @config[published?(uri) && published ? :target : :private]

  # target is a pathname so this makes a pathname
  target + "#{URI(uri.to_s).uuid}.xml"
end

#visit(uri) ⇒ RDF::SAK::Context::Document

Visit (open) the document at the given URI.

Parameters:

  • uri (RDF::URI, URI, :to_s)

Returns:



1750
1751
1752
1753
1754
1755
 
# File 'lib/rdf/sak.rb', line 1750

def visit uri
  uri  = canonical_uuid uri
  path = locate uri
  return unless path
  Document.new self, uri, uri: canonical_uri(uri), doc: path
end

#write_feeds(type: RDF::Vocab::DCAT.Distribution, published: true) ⇒ Object 



1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
 
# File 'lib/rdf/sak.rb', line 1357

def write_feeds type: RDF::Vocab::DCAT.Distribution, published: true
  feeds = all_of_type type
  target = @config[published ? :target : :private]
  feeds.each do |feed|
    tu  = URI(feed.to_s)
    doc = generate_atom_feed feed, published: published, related: feeds
    fh  = (target + "#{tu.uuid}.xml").open('w')
    doc.write_to fh
    fh.close
  end
end

#write_gone_map(published: false, docs: nil) ⇒ Object 



1535
1536
1537
1538
1539
 
# File 'lib/rdf/sak.rb', line 1535

def write_gone_map published: false, docs: nil
  data = generate_gone_map published: published, docs: docs
  loc  = map_location :gone
  write_map_file loc, data
end

#write_map_file(location, data) ⇒ Object

private?



1514
1515
1516
1517
1518
1519
 
# File 'lib/rdf/sak.rb', line 1514

def write_map_file location, data
  # open file
  fh = File.new location, 'w'
  data.sort.each { |k, v| fh.write "#{k}\t#{v}\n" }
  fh.close # return value is return value from close
end

#write_maps(published: true, docs: nil) ⇒ Object 



1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
 
# File 'lib/rdf/sak.rb', line 1541

def write_maps published: true, docs: nil
  docs ||= reachable published: false
  # slug to uuid (internal)
  write_rewrite_map docs: docs
  # uuid/slug to canonical slug (308)
  write_redirect_map docs: docs
  # retired slugs/uuids (410)
  write_gone_map docs: docs
  true
end

#write_reading_lists(published: true) ⇒ Object 



1570
1571
1572
1573
1574
1575
1576
1577
1578
 
# File 'lib/rdf/sak.rb', line 1570

def write_reading_lists published: true
  reading_lists(published: published).each do |rl|
    tu  = URI(rl.to_s)
    doc = generate_reading_list rl, published: published
    fh  = (target + "#{tu.uuid}.xml").open('w')
    doc.write_to fh
    fh.close
  end
end

#write_redirect_map(published: false, docs: nil) ⇒ Object 



1529
1530
1531
1532
1533
 
# File 'lib/rdf/sak.rb', line 1529

def write_redirect_map published: false, docs: nil
  data = generate_redirect_map published: published, docs: docs
  loc  = map_location :redirect
  write_map_file loc, data
end

#write_rewrite_map(published: false, docs: nil) ⇒ Object

public again



1523
1524
1525
1526
1527
 
# File 'lib/rdf/sak.rb', line 1523

def write_rewrite_map published: false, docs: nil
  data = generate_rewrite_map published: published, docs: docs
  loc  = map_location :rewrite
  write_map_file loc, data
end

#write_sitemap(published: true) ⇒ Object 



1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
 
# File 'lib/rdf/sak.rb', line 1159

def write_sitemap published: true
  sitemap = generate_sitemap published: published
  file    = @config[:sitemap] || '.well-known/sitemap.xml'
  target  = @config[published ? :target : :private]
  target.mkpath unless target.directory?

  fh = (target + file).open(?w)
  sitemap.write_to fh
  fh.close
end

#write_stats(published: true) ⇒ Object 



1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
 
# File 'lib/rdf/sak.rb', line 1683

def write_stats published: true
  target = @config[published ? :target : :private]
  target.mkpath unless target.directory?
  generate_stats(published: published).each do |uu, doc|
    bn = URI(uu.to_s).uuid + '.xml'
    fh = (target + bn).open ?w
    doc.write_to fh
    fh.flush
    fh.close
  end
end

#write_xhtml(published: true) ⇒ Object

write public and private variants to target



1834
1835
 
# File 'lib/rdf/sak.rb', line 1834

def write_xhtml published: true
end