Module: Bolognese::Readers::SchemaOrgReader

Included in:
MetadataUtils
Defined in:
lib/bolognese/readers/schema_org_reader.rb

Constant Summary collapse

SO_TO_DC_RELATION_TYPES =
{
  "citation" => "References",
  "sameAs" => "IsIdenticalTo",
  "isPartOf" => "IsPartOf",
  "hasPart" => "HasPart",
  "isPredecessor" => "IsPreviousVersionOf",
  "isSuccessor" => "IsNewVersionOf"
}

Instance Method Summary collapse

Instance Method Details

#get_schema_org(id: nil, **options) ⇒ Object



15
16
17
18
19
20
21
22
23
24
25
26
27
28
# File 'lib/bolognese/readers/schema_org_reader.rb', line 15

def get_schema_org(id: nil, **options)
  return { "string" => nil, "state" => "not_found" } unless id.present?

  id = normalize_id(id)
  response = Maremma.get(id)
  doc = Nokogiri::XML(response.body.fetch("data", nil), nil, 'UTF-8')

  # workaround for xhtml documents
  nodeset = doc.css("script")
  string = nodeset.find { |element| element["type"] == "application/ld+json" }
  string = string.text if string.present?

  { "string" => string }
end

#read_schema_org(string: nil, **options) ⇒ Object



30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
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
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
169
170
171
172
# File 'lib/bolognese/readers/schema_org_reader.rb', line 30

def read_schema_org(string: nil, **options)
  if string.present?
    errors = jsonlint(string)
    return { "errors" => errors } if errors.present?
  end

  read_options = ActiveSupport::HashWithIndifferentAccess.new(options.except(:doi, :id, :url, :sandbox, :validate))

  meta = string.present? ? Maremma.from_json(string) : {}

  identifiers = ([options[:doi] || meta.fetch("@id", nil)] + Array.wrap(meta.fetch("identifier", nil))).map do |r|
    r = normalize_id(r) if r.is_a?(String)
    if r.is_a?(String) && r.start_with?("https://doi.org")
      { "identifierType" => "DOI", "identifier" => r }
    elsif r.is_a?(String)
        { "identifierType" => "URL", "identifier" => r }
    elsif r.is_a?(Hash)
      { "identifierType" => get_identifier_type(r["propertyID"]), "identifier" => r["value"] }
    end
  end.compact.uniq

  id = Array.wrap(identifiers).first.to_h.fetch("identifier", nil)

  schema_org = meta.fetch("@type", nil) && meta.fetch("@type").camelcase
  resource_type_general = Bolognese::Utils::SO_TO_DC_TRANSLATIONS[schema_org]
  types = {
    "resourceTypeGeneral" => resource_type_general,
    "resourceType" => meta.fetch("additionalType", nil),
    "schemaOrg" => schema_org,
    "citeproc" => Bolognese::Utils::SO_TO_CP_TRANSLATIONS[schema_org] || "article-journal",
    "bibtex" => Bolognese::Utils::SO_TO_BIB_TRANSLATIONS[schema_org] || "misc",
    "ris" => Bolognese::Utils::SO_TO_RIS_TRANSLATIONS[resource_type_general.to_s.dasherize] || "GEN"
  }.compact
  authors = meta.fetch("author", nil) || meta.fetch("creator", nil)
  # Authors should be an object, if it's just a plain string don't try and parse it.
  if not authors.is_a?(String)
    creators = get_authors(from_schema_org_creators(Array.wrap(authors)))
  end
  contributors = get_authors(from_schema_org_contributors(Array.wrap(meta.fetch("editor", nil))))
  publisher = parse_attributes(meta.fetch("publisher", nil), content: "name", first: true)

  ct = (schema_org == "Dataset") ? "includedInDataCatalog" : "Periodical"
  container = if meta.fetch(ct, nil).present?
    url =  parse_attributes(from_schema_org(meta.fetch(ct, nil)), content: "url", first: true)

    {
      "type" => (schema_org == "Dataset") ? "DataRepository" : "Periodical",
      "title" => parse_attributes(from_schema_org(meta.fetch(ct, nil)), content: "name", first: true),
      "identifier" => url,
      "identifierType" => url.present? ? "URL" : nil,
      "volume" => meta.fetch("volumeNumber", nil),
      "issue" => meta.fetch("issueNumber", nil),
      "firstPage" => meta.fetch("pageStart", nil),
      "lastPage" => meta.fetch("pageEnd", nil)
    }.compact
  else
    {}
  end

  related_identifiers = Array.wrap(schema_org_is_identical_to(meta)) +
    Array.wrap(schema_org_is_part_of(meta)) +
    Array.wrap(schema_org_has_part(meta)) +
    Array.wrap(schema_org_is_previous_version_of(meta)) +
    Array.wrap(schema_org_is_new_version_of(meta)) +
    Array.wrap(schema_org_references(meta)) +
    Array.wrap(schema_org_is_referenced_by(meta)) +
    Array.wrap(schema_org_is_supplement_to(meta)) +
    Array.wrap(schema_org_is_supplemented_by(meta))

  rights_list = {
    "rightsUri" => parse_attributes(meta.fetch("license", nil), content: "id", first: true),
    "rights" => parse_attributes(meta.fetch("license", nil), content: "name", first: true)
  }

  funding_references = Array.wrap(meta.fetch("funder", nil)).compact.map do |fr|
    if fr["@id"].present?
      {
        "funderName" => fr["name"],
        "funderIdentifier" => fr["@id"],
        "funderIdentifierType" => fr["@id"].to_s.start_with?("https://doi.org/10.13039") ? "Crossref Funder ID" : "Other" }.compact
    else
      {
        "funderName" => fr["name"] }.compact
    end
  end
  dates = []
  dates << { "date" => meta.fetch("datePublished"), "dateType" => "Issued" } if meta.fetch("datePublished", nil).present?
  dates << { "date" => meta.fetch("dateCreated"), "dateType" => "Created" } if meta.fetch("dateCreated", nil).present?
  dates << { "date" => meta.fetch("dateModified"), "dateType" => "Updated" } if meta.fetch("dateModified", nil).present?
  publication_year = meta.fetch("datePublished")[0..3] if meta.fetch("datePublished", nil).present?

  state = meta.present? || read_options.present? ? "findable" : "not_found"
  geo_locations = Array.wrap(meta.fetch("spatialCoverage", nil)).map do |gl|
    if gl.dig("geo", "box")
      s, w, n, e = gl.dig("geo", "box").split(" ", 4)
      geo_location_box = {
        "westBoundLongitude" => w,
        "eastBoundLongitude" => e,
        "southBoundLatitude" => s,
        "northBoundLatitude" => n
      }.compact.presence
    else
      geo_location_box = nil
    end
    geo_location_point = { "pointLongitude" => gl.dig("geo", "longitude"), "pointLatitude" => gl.dig("geo", "latitude") }.compact.presence

    {
      "geoLocationPlace" => gl.dig("geo", "address"),
      "geoLocationPoint" => geo_location_point,
      "geoLocationBox" => geo_location_box
    }.compact
  end
  subjects = Array.wrap(meta.fetch("keywords", nil).to_s.split(", ")).map do |s|
    { "subject" => s }
  end

  { "id" => id,
    "types" => types,
    "doi" => validate_doi(id),
    "identifiers" => identifiers,
    "url" => normalize_id(meta.fetch("url", nil)),
    "content_url" => Array.wrap(meta.fetch("contentUrl", nil)),
    "sizes" => Array.wrap(meta.fetch("contenSize", nil)).presence,
    "formats" => Array.wrap(meta.fetch("encodingFormat", nil) || meta.fetch("fileFormat", nil)),
    "titles" => meta.fetch("name", nil).present? ? [{ "title" => meta.fetch("name", nil) }] : nil,
    "creators" => creators,
    "contributors" => contributors,
    "publisher" => publisher,
    "agency" => parse_attributes(meta.fetch("provider", nil), content: "name", first: true),
    "container" => container,
    "related_identifiers" => related_identifiers,
    "publication_year" => publication_year,
    "dates" => dates,
    "descriptions" => meta.fetch("description", nil).present? ? [{ "description" => sanitize(meta.fetch("description")), "descriptionType" => "Abstract" }] : nil,
    "rights_list" => rights_list,
    "version_info" => meta.fetch("version", nil).to_s.presence,
    "subjects" => subjects,
    "state" => state,
    "schema_version" => meta.fetch("schemaVersion", nil).to_s.presence,
    "funding_references" => funding_references,
    "geo_locations" => geo_locations
  }.merge(read_options)
end

#schema_org_has_part(meta) ⇒ Object



190
191
192
# File 'lib/bolognese/readers/schema_org_reader.rb', line 190

def schema_org_has_part(meta)
  schema_org_related_identifier(meta, relation_type: "hasPart")
end

#schema_org_is_identical_to(meta) ⇒ Object



182
183
184
# File 'lib/bolognese/readers/schema_org_reader.rb', line 182

def schema_org_is_identical_to(meta)
  schema_org_related_identifier(meta, relation_type: "sameAs")
end

#schema_org_is_new_version_of(meta) ⇒ Object



198
199
200
# File 'lib/bolognese/readers/schema_org_reader.rb', line 198

def schema_org_is_new_version_of(meta)
  schema_org_related_identifier(meta, relation_type: "SuccessorOf")
end

#schema_org_is_part_of(meta) ⇒ Object



186
187
188
# File 'lib/bolognese/readers/schema_org_reader.rb', line 186

def schema_org_is_part_of(meta)
  schema_org_related_identifier(meta, relation_type: "isPartOf")
end

#schema_org_is_previous_version_of(meta) ⇒ Object



194
195
196
# File 'lib/bolognese/readers/schema_org_reader.rb', line 194

def schema_org_is_previous_version_of(meta)
  schema_org_related_identifier(meta, relation_type: "PredecessorOf")
end

#schema_org_is_referenced_by(meta) ⇒ Object



206
207
208
# File 'lib/bolognese/readers/schema_org_reader.rb', line 206

def schema_org_is_referenced_by(meta)
  schema_org_reverse_related_identifier(meta, relation_type: "citation")
end

#schema_org_is_supplement_to(meta) ⇒ Object



210
211
212
# File 'lib/bolognese/readers/schema_org_reader.rb', line 210

def schema_org_is_supplement_to(meta)
  schema_org_reverse_related_identifier(meta, relation_type: "isBasedOn")
end

#schema_org_is_supplemented_by(meta) ⇒ Object



214
215
216
# File 'lib/bolognese/readers/schema_org_reader.rb', line 214

def schema_org_is_supplemented_by(meta)
  schema_org_related_identifier(meta, relation_type: "isBasedOn")
end

#schema_org_references(meta) ⇒ Object



202
203
204
# File 'lib/bolognese/readers/schema_org_reader.rb', line 202

def schema_org_references(meta)
  schema_org_related_identifier(meta, relation_type: "citation")
end


174
175
176
# File 'lib/bolognese/readers/schema_org_reader.rb', line 174

def schema_org_related_identifier(meta, relation_type: nil)
  normalize_ids(ids: meta.fetch(relation_type, nil), relation_type: SO_TO_DC_RELATION_TYPES[relation_type])
end


178
179
180
# File 'lib/bolognese/readers/schema_org_reader.rb', line 178

def schema_org_reverse_related_identifier(meta, relation_type: nil)
  normalize_ids(ids: meta.dig("@reverse", relation_type), relation_type: SO_TO_DC_RELATION_TYPES[relation_type])
end