data_fetcher.rb in relaton-nist-1.9.6

- old
+ new

@@ -9,32 +9,49 @@
       "isVersionOf" => "editionOf",
       "hasTranslation" => "hasTranslation",
       "isTranslationOf" => "translatedFrom",
       "hasPreprint" => "hasReprint",
       "isSupplementTo" => "complements",
+      "isPartOf" => "partOf",
+      "hasPart" => "hasPart",
     }.freeze
     URL = "https://raw.githubusercontent.com/usnistgov/NIST-Tech-Pubs/nist-pages/xml/allrecords.xml"
 
     def initialize(output, format)
       @output = output
       @format = format
       @ext = format.sub(/^bib/, "")
     end
 
-    def parse_docid(doc)
-      doi = doc.at("doi_data/doi").text
-      id = doc.at("publisher_item/item_number", "publisher_item/identifier").text.sub(%r{^/}, "")
-      case doi
-      when "10.6028/NBS.CIRC.12e2revjune" then id.sub!("13e", "12e")
-      when "10.6028/NBS.CIRC.36e2" then id.sub!("46e", "36e")
-      when "10.6028/NBS.HB.67suppJune1967" then id.sub!("1965", "1967")
-      when "10.6028/NBS.HB.105-1r1990" then id.sub!("105-1-1990", "105-1r1990")
-      when "10.6028/NIST.HB.150-10-1995" then id.sub!(/150-10$/, "150-10-1995")
-      end
-      [{ type: "NIST", id: id }, { type: "DOI", id: doi }]
+    def parse_docid(doc) # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
+      # case doi
+      # when "10.6028/NBS.CIRC.12e2revjune" then doi.sub!("13e", "12e")
+      # when "10.6028/NBS.CIRC.36e2" then doi.sub!("46e", "36e")
+      # when "10.6028/NBS.HB.67suppJune1967" then doi.sub!("1965", "1967")
+      # when "10.6028/NBS.HB.105-1r1990" then doi.sub!("105-1-1990", "105-1r1990")
+      # when "10.6028/NIST.HB.150-10-1995" then doi.sub!(/150-10$/, "150-10-1995")
+      # end
+      # anchor = doi.split("/")[1..-1].join "/"
+      [
+        { type: "NIST", id: pub_id(doc) },
+        { type: "DOI", id: doi(doc) },
+        { type: "NIST", id: anchor(doc), scope: "anchor" },
+      ]
     end
 
+    def pub_id(doc)
+      anchor(doc).gsub(".", " ")
+    end
+
+    def doi(doc)
+      doc.at("doi_data/doi").text
+    end
+
+    def anchor(doc)
+      doi(doc).split("/")[1..-1].join "/"
+    end
+
     # @param doc [Nokogiri::XML::Element]
     # @return [Array<RelatonBib::DocumentIdentifier>]
     def fetch_docid(doc)
       parse_docid(doc).map do |id|
         RelatonBib::DocumentIdentifier.new(type: id[:type], id: id[:id])
@@ -45,11 +62,11 @@
     # @return [RelatonBib::TypedTitleStringCollection, Array]
     def fetch_title(doc)
       t = doc.xpath("titles/title|titles/subtitle")
       return [] unless t.any?
 
-      RelatonBib::TypedTitleString.from_string t.map(&:text).join(" "), "en", "Latn"
+      RelatonBib::TypedTitleString.from_string t.map(&:text).join, "en", "Latn"
     end
 
     # @param doc [Nokogiri::XML::Element]
     # @return [Array<RelatonBib::BibliographicDate>]
     def fetch_date(doc)
@@ -74,16 +91,15 @@
     # @param doc [Nokogiri::XML::Element]
     # @return [Array<Hash>]
     def fetch_relation(doc)
       ns = "http://www.crossref.org/relations.xsd"
       doc.xpath("./ns:program/ns:related_item", ns: ns).map do |rel|
-        doi = rel.at_xpath("ns:intra_work_relation|ns:inter_work_relation", ns: ns)
-        # ref = doi_to_id doi.text
-        # ref, = parse_docid doc
-        fref = RelatonBib::FormattedRef.new content: doi.text
+        rdoi = rel.at_xpath("ns:intra_work_relation|ns:inter_work_relation", ns: ns)
+        fref = RelatonBib::FormattedRef.new content: rdoi.text
         bibitem = RelatonBib::BibliographicItem.new formattedref: fref
-        type = RELATION_TYPES[doi["relationship-type"]]
+        type = RELATION_TYPES[rdoi["relationship-type"]]
+        warn "Relation type #{rdoi['relationship-type']} not found" unless type
         { type: type, bibitem: bibitem }
       end
     end
 
     # @param doc [Nokogiri::XML::Element]
@@ -121,26 +137,46 @@
           RelatonBib::PersonIdentifier.new "orcid", id.text
         end
         fullname = RelatonBib::FullName.new(
           surname: surname, forename: forename, initial: initial, identifier: ident,
         )
-        person = RelatonBib::Person.new name: fullname
+        person = RelatonBib::Person.new name: fullname, affiliation: affiliation(doc)
         { entity: person, role: [{ type: p["contributor_role"] }] }
       end
       contribs + doc.xpath("publisher").map do |p|
         abbr = p.at("../institution/institution_acronym")&.text
-        org = RelatonBib::Organization.new(name: p.at("publisher_name").text, abbreviation: abbr)
+        place = p.at("./publisher_place")
+        cont = []
+        if place
+          city, state = place.text.split(", ")
+          cont << RelatonBib::Address.new(street: [], city: city, state: state, country: "US")
+        end
+        org = RelatonBib::Organization.new(
+          name: p.at("publisher_name").text, abbreviation: abbr, contact: cont,
+        )
         { entity: org, role: [{ type: "publisher" }] }
       end
     end
 
+    def affiliation(doc)
+      doc.xpath("./institution/institution_department").map do |id|
+        org = RelatonBib::Organization.new name: id.text
+        RelatonBib::Affiliation.new organization: org
+      end
+    end
+
     # @param doc [Nokogiri::XML::Element]
     # @return [Array<String>]
     def fetch_place(doc)
       doc.xpath("institution/institution_place").map(&:text)
     end
 
+    def fetch_series(doc)
+      title = RelatonBib::TypedTitleString.new(content: "NIST")
+      [RelatonBib::Series.new(title: title, number: pub_id(doc))]
+    end
+
     #
     # Save document
     #
     # @param bib [RelatonNist::NistBibliographicItem]
     #
@@ -172,18 +208,19 @@
       item = RelatonNist::NistBibliographicItem.new(
         type: "standard", docid: fetch_docid(doc), title: fetch_title(doc),
         link: fetch_link(doc), abstract: fetch_abstract(doc),
         date: fetch_date(doc), edition: fetch_edition(doc),
         contributor: fetch_contributor(doc), relation: fetch_relation(doc),
-        place: fetch_place(doc),
+        place: fetch_place(doc), series: fetch_series(doc),
         language: [doc["language"]], script: ["Latn"], doctype: "standard"
       )
       write_file item
     rescue StandardError => e
       warn "Document: #{doc.at('doi').text}"
       warn e.message
-      raise e
+      warn e.backtrace[0..5].join("\n")
+      # raise e
     end
 
     #
     # Fetch all the documnts from dataset
     #
@@ -200,9 +237,10 @@
       t2 = Time.now
       puts "Stopped at: #{t2}"
       puts "Done in: #{(t2 - t1).round} sec."
     rescue StandardError => e
       warn e.message
+      warn e.backtrace[0..5].join("\n")
     end
 
     #
     # Fetch all the documnts from dataset
     #