stanford-mods 2.3.1 → 2.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: bf9995bc6d5890bb8520ec918ef23ef8fc6a215e
-  data.tar.gz: 6a39d07658e000e296d40e3666b93ead0c3a32ca
+  metadata.gz: a4fe370e5f566e6a8aaf2cecf78f8d2f261e83d3
+  data.tar.gz: 7fd7c27ee40726c0509190ef9dfd398fba91d95d
 SHA512:
-  metadata.gz: 3fb187dbfc6ac4f2f1f21a68dc8cd43940f5cf8e2f67b1895392cc9adbd91eccf163cd6d8e35e5f7fbd72fadd1a0ff7019f2f8120e790f31b8e2d65c275fe5b6
-  data.tar.gz: 4d5e7dbf3099da891b5c15b5d3e64c27b72c1ea305c503b068a5890324dc818db9fb644d500e20afaa0729fb2202628d621f12fd3b8a95fdcfd020d22fdedc05
+  metadata.gz: ac960448cdf1c552d8e78822735bfacce0dfd193a729e8f8f8f4b6516c6eb06cb639546ce3183f5be5d6283251036e82b4f53ed04b2d482fe01da80440d208fc
+  data.tar.gz: bbdd468328c00a918ecf898a2217f58cf8b3737a4e541246da480103bf9405293350410f2629f4492ae591bb47d0dd8a6223dce1f0abf88a55c2de1c18ce1681

data/lib/stanford-mods/geo_spatial.rb

@@ -18,19 +18,36 @@ module Stanford
       #  <gml:upperCorner>-122.149475 37.4435369</gml:upperCorner>
       def geo_extensions_as_envelope
         mods_ng_xml.extension
-                    .xpath('//rdf:RDF/rdf:Description/gml:boundedBy/gml:Envelope',
-                      'gml' => GMLNS,
-                      'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'
-                    ).map do |v|
-                      uppers = v.xpath('gml:upperCorner', 'gml' => GMLNS).text.split
-                      lowers = v.xpath('gml:lowerCorner', 'gml' => GMLNS).text.split
-                      "ENVELOPE(#{lowers[0]}, #{uppers[0]}, #{uppers[1]}, #{lowers[1]})"
-                    end
+                   .xpath(
+                     '//rdf:RDF/rdf:Description/gml:boundedBy/gml:Envelope',
+                     'gml' => GMLNS,
+                     'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'
+                   ).map do |v|
+                     uppers = v.xpath('gml:upperCorner', 'gml' => GMLNS).text.split
+                     lowers = v.xpath('gml:lowerCorner', 'gml' => GMLNS).text.split
+                     "ENVELOPE(#{lowers[0]}, #{uppers[0]}, #{uppers[1]}, #{lowers[1]})"
+                   end
       rescue RuntimeError => e
         logger.warn "failure parsing <extension> element: #{e.message}"
         []
       end
 
+      # @return [Array{String}] values suitable for solr SRPT fields, like "-16.0 28.0"
+      # @note example xml leaf nodes
+      #  <gml:pos>-122.191292 37.4063388</gml:pos>
+      def geo_extensions_point_data
+        mods_ng_xml.extension
+                   .xpath(
+                     '//rdf:RDF/rdf:Description/gmd:centerPoint/gml:Point[gml:pos]',
+                     'gml' => GMLNS,
+                     'gmd' => 'http://www.isotc211.org/2005/gmd',
+                     'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'
+                   ).map do |v|
+                     lat, long = v.xpath('gml:pos', 'gml' => GMLNS).text.split
+                     "#{long} #{lat}"
+                   end
+      end
+
       # @return [Array{Stanford::Mods::Coordinate}] valid coordinates as objects
       def coordinates_objects
         coordinates.map { |n| Stanford::Mods::Coordinate.new(n) }.select(&:valid?)
@@ -47,6 +64,6 @@ module Stanford
       end
 
       alias point_bbox coordinates_as_bbox
-    end # class Record
-  end # Module Mods
-end # Module Stanford
+    end
+  end
+end

data/lib/stanford-mods/origin_info.rb

@@ -13,10 +13,9 @@ module Stanford
       # return pub year as an Integer
       # prefer dateIssued (any) before dateCreated (any) before dateCaptured (any)
       #  look for a keyDate and use it if there is one;  otherwise pick earliest date
-      # @param [Boolean] ignore_approximate true if approximate dates (per qualifier attribute)
-      #   should be ignored; false if approximate dates should be included
+      # @param [Boolean] ignore_approximate true if approximate dates (per qualifier attribute) should be ignored; false if approximate dates should be included
       # @return [Integer] publication year as an Integer
-      #   note that for sorting  5 B.C. => -5;  666 B.C. => -666
+      # @note for sorting:  5 B.C. => -5;  666 B.C. => -666
       def pub_year_int(ignore_approximate = false)
         single_pub_year(ignore_approximate, :year_int)
       end
@@ -24,11 +23,10 @@ module Stanford
       # return a single string intended for lexical sorting for pub date
       # prefer dateIssued (any) before dateCreated (any) before dateCaptured (any)
       #  look for a keyDate and use it if there is one;  otherwise pick earliest date
-      # @param [Boolean] ignore_approximate true if approximate dates (per qualifier attribute)
-      #   should be ignored; false if approximate dates should be included
+      # @param [Boolean] ignore_approximate true if approximate dates (per qualifier attribute) should be ignored; false if approximate dates should be included
       # @return [String] single String containing publication year for lexical sorting
-      #   note that for string sorting  5 B.C. = -5  => -995;  6 B.C. => -994  so 6 B.C. sorts before 5 B.C.
-      # @deprecated  use pub_year_int
+      # @note for string sorting  5 B.C. = -5  => -995;  6 B.C. => -994, so 6 B.C. sorts before 5 B.C.
+      # @deprecated use pub_year_int
       def pub_year_sort_str(ignore_approximate = false)
         single_pub_year(ignore_approximate, :year_sort_str)
       end
@@ -186,8 +184,7 @@ module Stanford
         result = send(method_sym, date_issued_elements(ignore_approximate))
         result ||= send(method_sym, date_created_elements(ignore_approximate))
         # dateCaptured for web archive seed records
-        result ||= send(method_sym, mods_ng_xml.origin_info.dateCaptured.to_a)
-        result
+        result || send(method_sym, mods_ng_xml.origin_info.dateCaptured.to_a)
       end
 
       # given the passed date elements, look for a single keyDate and use it if there is one;
@@ -210,10 +207,9 @@ module Stanford
         # get earliest parseable year from the passed date elements
         # @param [Array<Nokogiri::XML::Element>] date_el_array the elements from which to select a pub date
         # @param [Symbol] method_sym method name in DateParsing, as a symbol
-        # @return two values:
-        #   the first is either:  the lexically sortable String value of the earliest date or the Integer value of same,
-        #     depending on the method_sym passed in
-        #   the second is the original String value of the chosen element
+        # @return [Array<String,Integer>] two values: earliest date and the original element string
+        #   - first is earliest date either as lexically sortable String value or the Integer, depending on method_sym
+        #   - second is the original String value of the chosen element
         def earliest_year(date_el_array, method_sym)
           poss_results = {}
           date_el_array.each { |el|
@@ -228,8 +224,7 @@ module Stanford
 # ----   old date parsing methods used downstream of gem;  will be deprecated/replaced with new date parsing methods
 
       def place
-        vals = term_values([:origin_info, :place, :placeTerm])
-        vals
+        term_values([:origin_info, :place, :placeTerm])
       end
 
       # Values for the pub date facet. This is less strict than the 4 year date requirements for pub_date
@@ -237,28 +232,18 @@ module Stanford
       #   Spotlight:  pub_date field should be replaced by pub_year_w_approx_isi and pub_year_no_approx_isi
       #   SearchWorks:  pub_date field used for display in search results and show view; for sorting nearby-on-shelf
       #      these could be done with more approp fields/methods (pub_year_int for sorting;  new pub year methods to populate field)
-      # TODO:  prob should deprecate this in favor of pub_year_display_str;
+      # TODO: prob should deprecate this in favor of pub_year_display_str;
       #    need head-to-head testing with pub_year_display_str
-      # @return <Array[String]> with values for the pub date facet
+      # @return [String] value for the pub date facet
       def pub_date_facet
-        if pub_date
-          if pub_date.start_with?('-')
-            return (pub_date.to_i + 1000).to_s + ' B.C.'
-          end
-          if pub_date.include? '--'
-            cent = pub_date[0, 2].to_i
-            cent += 1
-            cent = cent.to_s + 'th century'
-            return cent
-          else
-            return pub_date
-          end
-        end
-        nil
+        return nil unless pub_date
+        return "#{pub_date.to_i + 1000} B.C." if pub_date.start_with?('-')
+        return pub_date unless pub_date.include? '--'
+        "#{pub_date[0, 2].to_i + 1}th century"
       end
 
       # creates a date suitable for sorting. Guarnteed to be 4 digits or nil
-      # @deprecated:  use pub_year_int, or pub_year_sort_str if you must have a string (why?)
+      # @deprecated use pub_year_int, or pub_year_sort_str if you must have a string (why?)
       def pub_date_sort
         if pub_date
           pd = pub_date
@@ -272,14 +257,14 @@ module Stanford
       # For the date display only, the first place to look is in the dates without encoding=marc array.
       # If no such dates, select the first date in the dates_marc_encoding array.  Otherwise return nil
       # @return [String] value for the pub_date_display Solr field for this document or nil if none
-      # @deprecated:  DO NOT USE: this is no longer used in SW, Revs or Spotlight Jan 2016
+      # @deprecated DO NOT USE: this is no longer used in SW, Revs or Spotlight Jan 2016
       def pub_date_display
         return dates_no_marc_encoding.first unless dates_no_marc_encoding.empty?
         return dates_marc_encoding.first unless dates_marc_encoding.empty?
         nil
       end
 
-# ----   old date parsing protected methods will be deprecated/replaced with new date parsing methods (see also DateParsing)
+# old date parsing protected methods to be deprecated/replaced with new methods (see also DateParsing)
 
     protected
 
@@ -292,30 +277,23 @@ module Stanford
           return @pub_year
         end
 
-        dates = pub_dates
-        if dates
-          pruned_dates = []
-          dates.each do |f_date|
-            # remove ? and []
-            if f_date.length == 4 && f_date.end_with?('?')
-              pruned_dates << f_date.tr('?', '0')
-            else
-              pruned_dates << f_date.delete('?[]')
-            end
+        dates = pub_dates.map do |f_date|
+          # remove ? and []
+          if f_date.length == 4 && f_date.end_with?('?')
+            f_date.tr('?', '0')
+          else
+            f_date.delete('?[]')
           end
+        end
+
+        if dates
           # try to find a date starting with the most normal date formats and progressing to more wonky ones
-          @pub_year = get_plain_four_digit_year pruned_dates
-          return @pub_year if @pub_year
-          # Check for years in u notation, e.g., 198u
-          @pub_year = get_u_year pruned_dates
-          return @pub_year if @pub_year
-          @pub_year = get_double_digit_century pruned_dates
-          return @pub_year if @pub_year
-          @pub_year = get_bc_year pruned_dates
-          return @pub_year if @pub_year
-          @pub_year = get_three_digit_year pruned_dates
-          return @pub_year if @pub_year
-          @pub_year = get_single_digit_century pruned_dates
+          @pub_year = get_plain_four_digit_year(dates) ||
+                      get_u_year(dates)                || # Check for years in u notation, e.g., 198u
+                      get_double_digit_century(dates)  ||
+                      get_bc_year(dates)               ||
+                      get_three_digit_year(dates)      ||
+                      get_single_digit_century(dates)
           return @pub_year if @pub_year
         end
         @pub_year = ''
@@ -384,7 +362,6 @@ module Stanford
           matches = f_date.scan(/\d{4}/)
           if matches.length == 1
             @pub_year = matches.first
-            return matches.first
           else
             # when there are multiple matches, check for ones with CE after them
             matches.each do |match|
@@ -396,8 +373,8 @@ module Stanford
                 return match
               end
             end
-            return matches.first
           end
+          return matches.first
         end
         nil
       end

data/lib/stanford-mods/physical_location.rb

@@ -8,78 +8,56 @@ module Stanford
     # Each method here expects to find at most ONE matching element.  Subsequent potential matches
     # are ignored.
     class Record < ::Mods::Record
-      # return box number (note: single valued and might be something like 35A)
-      #   data in location/physicalLocation or in relatedItem/location/physicalLocation
-      #   so use _location to get the data from either one of them
-      # TODO:  should it be hierarchical series/box/folder?
+      # data in location/physicalLocation or in relatedItem/location/physicalLocation
+      # so use _location to get the data from either one of them
+      # @return [String] box number (note: single valued and might be something like 35A)
+      # @todo should it be hierarchical series/box/folder?
       def box
-        #   _location.physicalLocation should find top level and relatedItem
-        box_num = @mods_ng_xml._location.physicalLocation.map do |node|
-          val = node.text
-          # note that this will also find Flatbox or Flat-box
-          match_data = val.match(/Box ?:? ?([^,|(Folder)]+)/i)
-          match_data[1].strip if match_data.present?
-        end.compact
-
-        # There should only be one box
-        box_num.first
+        mods_ng_xml._location.physicalLocation.each do |node|
+          match_data = node.text.match(/Box ?:? ?([^,|(Folder)]+)/i) # note that this will also find Flatbox or Flat-box
+          return match_data[1].strip if match_data.present?
+        end
+        nil
       end
 
-      # returns folder number (note: single valued)
-      #   data in location/physicalLocation or in relatedItem/location/physicalLocation
-      #   so use _location to get the data from either one of them
-      # TODO:  should it be hierarchical series/box/folder?
+      # data in location/physicalLocation or in relatedItem/location/physicalLocation
+      # so use _location to get the data from either one of them
+      # @return [String] folder number (note: single valued)
+      # @todo should it be hierarchical series/box/folder?
       def folder
-        #   _location.physicalLocation should find top level and relatedItem
-        folder_num = @mods_ng_xml._location.physicalLocation.map do |node|
+        mods_ng_xml._location.physicalLocation.each do |node|
           val = node.text
-
-          match_data = if val =~ /\|/
-                         # we assume the data is pipe-delimited, and may contain commas within values
-                         val.match(/Folder ?:? ?([^|]+)/)
-                       else
-                         # the data should be comma-delimited, and may not contain commas within values
-                         val.match(/Folder ?:? ?([^,]+)/)
-                       end
-
-          match_data[1].strip if match_data.present?
-        end.compact
-
-        # There should be one folder
-        folder_num.first
+          match_data = val =~ /\|/ ?
+                       val.match(/Folder ?:? ?([^|]+)/) : # expect pipe-delimited, may contain commas within values
+                       val.match(/Folder ?:? ?([^,]+)/)   # expect comma-delimited, may NOT contain commas within values
+          return match_data[1].strip if match_data.present?
+        end
+        nil
       end
 
-      # return entire contents of physicalLocation as a string (note: single valued)
-      #   but only if it has series, accession, box or folder data
-      #   data in location/physicalLocation or in relatedItem/location/physicalLocation
-      #   so use _location to get the data from either one of them
-      # TODO:  should it be hierarchical series/box/folder?
-      # NOTE: there is a "physicalLocation" and a "location" method defined in the mods gem, so we cannot use these names to avoid conflicts
+      # but only if it has series, accession, box or folder data
+      # data in location/physicalLocation or in relatedItem/location/physicalLocation
+      # so use _location to get the data from either one of them
+      # @return [String] entire contents of physicalLocation as a string (note: single valued)
+      # @note there is a "physicalLocation" and a "location" method defined in the mods gem, so we cannot use these names to avoid conflicts
+      # @todo should it be hierarchical series/box/folder?
       def physical_location_str
-        #   _location.physicalLocation should find top level and relatedItem
-        loc = @mods_ng_xml._location.physicalLocation.map do |node|
-          node.text if node.text =~ /.*(Series)|(Accession)|(Folder)|(Box).*/i
-        end.compact
-
-        # There should only be one location
-        loc.first
+        mods_ng_xml._location.physicalLocation.map(&:text).find do |text|
+          text =~ /.*(Series)|(Accession)|(Folder)|(Box).*/i
+        end
       end
 
-      # return series/accession 'number' (note: single valued)
-      #   data in location/physicalLocation or in relatedItem/location/physicalLocation
-      #   so use _location to get the data from either one of them
-      # TODO: should it be hierarchical series/box/folder?
+      # data in location/physicalLocation or in relatedItem/location/physicalLocation
+      # so use _location to get the data from either one of them
+      # @return [String] series/accession 'number' (note: single valued)
+      # @todo should it be hierarchical series/box/folder?
       def series
-        #   _location.physicalLocation should find top level and relatedItem
-        series_num = @mods_ng_xml._location.physicalLocation.map do |node|
-          val = node.text
+        mods_ng_xml._location.physicalLocation.each do |node|
           # feigenbaum uses 'Accession'
-          match_data = val.match(/(?:(?:Series)|(?:Accession)):? ([^,|]+)/i)
-          match_data[1].strip if match_data.present?
-        end.compact
-
-        # There should be only one series
-        series_num.first
+          match_data = node.text.match(/(?:(?:Series)|(?:Accession)):? ([^,|]+)/i)
+          return match_data[1].strip if match_data.present?
+        end
+        nil
       end
     end # class Record
   end # Module Mods

data/lib/stanford-mods/searchworks_subjects.rb

@@ -16,21 +16,15 @@ module Stanford
         result = term_values([:subject, :geographic]) || []
 
         # hierarchicalGeographic has sub elements
-        @mods_ng_xml.subject.hierarchicalGeographic.each { |hg_node|
-          hg_vals = []
-          hg_node.element_children.each { |e|
-            hg_vals << e.text unless e.text.empty?
-          }
+        mods_ng_xml.subject.hierarchicalGeographic.each { |hg_node|
+          hg_vals = hg_node.element_children.map(&:text).reject(&:empty?)
           result << hg_vals.join(sep) unless hg_vals.empty?
         }
 
-        trans_code_vals = @mods_ng_xml.subject.geographicCode.translated_value
-        if trans_code_vals
-          trans_code_vals.each { |val|
-            result << val unless result.include?(val)
-          }
-        end
-
+        trans_code_vals = mods_ng_xml.subject.geographicCode.translated_value || []
+        trans_code_vals.each { |val|
+          result << val unless result.include?(val)
+        }
         result
       end
 
@@ -40,12 +34,11 @@ module Stanford
       # @param [String] sep - the separator string for joining namePart sub elements
       # @return [Array<String>] values for names inside subject elements or [] if none
       def sw_subject_names(sep = ', ')
-        result = []
-        @mods_ng_xml.subject.name_el.select { |n_el| n_el.namePart }.each { |name_el_w_np|
-          parts = name_el_w_np.namePart.map { |npn| npn.text unless npn.text.empty? }.compact
-          result << parts.join(sep).strip unless parts.empty?
-        }
-        result
+        mods_ng_xml.subject.name_el
+                   .select { |n_el| n_el.namePart }
+                   .map { |name_el_w_np| name_el_w_np.namePart.map(&:text).reject(&:empty?) }
+                   .reject(&:empty?)
+                   .map { |parts| parts.join(sep).strip }
       end
 
       # Values are the contents of:
@@ -54,8 +47,8 @@ module Stanford
       # @return [Array<String>] values for titles inside subject elements or [] if none
       def sw_subject_titles(sep = ' ')
         result = []
-        @mods_ng_xml.subject.titleInfo.each { |ti_el|
-          parts = ti_el.element_children.map { |el| el.text unless el.text.empty? }.compact
+        mods_ng_xml.subject.titleInfo.each { |ti_el|
+          parts = ti_el.element_children.map(&:text).reject(&:empty?)
           result << parts.join(sep).strip unless parts.empty?
         }
         result
@@ -85,10 +78,7 @@ module Stanford
         vals.concat(subject_names) if subject_names
         vals.concat(subject_titles) if subject_titles
         vals.concat(subject_occupations) if subject_occupations
-        vals.map! { |val|
-          v = val.sub(/[\\,;]$/, '')
-          v.strip
-        }
+        vals.map! { |val| val.sub(/[\\,;]$/, '').strip }
         vals.empty? ? nil : vals
       end
 
@@ -119,9 +109,8 @@ module Stanford
           codes = term_values([:subject, :geographicCode])
           if codes && codes.size > xvals.size
             subject.geographicCode.each { |n|
-              if n.authority != 'marcgac' && n.authority != 'marccountry'
-                sw_logger.info("#{druid} has subject geographicCode element with untranslated encoding (#{n.authority}): #{n.to_xml}")
-              end
+              next unless n.authority != 'marcgac' && n.authority != 'marccountry'
+              sw_logger.info("#{druid} has subject geographicCode element with untranslated encoding (#{n.authority}): #{n.to_xml}")
             }
           end
 

data/lib/stanford-mods/version.rb

@@ -1,6 +1,6 @@
 module Stanford
   module Mods
     # this is the Ruby Gem version
-    VERSION = "2.3.1".freeze
+    VERSION = "2.4.0".freeze
   end
 end

data/spec/geo_spatial_spec.rb

@@ -98,7 +98,7 @@ describe '#geo_extension_as_envelope' do
     expect(mods.geo_extensions_as_envelope).to eq []
   end
 
-  describe 'with data' do
+  describe 'with geo extension bounding-box data' do
     let(:modsbody) do
       <<-EOF
         <extension displayLabel="geo">
@@ -128,4 +128,28 @@ describe '#geo_extension_as_envelope' do
       expect(mods.geo_extensions_as_envelope).to eq ["ENVELOPE(-122.191292, -122.149475, 37.4435369, 37.4063388)"]
     end
   end
+
+  describe 'with geo-extension point data' do
+    let(:modsbody) do
+      <<-EOF
+        <extension displayLabel="geo">
+          <rdf:RDF xmlns:gml="http://www.opengis.net/gml/3.2/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:gmd="http://www.isotc211.org/2005/gmd">
+            <rdf:Description rdf:about="http://www.stanford.edu/cm896kp1291">
+              <dc:format>image/jpeg</dc:format>
+              <dc:type>Image</dc:type>
+              <gmd:centerPoint>
+                <gml:Point gml:id="ID">
+                  <gml:pos>41.8898687280593 12.4913412520789</gml:pos>
+                </gml:Point>
+              </gmd:centerPoint>
+            </rdf:Description>
+          </rdf:RDF>
+        </extension>
+      EOF
+    end
+
+    it 'extract point strings' do
+      expect(mods.geo_extensions_point_data).to eq ["12.4913412520789 41.8898687280593"]
+    end
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: stanford-mods
 version: !ruby/object:Gem::Version
-  version: 2.3.1
+  version: 2.4.0
 platform: ruby
 authors:
 - Naomi Dushay
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-09-22 00:00:00.000000000 Z
+date: 2017-01-24 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: mods
@@ -177,7 +177,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.5.1
+rubygems_version: 2.5.2
 signing_key: 
 specification_version: 4
 summary: Stanford specific wrangling of MODS metadata