pennmarc 1.0.3 → 1.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 28dc70104803899434767b340182cb8eecff8c402cfb5e8547058d560a45a1e5
-  data.tar.gz: 1fc0897f12ffb24de8bc0f40b69474d29264c036e33ba4ea97390b60a9f60c43
+  metadata.gz: 50fd0383f0e78807f62f8abe784f75d8dace6e5b0ecf64877f79ad90b6d40354
+  data.tar.gz: 3df19d10534fc787c55814e30bd1b066a8c5ea10e7da15f9b320c4500891e0be
 SHA512:
-  metadata.gz: c82d8a22f480a768a8f6cce3cb7bc2ed25f7d5b202e3b492ac572a6f5bfec0d801098e1d74db8eb470573a2afbdd9314a90e18955a576c9ff5dee42aa2fbc5cd
-  data.tar.gz: d3c69ce76e13c50b0f2da649ea9499dcb93a734944d298cb783e71e0c4bc59c76fad0ea344353124f024524d11b5f8c92318771d780848506d50940f483c427b
+  metadata.gz: 277bb7c15e224c8134b8cea8b28de474d27022a1537904844af8c5542ff3f0080c37f13db38a5a9463a07453bcc403ccf346c2b5538c2fe26012df6a67dec24f
+  data.tar.gz: 1ca8e643758f86aeabdfbeb1576f513c866ba5ef1a40b6edaa279f687af7644baf6b953db6f078548fc72ca99b75b6377db7d80d98a19059c797b6ad41b5f648

data/.gitleaks.toml

@@ -0,0 +1,2 @@
+[file]
+paths-exclude = ["lib/pennmarc/mappings/language.yml"]

data/lib/pennmarc/helpers/date.rb

@@ -25,21 +25,17 @@ module PennMARC
       # @return [DateTime, nil] The date added, or nil if date found in record is invalid
       def added(record)
         record.fields(EnrichedMarc::TAG_ITEM).flat_map { |field|
-          field.filter_map do |subfield|
-            # skip unless field has date created subfield
-            next unless subfield_defined?(field, EnrichedMarc::SUB_ITEM_DATE_CREATED)
-
+          subfield_values(field, EnrichedMarc::SUB_ITEM_DATE_CREATED).filter_map do |date_added|
             # On 2022-05-02, this field value (as exported in enriched publishing
             # job from Alma) began truncating time to day-level granularity. We have
             # no guarantee that this won't switch back in the future, so for the
             # foreseeable future we should support both formats.
 
-            format = subfield.value.size == 10 ? '%Y-%m-%d' : '%Y-%m-%d %H:%M:%S'
-
-            DateTime.strptime(subfield.value, format)
+            format = date_added.size == 10 ? '%Y-%m-%d' : '%Y-%m-%d %H:%M:%S'
 
+            DateTime.strptime(date_added, format)
           rescue StandardError => e
-            puts "Error parsing date in date added subfield: #{subfield.value} - #{e}"
+            puts "Error parsing date in date added subfield: #{date_added} - #{e}"
             nil
           end
         }.max

data/lib/pennmarc/helpers/language.rb

@@ -4,10 +4,10 @@ module PennMARC
   # Logic for extracting and translating Language values for a record. Penn practice is to verify the value present in
   # the {https://www.oclc.org/bibformats/en/fixedfield/lang.html 008 control field} as a three letter code. This code
   # is then mapped to a display-friendly value using the a provided mapping hash.
-  # @todo should we consider values in the {https://www.oclc.org/bibformats/en/0xx/041.html 041 field}?
   class Language < Helper
     # Used when no value is present in the control field - still mapped
     UNDETERMINED_CODE = :und
+    LANGUAGE_SUBFIELDS = %w[a b d e g h i j k m n p q r t].freeze
 
     class << self
       # Get language values for display from the {https://www.oclc.org/bibformats/en/5xx/546.html 546 field} and
@@ -21,16 +21,31 @@ module PennMARC
         values + linked_alternate(record, '546', &subfield_not_in?(%w[6 8]))
       end
 
-      # Get language values for searching and faceting of a record. The value is extracted from a defined position in
-      # the 008 control field. Language facet and search values will typically be the same.
+      # Get language values for searching and faceting of a record. The values are extracted from subfields
+      # in the 041 field. Language facet and search values will typically be the same, with the exception of `zxx`,
+      # when no linguistic content is found.
+      #
+      # @note In franklin, we extracted the language code from the 008 control field. After engaging cataloging unit
+      #   representatives, we decided to extract these values from the 041 field: Includes records for multilingual
+      #   items, items that involve translation, and items where the medium of communication is a sign language.
+      #   https://www.loc.gov/marc/bibliographic/bd041.html
       #
       # @param [MARC::Record] record
-      # @param [Hash] language_map hash for language code translation
-      # @return [String] nice value for language
-      def search(record, language_map: Mappers.language)
+      # @param [Hash] iso_639_2_mapping iso-639-2 spec hash for language code translation
+      # @param [Hash] iso_639_3_mapping iso-639-3 spec hash for language code translation
+      # @return [Array] array of language values
+      def values(record, iso_639_2_mapping: Mappers.iso_639_2_language, iso_639_3_mapping: Mappers.iso_639_3_language)
+        values = record.fields('041').filter_map { |field|
+          mapper = subfield_value?(field, '2', /iso639-3/) ? iso_639_3_mapping : iso_639_2_mapping
+          field.filter_map do |sf|
+            next unless LANGUAGE_SUBFIELDS.include? sf.code
+
+            mapper[sf.value&.to_sym]
+          end
+        }.flatten
         control_field = record['008']&.value
-        language_code = control_field[35..37]
-        language_map[language_code.to_sym || UNDETERMINED_CODE]
+        values << iso_639_2_mapping[control_field[35..37]&.to_sym] if control_field.present?
+        values.empty? ? values << iso_639_2_mapping[UNDETERMINED_CODE] : values.uniq
       end
     end
   end

data/lib/pennmarc/mappers.rb

@@ -5,8 +5,12 @@ module PennMARC
   class Mappers
     class << self
       # @return [Hash]
-      def language
-        @language ||= load_map('language.yml')
+      def iso_639_2_language
+        @iso_639_2_language ||= load_map('iso639-2-languages.yml')
+      end
+
+      def iso_639_3_language
+        @iso_639_3_language ||= load_map('iso639-3-languages.yml')
       end
 
       # @return [Hash]