Merge pull request #198 from lyrasis/marc-lang

`Marc::LanguageCodeLookup` transform; fix meeting name subfield config

CHANGELOG.adoc

@@ -32,11 +32,13 @@ These changes are merged into the `main` branch, but have not been released. Aft
 
 === Bugfixes
 
+* Role term and subordinate body subfields for meeting names fixed in default config.
 * `IterativeCleanup` now automatically extends its extending module with `Dry::Configurable` prior to defining settings that depend on `Dry::Configurable`. (PR#192)
 * `Kiba::Extend::Job.output?` no longer fails if given job returns Nil (PR#194)
 * `Reshape::FieldsToFieldGroupWithConstant` constant value is no longer added to rows with no values in the renamed/remapped value fields, when fieldmap length == 1. (PR#195)
 
 === Added
+* `MARC::LanguageCodeLookup` transform
 * Ability to pass `find` argument to `Clean::RegexpFindReplaceFieldVals` as a `Regexp` object. Not sure why this was not the default initial behavior, but here we are! (PR#196)
 
 === Changed

lib/kiba/extend/marc.rb

@@ -60,7 +60,7 @@ module Marc
       # @return [Array<String>] subfields to be extracted as part of name values
       #   from {meeting_data_tags}
       setting :meeting_name_part_subfields,
-        default: %w[a q b c d g n u],
+        default: %w[a q b c e d g n u],
         reader: true
       # @return [Array<String>] subfields to be extracted as meeting role codes
       #   from {meeting_data_tags}
@@ -70,7 +70,7 @@ module Marc
       # @return [Array<String>] subfields to be extracted as meeting role terms
       #   from {meeting_data_tags}
       setting :meeting_role_term_subfields,
-        default: %w[e],
+        default: %w[j],
         reader: true
       # @return [Symbol] field in which to write the name value when
       #   converting MARC data to CSV row when extracting names

lib/kiba/extend/transforms/marc/extract_meeting_name_data.rb

@@ -12,14 +12,15 @@ module Marc
         # @example
         #   # =001  008000714-7
         #   # =711  2\$aAssociation of Child Psychology Annual Conference.
-        #   #          $esponsor$evenue
+        #   #          $eSubcommittee.$jsponsor$jvenue
         #   rec = get_marc_record(index: 6)
         #   xform = Marc::ExtractMeetingNameData.new
         #   results = []
         #   xform.process(rec){ |row| results << row }
         #   expect(results.length).to eq(1)
         #   row = {:sourcefield=>"711",
-        #          :name=>"Association of Child Psychology Annual Conference",
+        #          :name=>"Association of Child Psychology Annual "\
+        #          "Conference. Subcommittee",
         #          :nametype=>"meeting", :role_code=>"",
         #          :role_term=>"sponsor|venue", :marcid=>"008000714-7"}
         #   expect(results.first).to eq(row)

lib/kiba/extend/transforms/marc/extract_subfields_from_field.rb

@@ -6,7 +6,6 @@ module Kiba
   module Extend
     module Transforms
       module Marc
-        # rubocop:disable Layout/LineLength
         # For each occurrence of given field tag, outputs a row with the
         #   following columns: marcid, fullfield, and one column per specified
         #   subfield. If there are more than one occurrnces of a subfield
@@ -24,14 +23,16 @@ module Marc
         #   xform.process(rec){ |row| results << row }
         #   expect(results.length).to eq(1)
         #   first = {
-        #     :full260=>"260    $a Lahore : $b Zia-ul-Qurʾaan Publications, $c 1996. ",
+        #     :full260=>"260    $a Lahore : $b Zia-ul-Qurʾaan Publications, "\
+        #       "$c 1996. ",
         #     :_260a=>"Lahore :", :_260b=>"Zia-ul-Qurʾaan Publications,",
         #     :_260e=>nil, :_260f=>nil, :marcid=>"008000103-3"
         #   }
         #   expect(results[0]).to eq(first)
         # @example
         #   # =001  008000411-3
-        #   # =260  \\$aSan Jose, Calif. ;$aNew York, NY :$bH.M. Gousha Co.,$c[1986?]
+        #   # =260  \\$aSan Jose, Calif. ;$aNew York, NY :$bH.M. Gousha
+        #   #           Co.,$c[1986?]
         #   rec = get_marc_record(index: 3)
         #   xform = Marc::ExtractSubfieldsFromField.new(
         #     tag: '260', subfields: %w[a b e f]
@@ -40,27 +41,31 @@ module Marc
         #   xform.process(rec){ |row| results << row }
         #   expect(results.length).to eq(1)
         #   first = {
-        #     :full260=>"260    $a San Jose, Calif. ; $a New York, NY : $b H.M. Gousha Co., $c [1986?] ",
+        #     :full260=>"260    $a San Jose, Calif. ; $a New York, NY : $b "\
+        #       "H.M. Gousha Co., $c [1986?] ",
         #     :_260a=>"San Jose, Calif. ;|New York, NY :",
         #     :_260b=>"H.M. Gousha Co.,",
         #     :_260e=>nil, :_260f=>nil, :marcid=>"008000411-3"
         #   }
         #   expect(results[0]).to eq(first)
         #
         # @since 4.0.0
-        # rubocop:enable Layout/LineLength
         class ExtractSubfieldsFromField
           include FieldLinkable
           # @param tag [String] MARC tag from which to extract subfield values
           # @param subfields [Array<String>] subfield codes from which to
           #   extract values
+          # @param indicators [Boolean] whether to output indicators as separate
+          #   field values
+          # @param id_target [Symbol] name of field in which to write id value
           # @param delim [String] used when joining multiple values from
           #   recurring subfield
-          def initialize(tag:, subfields:,
+          def initialize(tag:, subfields:, indicators: false,
             id_target: Kiba::Extend::Marc.id_target_field,
             delim: Kiba::Extend.delim)
             @tag = tag
             @subfields = subfields
+            @indicators = indicators
             @id_target = id_target
             @delim = delim
             @idextractor = Kiba::Extend::Utils::MarcIdExtractor.new
@@ -84,14 +89,19 @@ def process(record)
 
           private
 
-          attr_reader :tag, :subfields, :id_target, :delim, :idextractor
+          attr_reader :tag, :subfields, :indicators, :id_target, :delim,
+            :idextractor
 
           def prepare_rows(fields, idhash)
             fields.map { |fld| prepare_row(fld, idhash) }
           end
 
           def prepare_row(field, idhash)
             row = {"full#{tag}".to_sym => field.to_s}.merge(idhash)
+            if indicators
+              row[:i1] = field.indicator1
+              row[:i2] = field.indicator2
+            end
             subfields.each do |code|
               row["_#{tag}#{code}".to_sym] = sf_val(field, code)
             end

lib/kiba/extend/transforms/marc/language_code_lookup.rb

@@ -0,0 +1,55 @@
+# frozen_string_literal: true
+
+require "net/http"
+require "uri"
+
+module Kiba
+  module Extend
+    module Transforms
+      module Marc
+        # Look up MARC List for Languages codes and provide preferred
+        # label value
+        #
+        # @example Known code
+        #   row = {code: "eng"}
+        #   result = Marc::LanguageCodeLookup.new(source: :code).process(row)
+        #   expect(result).to eq(row.merge({language: "English"}))
+        #
+        # @example Unknown code
+        #   row = {code: "foo"}
+        #   result = Marc::LanguageCodeLookup.new(source: :code).process(row)
+        #   expect(result).to eq(row.merge({language: nil}))
+        class LanguageCodeLookup
+          # @param source [Symbol] row field containing language code to
+          #   look up
+          # @param target [Symbol] row field into which language label value
+          #   will be written
+          def initialize(source:, target: :language)
+            @source = source
+            @target = target
+            @host = URI.parse("https://id.loc.gov").hostname
+          end
+
+          # @param row [Hash{ Symbol => String, nil }]
+          # @return [Hash{ Symbol => String, nil }]
+          def process(row)
+            Net::HTTP.start(host, use_ssl: true) do |http|
+              row[target] = api_result(http, row[source])
+            end
+            row
+          end
+
+          private
+
+          attr_reader :source, :target, :host
+
+          def api_result(http, code)
+            return if code.blank?
+
+            http.head("/vocabulary/languages/#{code}")["x-preflabel"]
+          end
+        end
+      end
+    end
+  end
+end