traject 3.8.0 → 3.8.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9e9868f3b83385402413a2fc6c8865dc6ab3dd3776c25c0f1f1b88bf20024005
-  data.tar.gz: f86c298c93905948ca9425983e65811b2091ba39d6105895744df183a7c43695
+  metadata.gz: 5e46b17b653fcdad34dbd5f7cf1d203c8d63a5c43eed9691bdb13b1f83d737a5
+  data.tar.gz: e2c7c086ffa93c167dd03be87953a9b2fd5b852ee1ff5695a9269da989c176f7
 SHA512:
-  metadata.gz: 7d1e1122020632ac10d4da030915e0f710c8dd9bb6e9780089129c1ec7febb76f7b0e23ac4828d1e4860429164fe6052c8db768d478ae6102f4819b8cd512f4d
-  data.tar.gz: '08eead90c2ddfebe141aa4bea2f280878c27af35509a9f5b220d2899fad820f9954dbc30392aa3cfd00b79e77b8f5266c69ac7101fe381d5e777e47b2e12aca7'
+  metadata.gz: d43267258736b94dcb8befdeec486892195239144855bb3afd7756e9d626af44c3cb511b7fab5da5328f4844ffc8c0bfa1dd9dc89295e96e3ac59e3cc39cc89d
+  data.tar.gz: 304de8c8c6bcd89d4cbc6d98bb3ff9a9a18e117d51df2c6869443c70c3c8dbea9d498fe92f04b192a42100ee979c4fe97ba91895ee5faa7645fa70f4166f09c1

data/.github/workflows/ruby.yml

@@ -12,7 +12,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        ruby: [ '2.4', '2.5', '2.6', '2.7', '3.0', '3.1', 'jruby-9.1', 'jruby-9.2' ]
+        ruby: [ '2.4', '2.5', '2.6', '2.7', '3.0', '3.1', '3.2', '3.3', 'jruby-9.2', 'jruby-9.3', 'jruby-9.4' ]
     name: Ruby ${{ matrix.ruby }}
     steps:
     - uses: actions/checkout@v2

data/CHANGES.md

@@ -4,7 +4,18 @@
 
 *
 
-*
+## 3.8.2
+
+Bug fix for the `#filing_version` logic, which was incorrectly assuming the 
+first subfield in a field would hold content (e.g., `$a`) and thus failed
+when it held a pointer to a linking field (e.g., `$6 245-01`)
+
+```
+
+
+## 3.8.1
+
+Ugh. Forgot about Jruby 9.1 problem with bundler 2. Changing the requirement back. 
 
 ## 3.8.0
 

data/lib/tasks/load_maps.rake

@@ -1,20 +1,16 @@
 require 'net/http'
 require 'open-uri'
+require 'csv'
 
 
 
+CODELIST_NS = 'info:lc/xmlns/codelist-v1'
 
 namespace :load_maps do
 
   desc "Load MARC geo codes by screen-scraping LC"
-  task :marc_geographic do
-    begin
-      require 'nokogiri'
-    rescue LoadError => e
-      $stderr.puts "\n  load_maps:marc_geographic task requires nokogiri"
-      $stderr.puts "  Try `gem install nokogiri` and try again. Exiting...\n\n"
-      exit 1
-    end
+  task :marc_geographic do |task|
+    require_nokogiri(task)
 
     source_url = "http://www.loc.gov/marc/geoareas/gacs_code.html"
 
@@ -45,4 +41,51 @@ namespace :load_maps do
     end
     $stderr.puts "Done."
   end
+
+  desc "Load MARC language codes from LOC and SIL"
+  task :marc_languages do |task|
+    require_nokogiri(task)
+    filename = ENV["OUTPUT_TO"] || File.expand_path("../../translation_maps/marc_languages.yaml", __FILE__)
+    file = File.open(filename, "w:utf-8")
+    $stderr.puts "Writing to `#{filename}` ..."
+    file.puts("# Map Language Codes (in 008[35-37], 041) to User Friendly Term\r")
+
+    marc_language_source_url = 'https://www.loc.gov/standards/codelists/languages.xml'
+    doc = Nokogiri::XML(URI.parse(marc_language_source_url).open)
+    marc_language_hash = doc.xpath('//codelist:language', codelist: CODELIST_NS)
+                            .to_h do |node|
+                              [node.xpath('./codelist:code/text()', codelist: CODELIST_NS).to_s,
+                               node.xpath('./codelist:name/text()', codelist: CODELIST_NS).to_s]
+                            end.reject { |key, _val| %w[und zxx].include? key }
+
+    file.puts "\r"
+    file.puts("# MARC language codes (including obsolete codes), from #{marc_language_source_url}\r\n")
+    marc_language_hash.sort_by { |k, _v| k }.each do |key, value|
+      file.puts("#{key}: #{escape_special_yaml_chars(value)}\r")
+    end
+
+    iso_639_3_url = 'https://iso639-3.sil.org/sites/iso639-3/files/downloads/iso-639-3.tab'
+    parsed_url = URI.parse(iso_639_3_url)
+    iso_languages = CSV.parse(parsed_url.read(encoding: 'UTF-8'), headers: true, col_sep: "\t", encoding: "UTF-8")
+    iso_language_hash = iso_languages.to_h { |row| [row['Id'], row['Ref_Name']] }
+                                     .reject { |key, _val| %w[und zxx].include? key }
+                                     .reject { |key, _val| marc_language_hash.keys.include? key }
+    file.puts "\r"
+    file.puts("# ISO 639-3 codes, from #{iso_639_3_url}\r")
+    iso_language_hash.sort_by { |k, _v| k }.each do |key, value|
+      file.puts("#{key}: #{escape_special_yaml_chars(value)}\r")
+    end
+  end
+
+  def require_nokogiri(task)
+    require 'nokogiri'
+  rescue LoadError
+    $stderr.puts "\n  #{task&.name} task requires nokogiri"
+    $stderr.puts "  Try `gem install nokogiri` and try again. Exiting...\n\n"
+    exit 1
+  end
+
+  def escape_special_yaml_chars(string)
+    string.match(/[\,\']/) ? %Q{"#{string}"} : string
+  end
 end

data/lib/traject/macros/marc21_semantics.rb

@@ -167,7 +167,13 @@ module Traject::Macros
       # (b) include the first subfield in the record
 
       subs = spec.subfields
-      return str unless subs && subs.include?(field.subfields[0].code)
+
+      # Get the code for the first alphabetic subfield, which would be
+      # the one getting characters shifted off
+
+      first_alpha_code = field.subfields.first{|sf| sf.code =~ /[a-z]/}.code
+
+      return str unless subs && subs.include?(first_alpha_code)
 
       # OK. If we got this far we actually need to strip characters off the string
 

data/lib/traject/version.rb

@@ -1,3 +1,3 @@
 module Traject
-  VERSION = "3.8.0"
+  VERSION = "3.8.2"
 end