arclight 0.1.4 → 0.2.0

data/app/views/shared/_main_menu_links.html.erb

@@ -1,6 +1,6 @@
 <li class="nav-item <%= repositories_active_class %>">
   <%= link_to t('arclight.routes.repositories'), arclight_engine.repositories_path, class: 'nav-link' %>
 </li>
-<li class="nav-item <%= collection_active_class %>">
+<li class="nav-item ml-3 <%= collection_active_class %>">
   <%= link_to t('arclight.routes.collections'), arclight_engine.collections_path, class: 'nav-link' %>
 </li>

data/arclight.gemspec

@@ -23,19 +23,23 @@ Gem::Specification.new do |spec|
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ['lib']
 
-  spec.add_dependency 'blacklight', '7.0.0.rc1'
-  spec.add_dependency 'blacklight_range_limit', '7.0.0.rc2'
+  spec.add_dependency 'blacklight', '~> 7.0', '>= 7.0.1'
+  spec.add_dependency 'blacklight_range_limit', '~> 7.1'
   spec.add_dependency 'rails', '~> 5.0'
   spec.add_dependency 'solr_ead'
+  spec.add_dependency 'traject', '~> 3.0'
+  spec.add_dependency 'traject_plus'
 
-  spec.add_development_dependency 'bundler', '~> 1.14'
+  spec.add_development_dependency 'bundler', '> 1.14'
   spec.add_development_dependency 'capybara'
-  spec.add_development_dependency 'coveralls'
   spec.add_development_dependency 'engine_cart'
-  spec.add_development_dependency 'poltergeist'
+  spec.add_development_dependency 'i18n-tasks'
   spec.add_development_dependency 'rake', '~> 12.0'
-  spec.add_development_dependency 'rubocop', '~> 0.48.1'
-  spec.add_development_dependency 'rubocop-rspec', '~> 1.15.0'
+  spec.add_development_dependency 'rubocop', '~> 0.74.0'
+  spec.add_development_dependency 'rubocop-rspec', '~> 1.35'
   spec.add_development_dependency 'rspec-rails', '~> 3.0'
+  spec.add_development_dependency 'selenium-webdriver'
+  spec.add_development_dependency 'simplecov'
   spec.add_development_dependency 'solr_wrapper'
+  spec.add_development_dependency 'webdrivers'
 end

data/config/i18n-tasks.yml

@@ -0,0 +1,132 @@
+# i18n-tasks finds and manages missing and unused translations: https://github.com/glebm/i18n-tasks
+
+# The "main" locale.
+base_locale: en
+## All available locales are inferred from the data by default. Alternatively, specify them explicitly:
+# locales: [es, fr]
+## Reporting locale, default: en. Available: en, ru.
+# internal_locale: en
+
+# Read and write translations.
+data:
+  ## Translations are read from the file system. Supported format: YAML, JSON.
+  ## Provide a custom adapter:
+  # adapter: I18n::Tasks::Data::FileSystem
+
+  # Locale files or `File.find` patterns where translations are read from:
+  read:
+    ## Default:
+    # - config/locales/%{locale}.yml
+    ## More files:
+    - config/locales/**/*.%{locale}.yml
+
+  # Locale files to write new keys to, based on a list of key pattern => file rules. Matched from top to bottom:
+  # `i18n-tasks normalize -p` will force move the keys according to these rules
+  write:
+    ## For example, write devise and simple form keys to their respective files:
+    # - ['{devise, simple_form}.*', 'config/locales/\1.%{locale}.yml']
+    ## Catch-all default:
+    # - config/locales/%{locale}.yml
+
+  # External locale data (e.g. gems).
+  # This data is not considered unused and is never written to.
+  external:
+    ## Example (replace %#= with %=):
+    # - "<%#= %x[bundle show vagrant].chomp %>/templates/locales/%{locale}.yml"
+
+  ## Specify the router (see Readme for details). Valid values: conservative_router, pattern_router, or a custom class.
+  # router: conservative_router
+
+  yaml:
+    write:
+      # do not wrap lines at 80 characters
+      line_width: -1
+
+  ## Pretty-print JSON:
+  # json:
+  #   write:
+  #     indent: '  '
+  #     space: ' '
+  #     object_nl: "\n"
+  #     array_nl: "\n"
+
+# Find translate calls
+search:
+  ## Paths or `File.find` patterns to search in:
+  # paths:
+  #  - app/
+
+  ## Root directories for relative keys resolution.
+  # relative_roots:
+  #   - app/controllers
+  #   - app/helpers
+  #   - app/mailers
+  #   - app/presenters
+  #   - app/views
+
+  ## Files or `File.fnmatch` patterns to exclude from search. Some files are always excluded regardless of this setting:
+  ##   %w(*.jpg *.png *.gif *.svg *.ico *.eot *.otf *.ttf *.woff *.woff2 *.pdf *.css *.sass *.scss *.less *.yml *.json)
+  exclude:
+    - app/assets/images
+    - app/assets/fonts
+    - app/assets/videos
+
+  ## Alternatively, the only files or `File.fnmatch patterns` to search in `paths`:
+  ## If specified, this settings takes priority over `exclude`, but `exclude` still applies.
+  # only: ["*.rb", "*.html.slim"]
+
+  ## If `strict` is `false`, guess usages such as t("categories.#{category}.title"). The default is `true`.
+  # strict: true
+
+  ## Multiple scanners can be used. Their results are merged.
+  ## The options specified above are passed down to each scanner. Per-scanner options can be specified as well.
+  ## See this example of a custom scanner: https://github.com/glebm/i18n-tasks/wiki/A-custom-scanner-example
+
+## Translation Services
+# translation:
+#   # Google Translate
+#   # Get an API key and set billing info at https://code.google.com/apis/console to use Google Translate
+#   google_translate_api_key: "AbC-dEf5"
+#   # DeepL Pro Translate
+#   # Get an API key and subscription at https://www.deepl.com/pro to use DeepL Pro
+#   deepl_api_key: "48E92789-57A3-466A-9959-1A1A1A1A1A1A"
+
+## Do not consider these keys missing:
+ignore_missing:
+- '{blacklight}.*'
+
+## Consider these keys used:
+# ignore_unused:
+# - 'activerecord.attributes.*'
+# - '{devise,kaminari,will_paginate}.*'
+# - 'simple_form.{yes,no}'
+# - 'simple_form.{placeholders,hints,labels}.*'
+# - 'simple_form.{error_notification,required}.:'
+
+## Exclude these keys from the `i18n-tasks eq-base' report:
+# ignore_eq_base:
+#   all:
+#     - common.ok
+#   fr,es:
+#     - common.brand
+
+## Exclude these keys from the `i18n-tasks check-consistent-interpolations` report:
+# ignore_inconsistent_interpolations:
+# - 'activerecord.attributes.*'
+
+## Ignore these keys completely:
+# ignore:
+#  - kaminari.*
+
+## Sometimes, it isn't possible for i18n-tasks to match the key correctly,
+## e.g. in case of a relative key defined in a helper method.
+## In these cases you can use the built-in PatternMapper to map patterns to keys, e.g.:
+#
+# <%# I18n::Tasks.add_scanner 'I18n::Tasks::Scanners::PatternMapper',
+#       only: %w(*.html.haml *.html.slim),
+#       patterns: [['= title\b', '.page_title']] %>
+#
+# The PatternMapper can also match key literals via a special %{key} interpolation, e.g.:
+#
+# <%# I18n::Tasks.add_scanner 'I18n::Tasks::Scanners::PatternMapper',
+#       patterns: [['\bSpree\.t[( ]\s*%{key}', 'spree.%{key}']] %>

data/config/locales/arclight.en.yml

@@ -1,65 +1,66 @@
+---
 en:
   arclight:
-    masthead_heading: 'Archival Collections at Institution'
+    breadcrumb_separator: " » "
     date_range_histogram:
-      show_hide: 'Show/hide date distribution'
+      hide: Hide date distribution
+      show: Show date distribution
     hierarchy:
-      view_all: 'View'
-      scope_and_contents: 'Scope and Contents'
+      scope_and_contents: Scope and Contents
+      view_all: View
+    masthead_heading: Archival Collections at Institution
     request:
-      container: 'Request'
+      container: Request
     routes:
-      home: 'Home'
-      collections: 'Collections'
-      repositories: 'Repositories'
-      search_results: 'Search results'
+      collections: Collections
+      home: Home
+      repositories: Repositories
+      search_results: Search results
+    truncation:
+      view_less: view less ▼
+      view_more: view more ▶
     views:
       index:
-        number_of_children:
-          zero: 'No children'
-          one: '1 child'
-          other: '%{count} children'
         collection_search:
-          count: 'collection'
-      online_content_indicator: 'online content'
+          count: collection
+        number_of_children:
+          one: 1 child
+          other: "%{count} children"
+          zero: No children
+      online_content_indicator: online content
+      repositories:
+        number_of_collections:
+          one: 1 collection
+          other: "%{count} collections"
+          zero: No collections
+        view_all_collections: View all of our collections
+        view_more: View more
       show:
         collection_id: 'Collection ID: %{id}'
-        overview: 'Overview'
-        online_content: 'Online content'
-        no_online_content: 'No online content'
-        no_contents: 'No content inventory'
-        search_within: 'Search within this collection'
-        navigation_sidebar:
-          title: 'Navigation overview'
         context_sidebar:
-          in_person_field: 'In person'
-          terms_field: 'Terms & Conditions'
-          cite_field: 'How to cite this collection'
-          component_terms_field: "Terms & Conditions"
-        sections:
-          summary_field: 'Summary'
-          access_field: 'Access and Use'
-          background_field: 'Background'
-          scope_and_arrangement_field: 'Scope and Arrangement'
-          related_field: 'Related'
-          indexed_terms_field: 'Indexed Terms'
-          admin_info_field: 'Administrative Information'
-          component_field: 'About this %{level}'
-          component_indexed_terms_field: 'Indexed Terms'
-          collection_context_field: 'Collection Context'
-        our_collections: 'Our Collections'
+          cite_field: How to cite this collection
+          component_terms_field: Terms & Conditions
+          in_person_field: In person
+          terms_field: Terms & Conditions
         download:
-          default: 'Download'
-          pdf: 'Collection PDF (%{size})'
-          ead: 'Collection EAD (%{size})'
-      repositories:
-        number_of_collections:
-          zero: 'No collections'
-          one: '1 collection'
-          other: '%{count} collections'
-        view_more: 'View more'
-        view_all_collections: 'View all of our collections'
-    breadcrumb_separator: ' » '
-    truncation:
-      view_more: 'view more ▶'
-      view_less: 'view less ▼'
+          default: Download
+          ead: Collection EAD (%{size})
+          pdf: Collection PDF (%{size})
+        navigation_sidebar:
+          title: Navigation overview
+        no_contents: No content inventory
+        no_online_content: No online content
+        online_content: Online content
+        our_collections: Our Collections
+        overview: Overview
+        sections:
+          access_field: Access and Use
+          admin_info_field: Administrative Information
+          background_field: Background
+          collection_context_field: Collection Context
+          component_field: About this %{level}
+          component_indexed_terms_field: Indexed Terms
+          indexed_terms_field: Indexed Terms
+          related_field: Related
+          scope_and_arrangement_field: Scope and Arrangement
+          summary_field: Summary

data/lib/arclight/engine.rb

@@ -2,6 +2,7 @@
 
 require 'blacklight'
 require 'solr_ead'
+require 'traject'
 require 'arclight/exceptions'
 require 'arclight/normalized_date'
 require 'arclight/normalized_id'

data/lib/arclight/hash_absolute_xpath.rb

@@ -0,0 +1,57 @@
+# frozen_string_literal: true
+
+require 'digest'
+
+module Arclight
+  ##
+  # Take a Nokogiri node and get its absolute path (inserting our own indexes for component levels)
+  # and hash that outout. This is intended as a potential strategy for handling missing IDs in EADs.
+  class HashAbsoluteXpath
+    class << self
+      attr_writer :hash_algorithm
+
+      def hash_algorithm
+        return Digest::SHA1 unless defined? @hash_algorithm
+
+        @hash_algorithm
+      end
+    end
+
+    COMPONENT_NODE_NAME_REGEX = /^c\d{,2}$/.freeze
+    attr_reader :node
+    def initialize(node)
+      @node = node
+    end
+
+    def to_hexdigest
+      self.class.hash_algorithm.hexdigest(absolute_xpath)
+    end
+
+    def absolute_xpath
+      ancestor_tree = node.ancestors.map do |ancestor|
+        if ancestor.name =~ COMPONENT_NODE_NAME_REGEX
+          index = component_siblings_for_node(ancestor).index(ancestor)
+          "#{ancestor.name}#{index}"
+        else
+          ancestor.name
+        end
+      end
+
+      "#{[ancestor_tree.reverse, node.name].flatten.join('/')}#{current_index}"
+    end
+
+    private
+
+    def current_index
+      siblings.index(node)
+    end
+
+    def component_siblings_for_node(xml_node)
+      xml_node.parent.children.select { |n| n.name =~ COMPONENT_NODE_NAME_REGEX }
+    end
+
+    def siblings
+      @siblings ||= component_siblings_for_node(node)
+    end
+  end
+end

data/lib/arclight/missing_id_strategy.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+require 'arclight/hash_absolute_xpath'
+
+module Arclight
+  ##
+  # A class to configure a selected MissingIdStrategy.
+  # Defaults to Arclight::HashAbsoluteXpath
+  # This can be updated in an initializer to be any other class
+  class MissingIdStrategy
+    class << self
+      attr_writer :selected
+
+      def selected
+        return Arclight::HashAbsoluteXpath unless defined? @selected
+
+        @selected
+      end
+    end
+  end
+end

data/lib/arclight/normalized_date.rb

@@ -7,16 +7,21 @@ module Arclight
   # @see http://www2.archivists.org/standards/DACS/part_I/chapter_2/4_date
   class NormalizedDate
     # @param [String | Array<String>] `inclusive` from the `unitdate`
-    # @param [String] `bulk` from the `unitdate`
-    # @param [String] `other` from the `unitdate` when type is not specified
-    def initialize(inclusive, bulk = nil, other = nil)
-      if inclusive.is_a? Array # of YYYY-YYYY for ranges
-        @inclusive = YearRange.new(inclusive.include?('/') ? inclusive : inclusive.map { |v| v.tr('-', '/') }).to_s
-      elsif inclusive.present?
-        @inclusive = inclusive.strip
-      end
-      @bulk = bulk.strip if bulk.present?
-      @other = other.strip if other.present?
+    # @param [Array<String>] `bulk` from the `unitdate`
+    # @param [Array<String>] `other` from the `unitdate` when type is not specified
+    def initialize(inclusive, bulk = [], other = [])
+      @inclusive = (inclusive || []).map do |inclusive_text|
+        if inclusive_text.is_a? Array # of YYYY-YYYY for ranges
+          # NOTE: This code is not routable AFAICT in actual indexing.
+          # We pass arrays of strings (or xml nodes) here, and never a multidimensional array
+          year_range(inclusive_text)
+        elsif inclusive_text.present?
+          inclusive_text.strip
+        end
+      end&.join(', ')
+
+      @bulk = Array.wrap(bulk).compact.map(&:strip).join(', ')
+      @other = Array.wrap(other).compact.map(&:strip).join(', ')
     end
 
     # @return [String] the normalized title/date
@@ -28,6 +33,10 @@ module Arclight
 
     attr_reader :inclusive, :bulk, :other
 
+    def year_range(date_array)
+      YearRange.new(date_array.include?('/') ? date_array : date_array.map { |v| v.tr('-', '/') }).to_s
+    end
+
     # @see http://www2.archivists.org/standards/DACS/part_I/chapter_2/4_date for rules
     def normalize
       if inclusive.present?

data/lib/arclight/repository.rb

@@ -7,31 +7,14 @@ module Arclight
   class Repository
     include ActiveModel::Conversion # for to_partial_path
 
-    FIELDS = %i[name
-                description
-                visit_note
-                building
-                address1
-                address2
-                city
-                state
-                zip
-                country
-                phone
-                contact_info
-                thumbnail_url
-                google_request_url
-                google_request_mappings
-                collection_count].freeze
-
-    attr_accessor :slug, *FIELDS
+    attr_accessor :slug, :collection_count
 
     # @param [String] `slug` the unique identifier for the repository
     # @param [Hash] `data`
     def initialize(slug, data = {})
       @slug = slug
-      FIELDS.each do |field|
-        value = data[field.to_s]
+      data.each do |field, value|
+        self.class.attr_accessor field.to_sym
         send("#{field}=", value) if value.present?
       end
     end