relaton-bib 2.0.0 → 2.1.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f0dda50ff6033e0640ffeb6e18596373c644c9704206159f4447a5ccc7f8ec4b
-  data.tar.gz: 2d0d134b521224759c9636fbb3aff0f283ec6028a22e8bc0e554aa3795dcdc1a
+  metadata.gz: 15d36c3ed1e4ae0042f96a06a6320dccbe089a87a691b77b8ad14e276217bd22
+  data.tar.gz: d6db31f185b72668da0c289bc462a62d1cfd1b9dda542f37b9dee7257d5282b4
 SHA512:
-  metadata.gz: cfaa4b06692c17d90018c4e54f5efaa91be82f49be3008730a46814433bf2f2dcd7ff3541b2369bf898401b620bbd19adad1752232e3b5a6eeb1ec3ab2914a38
-  data.tar.gz: ff217c62b41988af9300bf4dfafd1c469d76254dd2c9402199b0d88418c23c8d7e020ca069b4074cb1d0909fdb57abdd8a0391eb855b9d8d1528548050313caf
+  metadata.gz: 304fcb5b02e2c019d532164228acede4ed81c210eafededbbdd9446ac94fcd32d142be5be7689db09bc17b7aa3e6cfb333b892220ba36e64fcccc2bdda2df29c
+  data.tar.gz: 390cf8b808ab85b099d573b4bea833d265a2326d81c00936e5cebe6ae79c9243991843d44a46d3d24e403137f1ec27b433acc532c51f99a0d5e91d0ed66d6744

data/.github/workflows/integration-tests.yml

@@ -26,7 +26,7 @@ jobs:
 
     - uses: ruby/setup-ruby@v1
       with:
-        ruby-version: '3.1'
+        ruby-version: '3.3'
         bundler-cache: true
 
     - uses: actions/checkout@v3

data/.github/workflows/rake.yml

@@ -4,7 +4,7 @@ name: rake
 
 on:
   push:
-    branches: [ master, main ]
+    branches: [ master, main, lutaml-integration ]
     tags: [ v* ]
   pull_request:
 

data/CLAUDE.md

@@ -37,6 +37,18 @@ bin/console
 
 The codebase uses [lutaml-model](https://github.com/lutaml/lutaml-model) for serialization. Each model class in `lib/relaton/bib/model/` inherits from `Lutaml::Model::Serializable` and declares attributes with XML/YAML/JSON mappings.
 
+### Constant loading (autoload)
+
+Model and converter constants are wired via Ruby `autoload` from a central manifest,
+[lib/relaton/bib/model_autoload.rb](lib/relaton/bib/model_autoload.rb). This makes load order
+irrelevant and avoids circular requires between mutually-referencing classes (e.g. `OrganizationType`
+↔ `Subdivision`). **When adding a new model class, add an `autoload` entry to the manifest** rather
+than a `require_relative` in a sibling file. The manifest also eagerly loads `lutaml/model`/`lutaml/xml`
+and runs the `Lutaml::Model::Config.configure` block so the XML adapter is set regardless of which
+class is referenced first. Exceptions that keep `require_relative`: files that only *reopen* an
+already-defined module to add methods (the converter method-partials under
+`lib/relaton/bib/converter/*/`), which `autoload` cannot express.
+
 ### Core Classes
 
 **`Relaton::Bib::Item`** ([lib/relaton/bib/model/item.rb](lib/relaton/bib/model/item.rb)) - The main serialization class defining all bibliographic attributes and their XML mappings. Uses `ItemData` as its underlying model.

data/lib/relaton/bib/converter/asciibib/to_asciibib.rb

@@ -123,8 +123,8 @@ module Relaton
           def render_version(ver, prefix, count)
             pref = prefix.empty? ? prefix : "#{prefix}."
             out = count > 1 ? "#{prefix}version::\n" : ""
-            out << "#{pref}version.revision_date:: #{ver.revision_date}\n" if ver.revision_date
-            out << "#{pref}version.draft:: #{ver.draft}\n" if ver.draft
+            out << "#{pref}version:: #{ver.content}\n" if ver.content
+            out << "#{pref}version.type:: #{ver.type}\n" if ver.type
             out
           end
 

data/lib/relaton/bib/converter/bibxml/from_rfcxml.rb

@@ -1,3 +1,5 @@
+require "cgi"
+
 module Relaton
   module Bib
     module Converter
@@ -138,7 +140,8 @@ module Relaton
             return [] unless @reference.front.abstract&.t
 
             @reference.front.abstract.t.map do |t|
-              Abstract.new(content: t.content, language: "en", script: "Latn")
+              text = CGI.escapeHTML(t.content.join.strip)
+              Abstract.new(content: "<p>#{text}</p>", language: "en", script: "Latn")
             end
           end
 
@@ -218,7 +221,7 @@ module Relaton
             org = author.organization
             return if org.nil? || org.content.empty?
 
-            name = ORGNAMES[org.content] || org.content
+            name = ORGNAMES[org.content.join] || org.content.join
             orgname = TypedLocalizedString.new(content: name, language: "en")
             abbrev = LocalizedString.new(content: org.abbrev, language: "en") if org.abbrev
             Organization.new(

data/lib/relaton/bib/converter/bibxml/to_rfcxml.rb

@@ -1,3 +1,5 @@
+require "cgi"
+
 module Relaton
   module Bib
     module Converter
@@ -120,7 +122,7 @@ module Relaton
             abbrev = org.abbreviation&.content
             orgname = resolve_orgname(org, abbrev)
             Rfcxml::V3::Organization.new(
-              content: orgname, abbrev: abbrev,
+              content: [orgname], abbrev: abbrev,
             )
           end
 
@@ -282,9 +284,10 @@ module Relaton
             return unless @item.abstract.any?
 
             content = @item.abstract[0].content
-            Rfcxml::V3::Abstract.new(
-              t: [Rfcxml::V3::Text.new(content: content)],
-            )
+            paragraphs = content.scan(%r{<p>(.*?)</p>}m).flatten
+            paragraphs = [content] if paragraphs.empty?
+            ts = paragraphs.map { |p| Rfcxml::V3::Text.new(content: CGI.unescapeHTML(p)) }
+            Rfcxml::V3::Abstract.new(t: ts)
           end
 
           FORMAT_TYPES = %w[TXT HTML PDF XML DOC].freeze

data/lib/relaton/bib/hash_parser_v1.rb

@@ -196,10 +196,23 @@ module Relaton
       def version_hash_to_bib(ret)
         return unless ret[:version]
 
-        ret[:version] = array(ret[:version]).map do |v|
-          v[:revision_date] &&= parse_date(v[:revision_date], str: false)
-          Bib::Version.new(**v)
-        end
+        ret[:version] = array(ret[:version]).map { |v| version_entry(v) }
+      end
+
+      def version_entry(hash)
+        return Bib::Version.new(**hash) if hash[:content]
+
+        attrs = { content: legacy_version_content(hash) }
+        attrs[:type] = hash[:type] if hash[:type]
+        Bib::Version.new(**attrs)
+      end
+
+      def legacy_version_content(hash)
+        draft = hash[:draft]
+        revdate = hash[:revision_date]
+        return "#{draft} (#{revdate})" if draft && revdate
+
+        draft || revdate
       end
 
       def biblionote_hash_to_bib(ret)

data/lib/relaton/bib/model/bibdata.rb

@@ -1,23 +1,8 @@
 module Relaton
   module Bib
-    module BibdataShared
-      def self.included(base)
-        base.class_eval do
-          model ItemData
-
-          # Bibdata doesn't have id attribute.
-          attributes.delete :id
-          mappings[:xml].instance_variable_get(:@attributes).delete("id")
-
-          xml do
-            root "bibdata"
-          end
-        end
-      end
-    end
-
-    # This class represents a bibliographic item as a bibdata.
+    # Bibliographic item serialized as <bibdata>. Has ext, no id.
     class Bibdata < Item
+      model ItemData
       include BibdataShared
     end
   end

data/lib/relaton/bib/model/bibdata_shared.rb

@@ -0,0 +1,10 @@
+module Relaton
+  module Bib
+    module BibdataShared
+      def self.included(base)
+        base.xml { root "bibdata" }
+        ItemShared.prune_attribute(base, :id, "id")
+      end
+    end
+  end
+end

data/lib/relaton/bib/model/bibitem.rb

@@ -1,22 +1,8 @@
 module Relaton
   module Bib
-    module BibitemShared
-      def self.included(base)
-        base.class_eval do
-          model ItemData
-
-          # This class represents a bibliographic item as a bibitem.
-          attributes.delete :ext
-          mappings[:xml].instance_variable_get(:@elements).delete("ext")
-
-          xml do
-            root "bibitem"
-          end
-        end
-      end
-    end
-
+    # Bibliographic item serialized as <bibitem>. Has id, no ext.
     class Bibitem < Item
+      model ItemData
       include BibitemShared
     end
   end

data/lib/relaton/bib/model/bibitem_shared.rb

@@ -0,0 +1,10 @@
+module Relaton
+  module Bib
+    module BibitemShared
+      def self.included(base)
+        base.xml { root "bibitem" }
+        ItemShared.prune_attribute(base, :ext, "ext")
+      end
+    end
+  end
+end

data/lib/relaton/bib/model/contact.rb

@@ -1,7 +1,3 @@
-require_relative "address"
-require_relative "phone"
-require_relative "uri"
-
 module Relaton
   module Bib
     module Contact

data/lib/relaton/bib/model/contributor.rb

@@ -16,56 +16,14 @@ module Relaton
       end
 
       attribute :role, Role, collection: true, initialize_empty: true
-      # attribute :entity, ContributionInfo
-      # @TODO: use `import_model ContributionInfo` for person and organization when Lutaml supports it
-      choice(min: 1, max: 1) do
-        attribute :person, Person
-        attribute :organization, Organization
-      end
-      # import_model_attributes ContributionInfo
+      import_model_attributes ContributionInfo
 
       xml do
         root "contributor"
 
         map_element "role", to: :role
-        # map_content to: :entity, with: { from: :entity_from_xml, to: :entity_to_xml }
-        map_element "person", to: :person
-        map_element "organization", to: :organization # , with: { from: :organization_from_xml, to: :organization_to_xml }
+        import_model_mappings ContributionInfo
       end
-
-      # def entity
-      #   person || organization
-      # end
-
-      # def entity=(value)
-      #   if value.is_a? Person
-      #     self.person = value
-      #     self.organization = nil
-      #   elsif value.is_a? Organization
-      #     self.organization = value
-      #     self.person = nil
-      #   else
-      #     raise ArgumentError, "value must be a Person or Organization"
-      #   end
-      # end
-
-      # def organization_from_xml(model, node)
-      #   model.entity = Organization.of_xml node
-      # end
-
-      # def organization_to_xml(model, parent, _doc)
-      #   parent << model.entity.to_xml
-      # end
-
-      # def entity_from_xml(model, node)
-      #   n = node.instance_variable_get(:@node) || node
-      #   model.content = ContributionInfo.of_xml n
-      # rescue StandardError
-      # end
-
-      # def entity_to_xml(model, parent, _doc)
-      #   parent << model.content.to_xml
-      # end
     end
   end
 end

data/lib/relaton/bib/model/date.rb

@@ -1,5 +1,3 @@
-require_relative "type/string_date"
-
 module Relaton
   module Bib
     class Date < Lutaml::Model::Serializable

data/lib/relaton/bib/model/docidentifier.rb

@@ -6,6 +6,7 @@ module Relaton
       attribute :primary, :boolean
 
       xml do
+        root "docidentifier"
         map_attribute "type", to: :type
         map_attribute "scope", to: :scope
         map_attribute "primary", to: :primary

data/lib/relaton/bib/model/ext.rb

@@ -1,11 +1,7 @@
-require_relative "doctype"
-require_relative "ics"
-require_relative "structured_identifier"
-
 module Relaton
   module Bib
     class Ext < Lutaml::Model::Serializable
-      attribute :schema_version, :string
+      attribute :schema_version, :string, method: :get_schema_version
       attribute :doctype, Doctype
       attribute :subdoctype, :string
       attribute :flavor, :string
@@ -14,13 +10,26 @@ module Relaton
 
       xml do
         root "ext"
-        map_attribute "schema-version", to: :schema_version
+        map_attribute "schema-version", to: :schema_version, render_default: true
         map_element "doctype", to: :doctype
         map_element "subdoctype", to: :subdoctype
         map_element "flavor", to: :flavor
         map_element "ics", to: :ics
         map_element "structuredidentifier", to: :structuredidentifier
       end
+
+      key_value do
+        map_element "schema_version", to: :schema_version, render_default: true
+        map_element "doctype", to: :doctype
+        map_element "subdoctype", to: :subdoctype
+        map_element "flavor", to: :flavor
+        map_element "ics", to: :ics
+        map_element "structuredidentifier", to: :structuredidentifier
+      end
+
+      # Base returns nil so direct use omits schema-version. Subclasses in
+      # downstream relaton gems override to return their own version.
+      def get_schema_version = nil
     end
   end
 end

data/lib/relaton/bib/model/ics.rb

@@ -12,11 +12,37 @@ module Relaton
         map_element "text", to: :text
       end
 
+      # Returns the explicit text if set, else the Isoics description for
+      # `code`. Kept for consumers that read `.text` directly.
       def text
-        val = @text
-        return val if val && !val.empty?
+        return @text if @text.is_a?(String) && !@text.empty?
 
-        Isoics.fetch(code)&.description if code
+        Isoics.fetch(code)&.description if code.is_a?(String) && !code.empty?
+      end
+
+      # When code is assigned, eagerly populate text from Isoics if no
+      # explicit text has been set. Going through the public writer
+      # registers the value with the lutaml-model `value_set_for` tracker
+      # so the attribute is emitted on serialization.
+      def code=(val)
+        super
+        return unless val.is_a?(String) && !val.empty?
+        return if @text.is_a?(String) && !@text.empty?
+
+        description = Isoics.fetch(val)&.description
+        self.text = description if description
+      end
+
+      # When the deserializer reaches the end of the XML element and
+      # records that <text> was absent, it calls `using_default_for(:text)`
+      # to mark the attribute as default-valued (suppressing serialization).
+      # Refuse that mark if we've already populated text from Isoics so the
+      # value survives round-trip. See #112.
+      def using_default_for(attribute_name)
+        return if attribute_name == :text &&
+          @text.is_a?(String) && !@text.empty?
+
+        super
       end
     end
   end

data/lib/relaton/bib/model/item.rb

@@ -1,137 +1,44 @@
-require "lutaml/model"
-require "lutaml/model/xml_adapter/nokogiri_adapter"
-require_relative "localized_string_attrs"
-require_relative "localized_string"
-require_relative "formattedref"
-require_relative "abstract"
-require_relative "date"
-require_relative "locality"
-require_relative "locality_stack"
-require_relative "image"
-require_relative "title"
-require_relative "docidentifier"
-require_relative "note"
-require_relative "full_name_type"
-require_relative "fullname"
-require_relative "contact"
-require_relative "logo"
-require_relative "organization"
-require_relative "affiliation"
-require_relative "person"
-require_relative "contribution_info"
-require_relative "contributor"
-require_relative "edition"
-require_relative "version"
-require_relative "status"
-require_relative "copyright"
-require_relative "place"
-require_relative "series"
-require_relative "medium"
-require_relative "uri"
-require_relative "price"
-require_relative "extent"
-require_relative "size"
-require_relative "keyword"
-require_relative "validity"
-require_relative "depiction"
-require_relative "source_locality_stack"
-require_relative "ext"
-
-Lutaml::Model::Config.configure do |config|
-  config.xml_adapter = Lutaml::Model::Xml::NokogiriAdapter
-end
-
+# Model classes are wired via autoload (see model_autoload.rb); the lutaml
+# requires and XML adapter configuration live there so they run eagerly,
+# independent of which model class is referenced first.
 module Relaton
   module Bib
-    class Relation < Lutaml::Model::Serializable
-    end
-
     # Item class repesents bibliographic item metadata.
+    # Used for YAML/JSON parsing and as the XML dispatch entry point.
     class Item < Lutaml::Model::Serializable
+      include NamespaceHelper
+
       attr_accessor :type # in some cases mehod type is unavailable
 
       model ItemData
 
+      def self.from_xml(xml, options = {})
+        return super unless self == namespace::Item
+
+        # lutaml-model has no built-in dispatch on root element name
+        # (polymorphic_map only works on attribute discriminators), so we
+        # peek at the root tag with Nokogiri and forward to the right class.
+        root_name = Nokogiri::XML(xml.to_s).root&.name
+        klass = root_name == "bibdata" ? namespace::Bibdata : namespace::Bibitem
+        klass.from_xml(xml, options)
+      end
+
       attribute :id, :string
-      attribute :type, :string, values: %W[
-        article book booklet manual proceedings presentation thesis techreport standard
-        unpublished map electronic\sresource audiovisual film video boradcast software
-        graphic_work music patent inbook incollection inproceedings journal website
-        webresource dataset archival social_media alert message convesation misc
-      ]
-      attribute :schema_version, method: :get_schema_version
-      attribute :fetched, :date
-      attribute :formattedref, Formattedref
-      attribute :title, Title, collection: true, initialize_empty: true
-      attribute :source, Uri, collection: true, initialize_empty: true
-      attribute :docidentifier, Docidentifier, collection: true, initialize_empty: true
-      attribute :docnumber, :string
-      attribute :date, Date, collection: true, initialize_empty: true
-      attribute :contributor, Contributor, collection: true, initialize_empty: true
-      attribute :edition, Edition
-      attribute :version, Version, collection: true, initialize_empty: true
-      attribute :note, Note, collection: true, initialize_empty: true
-      attribute :language, :string, collection: true, initialize_empty: true
-      attribute :locale, :string, collection: true, initialize_empty: true
-      attribute :script, :string, collection: true, initialize_empty: true
-      attribute :abstract, Abstract, collection: true, initialize_empty: true
-      attribute :status, Status
-      attribute :copyright, Copyright, collection: true, initialize_empty: true
-      attribute :relation, Relation, collection: true, initialize_empty: true
-      attribute :series, Series, collection: true, initialize_empty: true
-      attribute :medium, Medium
-      attribute :place, Place, collection: true, initialize_empty: true
-      attribute :price, Price, collection: true, initialize_empty: true
-      attribute :extent, Extent, collection: true, initialize_empty: true
-      attribute :size, Size
-      attribute :accesslocation, :string, collection: true, initialize_empty: true
-      attribute :license, :string, collection: true, initialize_empty: true
-      attribute :classification, Docidentifier, collection: true, initialize_empty: true
-      attribute :keyword, Keyword, collection: true, initialize_empty: true
-      attribute :validity, Validity
-      attribute :depiction, Depiction, collection: true, initialize_empty: true
+      attribute :schema_version, :string, method: :get_schema_version
+      attribute :fetched, PlainDate
+      instance_exec(&ItemShared::ATTRIBUTES)
       attribute :ext, Ext
 
-      xml do # rubocop:disable Metrics/BlockLength
+      xml do
         map_attribute "id", to: :id
         map_attribute "type", to: :type
-        map_attribute "schema-version", to: :schema_version
+        map_attribute "schema-version", to: :schema_version, render_default: true
         map_element "fetched", to: :fetched
-        map_element "formattedref", to: :formattedref
-        map_element "title", to: :title # , with: { from: :title_from_xml, to: :title_to_xml }
-        map_element "uri", to: :source
-        map_element "docidentifier", to: :docidentifier
-        map_element "docnumber", to: :docnumber
-        map_element "date", to: :date
-        map_element "contributor", to: :contributor
-        map_element "edition", to: :edition
-        map_element "version", to: :version
-        map_element "note", to: :note
-        map_element "language", to: :language
-        map_element "locale", to: :locale
-        map_element "script", to: :script
-        map_element "abstract", to: :abstract
-        map_element "status", to: :status
-        map_element "copyright", to: :copyright
-        map_element "relation", to: :relation # , with: { from: :relation_from_xml, to: :relation_to_xml }
-        map_element "series", to: :series
-        map_element "medium", to: :medium
-        map_element "place", to: :place
-        map_element "price", to: :price
-        map_element "extent", to: :extent
-        map_element "size", to: :size
-        map_element "accesslocation", to: :accesslocation
-        map_element "license", to: :license
-        map_element "classification", to: :classification
-        map_element "keyword", to: :keyword
-        map_element "validity", to: :validity
-        map_element "depiction", to: :depiction
+        instance_exec(&ItemShared::XML_BODY)
         map_element "ext", to: :ext
       end
 
-      def get_schema_version
-        Relaton.schema_versions["relaton-models"]
-      end
+      def get_schema_version = Relaton.schema_versions["relaton-models"]
     end
   end
 end

data/lib/relaton/bib/model/item_base.rb

@@ -1,27 +1,20 @@
 module Relaton
   module Bib
-    # Module to remove id, schema-version, fetched & ext attributes from Item subclasses.
-    # Used for bibitem/relation instances that don't need these attributes.
-    module ItemBaseAttributes
-      def self.included(base)
-        # we don't need id, schema-version & fetched attributes in relation/bibitem
-        base.mappings[:xml].instance_variable_get(:@attributes).delete("id")
-        base.mappings[:xml].instance_variable_get(:@attributes).delete("schema-version")
-        base.mappings[:xml].instance_variable_get(:@elements).delete("fetched")
-        base.mappings[:xml].instance_variable_get(:@elements).delete("ext")
-        base.attributes.delete :id
-        base.attributes.delete :schema_version
-        base.attributes.delete :fetched
-        base.attributes.delete :ext
-      end
-    end
+    # Bibliographic item used as a nested element inside Relation.
+    # Has neither id, schema_version, fetched, nor ext.
+    class ItemBase < Lutaml::Model::Serializable
+      include NamespaceHelper
+
+      attr_accessor :type
 
-    # The class is for relaton bibitem instances.
-    # The in relaton bibitem instances dosn't have schema-version & fetched attributes.
-    class ItemBase < Item
       model ItemData
 
-      include ItemBaseAttributes
+      instance_exec(&ItemShared::ATTRIBUTES)
+
+      xml do
+        map_attribute "type", to: :type
+        instance_exec(&ItemShared::XML_BODY)
+      end
     end
   end
 end