relaton-doi 2.1.3 → 2.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 953aa1acd54cd1a21fb180ba132e24da1a1f9bb310496bfc4bc454e6d8d73621
-  data.tar.gz: 3a60c3a21e9eb9b1b32bc62a57bdea14f8efbad2beddf490648dbc073c63cf7d
+  metadata.gz: 1f985958ac631dbcb90d8d435d4cd8d771ef1eb661096bd62d0aa05ea0739291
+  data.tar.gz: f0e7a7392db7ff8e082682c92afa15eabdad7230d35cb96335068f5ddaebd14f
 SHA512:
-  metadata.gz: 3061d50b8b594725c9ae4ca531fe593ccda1d9825a51ea6fce7b046367684920e068470fbb8b95f3ef29a76469f3b5f9281938bf9adffadd573c046bfc9a526b
-  data.tar.gz: 4a04be0cdde3781c793db668665c5a4d2f66530af19b505c646a9bb535a9276ab9984b09efee690aad0ff79a32ec128f6f48b5021cfabed7ac204783e18b5ec2
+  metadata.gz: e1474c8687f0772f77aa5a29cd5394722c12700a041bdb70cfab7fd2f84cb36d16f4cee676de366d262a2344075197b2fd6b18ea615c373495fc83f94756591d
+  data.tar.gz: fd1d94c456aed217671bd968d60815f5e06ed5d87bad05efd950f04fe43e6e1a46cbb1c8f8c28f8a57f90520894c384de29a29da7fd1a051533342ea5eac6fb8

data/CLAUDE.md

@@ -43,9 +43,19 @@ Key classes in `lib/relaton/doi/`:
 - `TYPES` — maps 23 Crossref document types to Relaton types (e.g., `"book-chapter"` → `"inbook"`)
 - `REALATION_TYPES` — maps 37 Crossref relation types to Relaton relation types
 - `COUNTRIES` — `%w[USA]`, used by `parse_place` to distinguish country vs region
-- `TAG_RE`, `NS_PREFIX_RE`, `NS_PLACEHOLDER`, `SAVE_OPTS` — used by `normalize_markup` and
-  `drop_namespaces` to remove the `jats:` and `xlink:` prefixes that Crossref puts on
-  abstract and title markup. Relaton never declares those prefixes, so the output would
-  otherwise fail every namespace-aware parser downstream, including relaton-render
-  (metanorma-pdfa#99). The helper declares each prefix on a wrapper element, removes the
-  namespaces with Nokogiri, and returns the content unchanged when it does not parse.
+
+## Crossref Markup Handling
+
+`Parser#normalize_markup` decodes the HTML entities of raw Crossref content. Crossref
+sometimes returns the JATS markup entity-encoded, and the relaton-bib sanitizer detects a
+tag by a real `<`, so it would treat the encoded form as plain text.
+
+`Bib::Title` and `Bib::Abstract` are `LocalizedMarkedUpString` subclasses, so their
+`content=` setter runs `Relaton::Bib::Sanitizer` on assignment. Since relaton-bib 2.1.9 that
+sanitizer also removes the `jats:` and `xlink:` namespace prefixes. Relaton never declares
+those prefixes, so leaving them in place made the output fail every namespace-aware parser
+downstream, including relaton-render (metanorma-pdfa#99). relaton-doi carried its own prefix
+stripping until 2.1.9 shipped that fix, which is why the gemspec pins `~> 2.1.9`.
+
+The sanitizer removes only the prefixes it could not resolve. A namespace that the content
+declares itself survives, such as the MathML `xmlns` inside a `<stem>`.

data/lib/relaton/doi/parser.rb

@@ -73,22 +73,6 @@ module Relaton
       ATTRS = %i[type fetched title docidentifier date source abstract contributor place
                  ext relation extent series medium].freeze
 
-      # Matches content that looks like markup. Same shape as the tag detector
-      # of `Relaton::Bib::Sanitizer`, so both agree on what to parse.
-      TAG_RE = %r{<[a-zA-Z/!?]}
-
-      # Captures a namespace prefix, on a tag or on an attribute: the `jats` of
-      # `<jats:p>` and `</jats:italic>`, and the `xlink` of `xlink:href`.
-      NS_PREFIX_RE = %r{(?:</?|\s)([A-Za-z_][\w.-]*):(?=[A-Za-z_])}
-
-      # Placeholder namespace, used only to make an undeclared prefix parseable.
-      NS_PLACEHOLDER = "urn:x-relaton-doi:%s".freeze
-
-      # Serialize without the FORMAT option, so the round trip through the
-      # parser does not add indentation to the content.
-      SAVE_OPTS = Nokogiri::XML::Node::SaveOptions::AS_XML |
-        Nokogiri::XML::Node::SaveOptions::NO_DECLARATION
-
       CROSSREF_API_URL = "https://api.crossref.org/works?query=%{query}&filter=%{filter}".freeze
       MAX_RETRIES = 3
 
@@ -230,62 +214,22 @@ module Relaton
       end
 
       #
-      # Prepare raw Crossref markup for the relaton-bib sanitizer.
+      # Decode the HTML entities of raw Crossref content.
+      #
+      # Crossref sometimes returns the JATS markup as HTML entities. The
+      # relaton-bib sanitizer detects a tag by a real `<`, so it treats the
+      # encoded form as plain text and leaves it alone. Decode the entities
+      # here, and the sanitizer maps the markup to the basicdoc set.
       #
-      # Crossref returns JATS markup. It sometimes encodes the markup as HTML
-      # entities, and it prefixes the elements with the `jats:` namespace
-      # prefix. The prefix is never declared in the Relaton output, so a
-      # namespace-aware parser rejects the document, and the relaton-bib
-      # sanitizer skips content that it cannot parse. Decode the entities and
-      # remove the prefixes, so the sanitizer maps the elements to the
-      # basicdoc set. See metanorma-pdfa#99.
+      # The sanitizer removes the `jats:` and `xlink:` namespace prefixes
+      # itself since relaton-bib 2.1.9, so this method no longer does.
       #
       # @param [String] str The raw Crossref content.
       #
-      # @return [String] The content without entities and namespace prefixes.
+      # @return [String] The content without HTML entities.
       #
       def normalize_markup(str)
-        cnt = CGI.unescapeHTML(str)
-        cnt.match?(TAG_RE) ? drop_namespaces(cnt) : cnt
-      end
-
-      #
-      # Remove the namespace prefixes from markup.
-      #
-      # Declare every prefix that the content uses on a wrapper element, so the
-      # parser accepts the content, then let Nokogiri remove the namespaces from
-      # both the elements and the attributes. Return the content unchanged when
-      # it does not parse, which keeps the conservative behaviour of the
-      # relaton-bib sanitizer for text that only looks like markup.
-      #
-      # @param [String] cnt The markup.
-      #
-      # @return [String] The markup without namespace prefixes.
-      #
-      def drop_namespaces(cnt)
-        doc = parse_with_prefixes(cnt) or return cnt
-
-        doc.remove_namespaces!
-        doc.root.children.map do |c|
-          c.to_xml encoding: "UTF-8", save_with: SAVE_OPTS
-        end.join
-      end
-
-      #
-      # Parse markup that uses undeclared namespace prefixes.
-      #
-      # @param [String] cnt The markup.
-      #
-      # @return [Nokogiri::XML::Document, nil] The document, or nil when the
-      #   content uses no prefix or does not parse.
-      #
-      def parse_with_prefixes(cnt)
-        prefixes = cnt.scan(NS_PREFIX_RE).flatten.uniq - ["xmlns"]
-        return if prefixes.empty?
-
-        decl = prefixes.map { |p| %(xmlns:#{p}="#{format NS_PLACEHOLDER, p}") }
-        doc = Nokogiri::XML "<r #{decl.join ' '}>#{cnt}</r>"
-        doc if doc.errors.empty?
+        CGI.unescapeHTML(str)
       end
 
       #

data/lib/relaton/doi/version.rb

@@ -2,6 +2,6 @@
 
 module Relaton
   module Doi
-    VERSION = "2.1.3"
+    VERSION = "2.1.4"
   end
 end

data/relaton-doi.gemspec

@@ -34,7 +34,7 @@ Gem::Specification.new do |spec|
   # Uncomment to register a new dependency of your gem
   # spec.add_dependency "example-gem", "~> 1.0"
 
-  spec.add_dependency "relaton-bib", "~> 2.1.8"
+  spec.add_dependency "relaton-bib", "~> 2.1.9"
   spec.add_dependency "relaton-bipm", "~> 2.1.0"
   spec.add_dependency "relaton-ieee", "~> 2.1.0"
   spec.add_dependency "relaton-ietf", "~> 2.1.0"

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: relaton-doi
 version: !ruby/object:Gem::Version
-  version: 2.1.3
+  version: 2.1.4
 platform: ruby
 authors:
 - Ribose Inc.
@@ -16,14 +16,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.1.8
+        version: 2.1.9
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.1.8
+        version: 2.1.9
 - !ruby/object:Gem::Dependency
   name: relaton-bipm
   requirement: !ruby/object:Gem::Requirement