moxml 0.5.171 → 0.5.172

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: dd5e790791e306829e87770a45424b006c9a535c95c825adfc7d0800816ceb7d
-  data.tar.gz: c3348f9fb7c6c0ccf00ab4bc16bba51b0812331d12d8478d2a77e09b8a600e58
+  metadata.gz: f8719e57bb76ceb63baa171d151028d8aff538fc351be7397b6ec5a492c8bdf1
+  data.tar.gz: 8fa85d23570f0588da0eb5d4ce5f83d59fb28e325d587f7210d3380a8afd5e33
 SHA512:
-  metadata.gz: 2972a94dc0bf75896d29432c1856b949090cbd8547b60c402376f2d715c743c8e029af2181531fbd327d6615a3a6aef5cb83144c5ff2b6389f35bd3ab6dfea59
-  data.tar.gz: c89e1be189130b17f0a7f61c67c9caba0624f589c29a2407effba7fb50da72304d33db77001185cb7b079729a7a29140d32e8a492269a58cf748a8e99f0276a2
+  metadata.gz: b93fc457a9d7435e2a55cb4fb5e3320d37b5949e77b184c640dd65a8b21954aab9987f204c8916411d9bfe6e11f7f0e358faa78f1ee9cfe3477f52b2d209e857
+  data.tar.gz: 78c66d6b3bdbb800b8cf23dde851235ab3a621bbb30fd0c413db06f9f856ba245b56f9a2c4a4d8220ca313b3bb2ef447d5737733387052bd82f2151cfd7bcdc5

data/Rakefile

@@ -350,4 +350,27 @@ namespace :docs do
   end
 end
 
+namespace :conformance do
+  desc "Export the cross-language conformance dataset to JSON (#496)"
+  task :export do
+    require "json"
+    require_relative "lib/moxml"
+    dataset = Moxml::XSD::BuiltinLexical::CONFORMANCE_CASES
+    payload = {
+      schema: 1,
+      generator: "moxml #{Moxml::VERSION}",
+      cases: {
+        "builtin_lexical" => dataset.map do |type, lexical, expected|
+          { "type" => type, "lexical" => lexical, "valid" => expected }
+        end,
+      },
+    }
+    dir = File.expand_path("conformance", __dir__)
+    FileUtils.mkdir_p(dir)
+    target = File.join(dir, "moxml-conformance.json")
+    File.write(target, "#{JSON.pretty_generate(payload)}\n")
+    puts "conformance: exported #{dataset.size} cases to #{target}"
+  end
+end
+
 task default: :spec

data/conformance/moxml-conformance.json

@@ -0,0 +1,413 @@
+{
+  "schema": 1,
+  "generator": "moxml 0.5.171",
+  "cases": {
+    "builtin_lexical": [
+      {
+        "type": "xs:token",
+        "lexical": "a  b",
+        "valid": true
+      },
+      {
+        "type": "xs:ID",
+        "lexical": " id ",
+        "valid": true
+      },
+      {
+        "type": "xs:NMTOKEN",
+        "lexical": "\tx ",
+        "valid": true
+      },
+      {
+        "type": "xs:ID",
+        "lexical": "1d",
+        "valid": false
+      },
+      {
+        "type": "xs:integer",
+        "lexical": "11111111111111111111111111111",
+        "valid": true
+      },
+      {
+        "type": "xs:positiveInteger",
+        "lexical": "11111111111111111111111111111",
+        "valid": true
+      },
+      {
+        "type": "xs:nonNegativeInteger",
+        "lexical": "11111111111111111111111111111",
+        "valid": true
+      },
+      {
+        "type": "xs:negativeInteger",
+        "lexical": "-100000000000000000000",
+        "valid": true
+      },
+      {
+        "type": "xs:nonPositiveInteger",
+        "lexical": "0",
+        "valid": true
+      },
+      {
+        "type": "xs:nonPositiveInteger",
+        "lexical": "-0",
+        "valid": true
+      },
+      {
+        "type": "xs:long",
+        "lexical": "9223372036854775807",
+        "valid": true
+      },
+      {
+        "type": "xs:long",
+        "lexical": "9223372036854775808",
+        "valid": false
+      },
+      {
+        "type": "xs:unsignedLong",
+        "lexical": "18446744073709551615",
+        "valid": true
+      },
+      {
+        "type": "xs:unsignedLong",
+        "lexical": "18446744073709551616",
+        "valid": false
+      },
+      {
+        "type": "xs:byte",
+        "lexical": "128",
+        "valid": false
+      },
+      {
+        "type": "xs:unsignedByte",
+        "lexical": "256",
+        "valid": false
+      },
+      {
+        "type": "xs:float",
+        "lexical": "+INF",
+        "valid": true
+      },
+      {
+        "type": "xs:double",
+        "lexical": "-INF",
+        "valid": true
+      },
+      {
+        "type": "xs:float",
+        "lexical": "NaN",
+        "valid": true
+      },
+      {
+        "type": "xs:float",
+        "lexical": "1.5E-3",
+        "valid": true
+      },
+      {
+        "type": "xs:decimal",
+        "lexical": "5.",
+        "valid": true
+      },
+      {
+        "type": "xs:decimal",
+        "lexical": ".5",
+        "valid": true
+      },
+      {
+        "type": "xs:decimal",
+        "lexical": "1e3",
+        "valid": false
+      },
+      {
+        "type": "xs:date",
+        "lexical": "2024-02-29",
+        "valid": true
+      },
+      {
+        "type": "xs:date",
+        "lexical": "1999-02-29",
+        "valid": false
+      },
+      {
+        "type": "xs:date",
+        "lexical": "2024-04-31",
+        "valid": false
+      },
+      {
+        "type": "xs:date",
+        "lexical": "2024-13-01",
+        "valid": false
+      },
+      {
+        "type": "xs:date",
+        "lexical": "not-a-date",
+        "valid": false
+      },
+      {
+        "type": "xs:dateTime",
+        "lexical": "2024-02-29T23:59:59Z",
+        "valid": true
+      },
+      {
+        "type": "xs:dateTime",
+        "lexical": "1999-02-29T00:00:00",
+        "valid": false
+      },
+      {
+        "type": "xs:dateTime",
+        "lexical": "not-a-date",
+        "valid": false
+      },
+      {
+        "type": "xs:dateTimeStamp",
+        "lexical": "2024-02-29T00:00:00Z",
+        "valid": true
+      },
+      {
+        "type": "xs:dateTimeStamp",
+        "lexical": "2024-02-29T00:00:00",
+        "valid": false
+      },
+      {
+        "type": "xs:time",
+        "lexical": "23:59:59.999+05:30",
+        "valid": true
+      },
+      {
+        "type": "xs:time",
+        "lexical": "24:00:00",
+        "valid": false
+      },
+      {
+        "type": "xs:time",
+        "lexical": "25:20:00",
+        "valid": false
+      },
+      {
+        "type": "xs:time",
+        "lexical": "13:20:60",
+        "valid": false
+      },
+      {
+        "type": "xs:gYearMonth",
+        "lexical": "2024-02Z",
+        "valid": true
+      },
+      {
+        "type": "xs:gYearMonth",
+        "lexical": "2024-13",
+        "valid": false
+      },
+      {
+        "type": "xs:gMonthDay",
+        "lexical": "--02-29",
+        "valid": true
+      },
+      {
+        "type": "xs:gMonthDay",
+        "lexical": "--02-30",
+        "valid": false
+      },
+      {
+        "type": "xs:gDay",
+        "lexical": "---31",
+        "valid": true
+      },
+      {
+        "type": "xs:gDay",
+        "lexical": "---32",
+        "valid": false
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "P1Y2M3DT4H5M6.5S",
+        "valid": true
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "PT0S",
+        "valid": true
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "-P1D",
+        "valid": true
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "P",
+        "valid": false
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "PT",
+        "valid": false
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "P1Y2MT",
+        "valid": false
+      },
+      {
+        "type": "xs:duration",
+        "lexical": "P200.5Y",
+        "valid": false
+      },
+      {
+        "type": "xs:dayTimeDuration",
+        "lexical": "P1DT2H",
+        "valid": true
+      },
+      {
+        "type": "xs:dayTimeDuration",
+        "lexical": "P1Y",
+        "valid": false
+      },
+      {
+        "type": "xs:yearMonthDuration",
+        "lexical": "P1Y2M",
+        "valid": true
+      },
+      {
+        "type": "xs:yearMonthDuration",
+        "lexical": "P1D",
+        "valid": false
+      },
+      {
+        "type": "xs:yearMonthDuration",
+        "lexical": "P1YT1H",
+        "valid": false
+      },
+      {
+        "type": "xs:base64Binary",
+        "lexical": "aGVsbG8=",
+        "valid": true
+      },
+      {
+        "type": "xs:base64Binary",
+        "lexical": "abcd",
+        "valid": true
+      },
+      {
+        "type": "xs:base64Binary",
+        "lexical": "abcde",
+        "valid": false
+      },
+      {
+        "type": "xs:base64Binary",
+        "lexical": "a b",
+        "valid": false
+      },
+      {
+        "type": "xs:hexBinary",
+        "lexical": "0A",
+        "valid": true
+      },
+      {
+        "type": "xs:hexBinary",
+        "lexical": "0AB",
+        "valid": false
+      },
+      {
+        "type": "xs:QName",
+        "lexical": "foo",
+        "valid": true
+      },
+      {
+        "type": "xs:QName",
+        "lexical": "ns:foo",
+        "valid": true
+      },
+      {
+        "type": "xs:QName",
+        "lexical": "xmlns:xsi",
+        "valid": false
+      },
+      {
+        "type": "xs:QName",
+        "lexical": ":foo",
+        "valid": false
+      },
+      {
+        "type": "xs:QName",
+        "lexical": "a:b:c",
+        "valid": false
+      },
+      {
+        "type": "xs:NOTATION",
+        "lexical": "foo",
+        "valid": true
+      },
+      {
+        "type": "xs:IDREFS",
+        "lexical": "a b",
+        "valid": true
+      },
+      {
+        "type": "xs:IDREFS",
+        "lexical": "a",
+        "valid": true
+      },
+      {
+        "type": "xs:IDREFS",
+        "lexical": "",
+        "valid": false
+      },
+      {
+        "type": "xs:NMTOKENS",
+        "lexical": "a.b-c:d e",
+        "valid": true
+      },
+      {
+        "type": "xs:ENTITIES",
+        "lexical": "a 1",
+        "valid": false
+      },
+      {
+        "type": "xs:string",
+        "lexical": "anything <>&",
+        "valid": true
+      },
+      {
+        "type": "xs:anyURI",
+        "lexical": "http://example.com/a b",
+        "valid": true
+      },
+      {
+        "type": "xs:boolean",
+        "lexical": "true",
+        "valid": true
+      },
+      {
+        "type": "xs:boolean",
+        "lexical": "0",
+        "valid": true
+      },
+      {
+        "type": "xs:boolean",
+        "lexical": "TRUE",
+        "valid": false
+      },
+      {
+        "type": "xs:language",
+        "lexical": "en-US",
+        "valid": true
+      },
+      {
+        "type": "xs:language",
+        "lexical": "en_US",
+        "valid": false
+      },
+      {
+        "type": "xs:Name",
+        "lexical": "a:b",
+        "valid": true
+      },
+      {
+        "type": "xs:NCName",
+        "lexical": "a:b",
+        "valid": false
+      }
+    ]
+  }
+}

data/lib/moxml/adapter/customized_leptris/declaration.rb

@@ -1,5 +1,14 @@
 # frozen_string_literal: true
 
+# Wrapper for an XML declaration.
+#
+# Lifecycle record: libleptris models the declaration as document-state
+# flags — setters shipped (leptris#1094) and clear shipped
+# (leptris#1229, binding 1.9.204.0) — but there is still no first-class
+# declaration NODE, so the facade keeps this mutable, removable value
+# object as the node stand-in. Dissolves when libleptris exposes
+# declaration node identity (leptris#1904's successor surface).
+
 module Moxml
   module Adapter
     module CustomizedLeptris

data/lib/moxml/adapter/customized_leptris/doctype.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# Wrapper for a programmatic DOCTYPE.
+#
+# Lifecycle record: set shipped (leptris#212, binding 1.9.176) and unset
+# shipped (leptris#1229, binding 1.9.204.0 remove_doctype), but there is
+# still no first-class DOCTYPE NODE — the facade keeps this value object
+# as the node stand-in in the document's attachments. Dissolves with
+# declaration node identity (see customized_leptris/declaration.rb).
+
 module Moxml
   module Adapter
     module CustomizedLeptris

data/lib/moxml/adapter/customized_leptris/entity_reference.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# Parse-path face: libleptris expands the five built-in entities at
+# parse time and has no custom ones, so parsed references are carried
+# as Moxml::Entity::Reference values over marker-bearing text.
+# Programmatic creation is native (NATIVE_ENTITY_REFS, binding >= 1.9.177).
+#
+# Lifecycle record: dissolves when parse-time entity reference retention
+# ships — leptris#1904 (references as nodes or bulk positions).
+
 module Moxml
   module Adapter
     module CustomizedLeptris

data/lib/moxml/adapter/customized_leptris/text_segment.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# A text run split out of a native text node that carried entity
+# markers. Value object: the moxml contract exposes text nodes as
+# separate children around entity references, while libleptris stores
+# them as one marker-bearing text node.
+#
+# Lifecycle record: dissolves with entity reference retention —
+# leptris#1904 (see customized_leptris/entity_reference.rb).
+
 module Moxml
   module Adapter
     module CustomizedLeptris

data/lib/moxml/adapter/customized_leptris.rb

@@ -7,6 +7,11 @@ module Moxml
     # and document-level processing instructions.
     # The adapter stores them via NativeAttachment and serializes them
     # from #to_xml.
+    #
+    # Lifecycle record (#495): a tracked shim, not an adapter-fate
+    # case — the parse-time entity-retention gap is the last engine
+    # ask of the #275 mutation-completion round; when the engine ships
+    # native retention these classes dissolve into Base.
     module CustomizedLeptris
       autoload :Declaration, "moxml/adapter/customized_leptris/declaration"
       autoload :Doctype, "moxml/adapter/customized_leptris/doctype"

data/lib/moxml/adapter/customized_libxml/cdata.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/comment.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/declaration.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/element.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/entity_reference.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/node.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/processing_instruction.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml/text.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+# Lifecycle record: the :libxml adapter is deprecated (#491); this wrapper dies with it.
+
 module Moxml
   module Adapter
     module CustomizedLibxml

data/lib/moxml/adapter/customized_libxml.rb

@@ -2,6 +2,9 @@
 
 module Moxml
   module Adapter
+    # Lifecycle record (#495): engine-gap shims for the :libxml
+    # adapter, deprecated in moxml#491 -- removed with it after the
+    # cycle. No separate upstream ask; the adapter's fate governs.
     module CustomizedLibxml
       autoload :Cdata, "moxml/adapter/customized_libxml/cdata"
       autoload :Comment, "moxml/adapter/customized_libxml/comment"

data/lib/moxml/adapter/customized_oga/entity_decoder.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# Override Oga::EntityDecoder.decode for XML inputs.
+#
+# Engine gap (upstream oga): Oga 3.4's stock Oga::XML::Entities.decode
+# runs multiple passes, turning the well-formed "&" into "&"
+# rather than the correct "&". Upstream tracker:
+# https://github.com/yorickpeterse/oga — the override stays until oga
+# fixes the decoder or the adapter is retired.
+
 require "oga"
 
 module Moxml

data/lib/moxml/adapter/customized_oga/raw_value_override.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# Bypass Oga's lazy decoder for user-authored values.
+#
+# Engine gap (upstream oga): Oga::XML::Attribute#value and
+# Oga::XML::Text#text decode @value/@text on first read via
+# Oga::EntityDecoder; EntityDecoderOverride (this directory) must hold
+# user-authored raw values. Upstream tracker:
+# https://github.com/yorickpeterse/oga
+
 require "oga"
 
 module Moxml

data/lib/moxml/adapter/customized_oga/xml_declaration.rb

@@ -1,5 +1,13 @@
 # frozen_string_literal: true
 
+# Stand-in XML declaration node.
+#
+# Engine gap (upstream oga): Oga::XML::XmlDeclaration is not writable —
+# version/encoding/standalone cannot be set on an instance the way the
+# facade contract requires, so this subclass owns the state and renders
+# the declaration itself. Upstream tracker:
+# https://github.com/yorickpeterse/oga
+
 require "oga"
 
 module Moxml

data/lib/moxml/adapter/customized_oga/xml_generator.rb

@@ -1,6 +1,6 @@
 # rubocop:disable Style/FrozenStringLiteralComment
 
-require "oga"
+# Lifecycle record: engine gap (upstream oga); see the patched-class URL below.require "oga"
 
 # monkey patch the Oga generator because it's not configurable
 # https://github.com/yorickpeterse/oga/blob/main/lib/oga/xml/generator.rb

data/lib/moxml/adapter/customized_oga.rb

@@ -2,6 +2,10 @@
 
 module Moxml
   module Adapter
+    # Lifecycle record (#495): every override serves the :oga adapter,
+    # deprecated in moxml#491 and deleted with the Opal surface in
+    # moxml#489 phase 4. No separate upstream ask -- the adapter's
+    # fate governs this file.
     module CustomizedOga
       autoload :EntityDecoderOverride, "moxml/adapter/customized_oga/entity_decoder"
       autoload :RawValueOverride, "moxml/adapter/customized_oga/raw_value_override"

data/lib/moxml/adapter/customized_ox/attribute.rb

@@ -1,5 +1,10 @@
 # frozen_string_literal: true
 
+# Engine gap (upstream ox): an ::Ox::Attribute carries no parent
+# back-reference and no prefix slots, so the facade's attribute node
+# contract needs this Node subclass. Upstream tracker:
+# https://github.com/ohler55/ox
+
 module Moxml
   module Adapter
     module CustomizedOx

data/lib/moxml/adapter/customized_ox/entity_reference.rb

@@ -1,5 +1,10 @@
 # frozen_string_literal: true
 
+# Parse-path alias: ox expands the built-in entities at parse time and
+# has no reference nodes, so parsed references are the facade's
+# Moxml::Entity::Reference values. Lifecycle: the :ox adapter is
+# deprecated (#491) — this record dies with it.
+
 module Moxml
   module Adapter
     module CustomizedOx

data/lib/moxml/adapter/customized_ox/namespace.rb

@@ -1,5 +1,10 @@
 # frozen_string_literal: true
 
+# Engine gap (upstream ox): ox has no Namespace node class — prefixes
+# live as attribute spellings — so the facade's namespace node contract
+# needs this value object. Upstream tracker:
+# https://github.com/ohler55/ox
+
 module Moxml
   module Adapter
     module CustomizedOx

data/lib/moxml/adapter/customized_ox/text.rb

@@ -1,5 +1,11 @@
 # frozen_string_literal: true
 
+# Engine gap (upstream ox): Ox uses Strings for text content, but a
+# String cannot carry a @parent back-reference; this ::Ox::Node subclass
+# gives text nodes node identity. NOTE: an ::Ox::Node subclass that is
+# neither ::Ox::Element nor ::Ox::Document changes Ox's dump behavior.
+# Upstream tracker: https://github.com/ohler55/ox
+
 module Moxml
   module Adapter
     module CustomizedOx

data/lib/moxml/adapter/customized_ox.rb

@@ -2,6 +2,9 @@
 
 module Moxml
   module Adapter
+    # Lifecycle record (#495): engine-gap shims for the :ox adapter,
+    # deprecated in moxml#491 -- removed with it after the cycle.
+    # No separate upstream ask; the adapter's fate governs.
     module CustomizedOx
       autoload :Attribute, "moxml/adapter/customized_ox/attribute"
       autoload :EntityReference, "moxml/adapter/customized_ox/entity_reference"

data/lib/moxml/adapter/customized_rexml/entity_reference.rb

@@ -1,5 +1,10 @@
 # frozen_string_literal: true
 
+# Parse-path alias: rexml expands the built-in entities at parse time,
+# so parsed references are the facade's Moxml::Entity::Reference values.
+# Lifecycle: the :rexml adapter is Opal-only and dies with the Opal
+# surface (#489 phase 4); third-party engine, no upstream ask.
+
 module Moxml
   module Adapter
     module CustomizedRexml

data/lib/moxml/adapter/customized_rexml/formatter.rb

@@ -1,5 +1,11 @@
 # frozen_string_literal: true
 
+# Custom REXML formatter that fixes indentation and wrapping issues.
+# Engine gap (upstream rexml): REXML's Formatters::Pretty cannot be
+# configured to the facade's whitespace contract. Third-party engine,
+# no upstream ask; the :rexml adapter is Opal-only and dies with the
+# Opal surface (#489 phase 4).
+
 require "rexml/formatters/pretty"
 
 module Moxml

data/lib/moxml/adapter/customized_rexml.rb

@@ -2,6 +2,10 @@
 
 module Moxml
   module Adapter
+    # Lifecycle record (#495): engine-gap shims for the :rexml
+    # adapter (Opal fallback; deleted with the Opal surface in
+    # moxml#489 phase 4). No separate upstream ask; the adapter's
+    # fate governs this file.
     module CustomizedRexml
       autoload :EntityReference,
                "moxml/adapter/customized_rexml/entity_reference"

data/lib/moxml/adapter.rb

@@ -2,6 +2,19 @@
 
 module Moxml
   module Adapter
+    # Extension contract (#494): adding an adapter is exactly the trio
+    # below — nothing else, anywhere. The dispatch is open for
+    # extension and closed against edit: registration is data, the
+    # load path is generic, and spec/moxml/adapter_registry_spec.rb
+    # walks the registry mechanically so a skipped leg fails there.
+    #
+    #   1. one entry in AVAILABLE_ADAPTERS,
+    #   2. a requireable file at moxml/adapter/<name> (a
+    #      `Customized<Name>` namespace module is optional),
+    #   3. a subclass of Adapter::Base in that file.
+    #
+    # Deprecated adapters live in DEPRECATED_ADAPTERS (still fully
+    # registered and loadable) until their removal milestone.
     autoload :Base, "moxml/adapter/base"
     autoload :CustomizedOga, "moxml/adapter/customized_oga"
     autoload :CustomizedOx, "moxml/adapter/customized_ox"

data/lib/moxml/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Moxml
-  VERSION = "0.5.171"
+  VERSION = "0.5.172"
 end

data/lib/moxml/xsd/builtin_lexical.rb

@@ -255,6 +255,106 @@ module Moxml
         "xs:NOTATION" => ->(lexical) { qname?(collapsed(lexical)) },
       }.freeze
 
+      # The cross-language conformance dataset (#496): the model's
+      # builtin-lexical answers as data, pinned BY the Ruby suite
+      # (spec/moxml/xsd_conformance_cases_spec.rb) and exported to
+      # conformance/moxml-conformance.json by `rake conformance:export`
+      # — moxml-rs and moxml-ts run the same rows. Each row:
+      # [type, lexical, expected].
+      CONFORMANCE_CASES = [
+        # token family / collapse
+        ["xs:token", "a  b", true],
+        ["xs:ID", " id ", true],
+        ["xs:NMTOKEN", "\tx ", true],
+        ["xs:ID", "1d", false],
+        # unbounded integers
+        ["xs:integer", "1" * 29, true],
+        ["xs:positiveInteger", "1" * 29, true],
+        ["xs:nonNegativeInteger", "1" * 29, true],
+        ["xs:negativeInteger", "-1#{'0' * 20}", true],
+        ["xs:nonPositiveInteger", "0", true],
+        ["xs:nonPositiveInteger", "-0", true],
+        # bounded integers
+        ["xs:long", "9223372036854775807", true],
+        ["xs:long", "9223372036854775808", false],
+        ["xs:unsignedLong", "18446744073709551615", true],
+        ["xs:unsignedLong", "18446744073709551616", false],
+        ["xs:byte", "128", false],
+        ["xs:unsignedByte", "256", false],
+        # float/decimal spellings
+        ["xs:float", "+INF", true],
+        ["xs:double", "-INF", true],
+        ["xs:float", "NaN", true],
+        ["xs:float", "1.5E-3", true],
+        ["xs:decimal", "5.", true],
+        ["xs:decimal", ".5", true],
+        ["xs:decimal", "1e3", false],
+        # calendar
+        ["xs:date", "2024-02-29", true],
+        ["xs:date", "1999-02-29", false],
+        ["xs:date", "2024-04-31", false],
+        ["xs:date", "2024-13-01", false],
+        ["xs:date", "not-a-date", false],
+        ["xs:dateTime", "2024-02-29T23:59:59Z", true],
+        ["xs:dateTime", "1999-02-29T00:00:00", false],
+        ["xs:dateTime", "not-a-date", false],
+        ["xs:dateTimeStamp", "2024-02-29T00:00:00Z", true],
+        ["xs:dateTimeStamp", "2024-02-29T00:00:00", false],
+        ["xs:time", "23:59:59.999+05:30", true],
+        ["xs:time", "24:00:00", false],
+        ["xs:time", "25:20:00", false],
+        ["xs:time", "13:20:60", false],
+        ["xs:gYearMonth", "2024-02Z", true],
+        ["xs:gYearMonth", "2024-13", false],
+        ["xs:gMonthDay", "--02-29", true],
+        ["xs:gMonthDay", "--02-30", false],
+        ["xs:gDay", "---31", true],
+        ["xs:gDay", "---32", false],
+        # duration
+        ["xs:duration", "P1Y2M3DT4H5M6.5S", true],
+        ["xs:duration", "PT0S", true],
+        ["xs:duration", "-P1D", true],
+        ["xs:duration", "P", false],
+        ["xs:duration", "PT", false],
+        ["xs:duration", "P1Y2MT", false],
+        ["xs:duration", "P200.5Y", false],
+        ["xs:dayTimeDuration", "P1DT2H", true],
+        ["xs:dayTimeDuration", "P1Y", false],
+        ["xs:yearMonthDuration", "P1Y2M", true],
+        ["xs:yearMonthDuration", "P1D", false],
+        ["xs:yearMonthDuration", "P1YT1H", false],
+        # binary
+        ["xs:base64Binary", "aGVsbG8=", true],
+        ["xs:base64Binary", "abcd", true],
+        ["xs:base64Binary", "abcde", false],
+        ["xs:base64Binary", "a b", false],
+        ["xs:hexBinary", "0A", true],
+        ["xs:hexBinary", "0AB", false],
+        # QName family
+        ["xs:QName", "foo", true],
+        ["xs:QName", "ns:foo", true],
+        ["xs:QName", "xmlns:xsi", false],
+        ["xs:QName", ":foo", false],
+        ["xs:QName", "a:b:c", false],
+        ["xs:NOTATION", "foo", true],
+        # lists: items, not characters
+        ["xs:IDREFS", "a b", true],
+        ["xs:IDREFS", "a", true],
+        ["xs:IDREFS", "", false],
+        ["xs:NMTOKENS", "a.b-c:d e", true],
+        ["xs:ENTITIES", "a 1", false],
+        # open lexical spaces
+        ["xs:string", "anything <>&", true],
+        ["xs:anyURI", "http://example.com/a b", true],
+        ["xs:boolean", "true", true],
+        ["xs:boolean", "0", true],
+        ["xs:boolean", "TRUE", false],
+        ["xs:language", "en-US", true],
+        ["xs:language", "en_US", false],
+        ["xs:Name", "a:b", true],
+        ["xs:NCName", "a:b", false],
+      ].map(&:freeze).freeze
+
       private_constant :CHECKERS
     end
   end

data/spec/moxml/adapter/customized_lifecycle_audit_spec.rb

@@ -0,0 +1,49 @@
+# frozen_string_literal: true
+
+require "spec_helper"
+
+# #495: the Customized* adapter layer is either a removable-lifecycle
+# record (an engine gap with a tracked upstream issue) or it dissolves
+# into Base. This audit makes the lifecycle record checkable instead of
+# folklore: every override file must cite its issue (an #NNN reference
+# for our engines, or an upstream tracker URL for third-party engines
+# we do not file against), and every Customized module must be wired
+# into exactly its adapter (no dead records).
+RSpec.describe "Customized adapter lifecycle audit" do
+  let(:override_files) do
+    Dir[File.expand_path("../../lib/moxml/adapter/customized_*/*.rb", __dir__)].sort
+  end
+
+  it "cites a tracked issue or upstream tracker in every override file" do
+    broken = override_files.reject do |file|
+      head = File.readlines(file).first(25).join
+      head.match?(/#\s*#\d{2,5}\b/) || head.include?("github.com/")
+    end
+
+    expect(broken).to be_empty, <<~MSG
+      Customized* override files without a lifecycle citation:
+      #{broken.join("\n")}
+      Each file's header must cite the open upstream issue that explains
+      why the override exists (#NNN for leptris; the upstream tracker URL
+      for third-party engines), or the #491 deprecation that retires it.
+    MSG
+  end
+
+  it "wires every Customized module into exactly its adapter" do
+    shims = Dir[File.expand_path("../../lib/moxml/adapter/customized_*.rb", __dir__)]
+
+    shims.each do |shim|
+      module_name = File.read(shim)[/module (Customized\w+)/, 1]
+      adapter_file = File.expand_path(
+        "../../lib/moxml/adapter/#{module_name.delete_prefix('Customized').downcase}.rb",
+        __dir__,
+      )
+
+      expect(File).to exist(adapter_file),
+                      "#{module_name} has no #{adapter_file} to customize"
+
+      expect(File.read(adapter_file)).to include(module_name),
+                                         "#{module_name} is not wired into its adapter"
+    end
+  end
+end

data/spec/moxml/adapter_registry_spec.rb

@@ -0,0 +1,44 @@
+# frozen_string_literal: true
+
+require "spec_helper"
+
+# The registry's extension contract (#494): adding an adapter is the
+# documented trio — one AVAILABLE_ADAPTERS entry, one requireable
+# adapter file, one Base subclass. The dispatch is open for extension:
+# this spec walks the registry mechanically so a new adapter that
+# skips any leg fails here, not in a consumer.
+RSpec.describe "Adapter registry extension contract" do
+  it "loads every registered adapter through the public face" do
+    Moxml::Adapter::AVAILABLE_ADAPTERS.each do |name|
+      adapter = Moxml::Adapter.load(name)
+      expect(adapter).not_to be_nil
+    end
+  end
+
+  it "resolves every registered adapter to an Adapter::Base subclass" do
+    Moxml::Adapter::AVAILABLE_ADAPTERS.each do |name|
+      adapter = Moxml::Adapter.load(name)
+      expect(adapter).to be <= Moxml::Adapter::Base
+    end
+  end
+
+  it "answers available? truthfully for every registered adapter" do
+    Moxml::Adapter::AVAILABLE_ADAPTERS.each do |name|
+      expect(Moxml::Adapter.available?(name)).to be(true)
+    end
+  end
+
+  it "rejects unregistered names before any require runs" do
+    expect do
+      Moxml::Adapter.load(:definitely_not_an_adapter)
+    end.to raise_error(Moxml::AdapterError, /not available on this platform/)
+  end
+
+  it "keeps the deprecated adapters inside the registry with guidance" do
+    described = Moxml::Adapter::DEPRECATED_ADAPTERS
+    expect(described.keys).to all(satisfy { |name| Moxml::Adapter::AVAILABLE_ADAPTERS.include?(name) })
+    described.each_value do |guidance|
+      expect(guidance).to match(/:nokogiri|:leptris/)
+    end
+  end
+end

data/spec/moxml/customized_lifecycle_spec.rb

@@ -0,0 +1,47 @@
+# frozen_string_literal: true
+
+require "spec_helper"
+
+# The customized_* layer's doctrine (#495): each namespace module is
+# either a tracked shim (an engine gap with a stated resolution) or an
+# adapter-fate case (dies with its deprecated adapter). A Customized*
+# file exists only when its header says which — the record is
+# checkable, not folklore.
+RSpec.describe "customized adapter layer lifecycle" do
+  def customized_files
+    Dir["lib/moxml/adapter/customized_*.rb"]
+  end
+
+  it "has customized namespaces to govern" do
+    expect(customized_files).not_to be_empty
+  end
+
+  it "cites its lifecycle in every module header" do
+    customized_files.each do |path|
+      header = File.readlines(path).first(30).join
+      expect(header).to match(/#\d{3}/),
+                        "#{path} carries no lifecycle citation (#495)"
+      expect(header).to match(/Lifecycle record|no native representation/),
+                        "#{path} header does not state its lifecycle"
+    end
+  end
+
+  it "marks adapter-fate cases as deprecated or Opal-bound" do
+    adapter_fate = customized_files.reject { |p| p.include?("leptris") }
+    expect(adapter_fate).not_to be_empty
+    adapter_fate.each do |path|
+      header = File.readlines(path).first(30).join
+      expect(header).to match(/moxml#491|moxml#489/),
+                        "#{path} is an adapter-fate case but cites neither " \
+                        "its deprecation (#491) nor the Opal removal (#489)"
+    end
+  end
+
+  it "keeps the tracked shim pointed at its resolution" do
+    header = File.readlines("lib/moxml/adapter/customized_leptris.rb")
+      .first(30).join
+    expect(header).to include("tracked shim"), "the leptris shim must say so"
+    expect(header).to include("dissolve into Base"),
+                      "the leptris shim must state its end state"
+  end
+end

data/spec/moxml/xsd_conformance_cases_spec.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+# Processing model §Validation (normative): pinned here.
+
+require "spec_helper"
+
+# The conformance dataset is the model's builtin-lexical answers AS
+# DATA (#496): the Ruby suite pins every row here, and the exported
+# JSON (rake conformance:export) is what moxml-rs and moxml-ts run.
+# A row that fails here breaks the cross-language contract, not just
+# this implementation.
+RSpec.describe "builtin lexical conformance dataset" do
+  def valid_lexical?(type, lexical)
+    Moxml::XSD.builtin_valid?(Moxml.new(:leptris), type, lexical)
+  end
+
+  it "carries the dataset" do
+    expect(Moxml::XSD::BuiltinLexical::CONFORMANCE_CASES.size).to be >= 80
+  end
+
+  it "answers every row exactly as the dataset expects" do
+    Moxml::XSD::BuiltinLexical::CONFORMANCE_CASES.each do |type, lexical, expected|
+      answer = valid_lexical?(type, lexical)
+      expect(answer).to eql(expected),
+                        "#{type} #{lexical.inspect}: got #{answer.inspect}, " \
+                        "dataset expects #{expected.inspect}"
+    end
+  end
+
+  it "keeps every dataset type in the table" do
+    Moxml::XSD::BuiltinLexical::CONFORMANCE_CASES.each do |type, _lexical, _expected|
+      expect do
+        valid_lexical?(type, "")
+      end.not_to raise_error
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: moxml
 version: !ruby/object:Gem::Version
-  version: 0.5.171
+  version: 0.5.172
 platform: ruby
 authors:
 - Ribose Inc.
@@ -53,6 +53,7 @@ files:
 - bin/console
 - bin/generate-model-coverage
 - bin/setup
+- conformance/moxml-conformance.json
 - data/w3c_entities.json
 - docs/ENTITY_SUPPORT_FOR_LUTAML_MODEL.md
 - docs/Gemfile
@@ -479,6 +480,7 @@ files:
 - spec/moxml/adapter/README.md
 - spec/moxml/adapter/base_spec.rb
 - spec/moxml/adapter/columnar_invalidation_spec.rb
+- spec/moxml/adapter/customized_lifecycle_audit_spec.rb
 - spec/moxml/adapter/entity_restoration_spec.rb
 - spec/moxml/adapter/headed_ox_spec.rb
 - spec/moxml/adapter/leptris_adoption_spec.rb
@@ -513,6 +515,7 @@ files:
 - spec/moxml/adapter/shared_examples/.gitkeep
 - spec/moxml/adapter/shared_examples/adapter_contract.rb
 - spec/moxml/adapter_deprecation_spec.rb
+- spec/moxml/adapter_registry_spec.rb
 - spec/moxml/adapter_spec.rb
 - spec/moxml/allocation_benchmark_spec.rb
 - spec/moxml/allocation_guard_spec.rb
@@ -538,6 +541,7 @@ files:
 - spec/moxml/config_spec.rb
 - spec/moxml/context_spec.rb
 - spec/moxml/css_spec.rb
+- spec/moxml/customized_lifecycle_spec.rb
 - spec/moxml/cycle_rejection_spec.rb
 - spec/moxml/declaration_preservation_spec.rb
 - spec/moxml/declaration_spec.rb
@@ -634,6 +638,7 @@ files:
 - spec/moxml/xpath_capabilities_spec.rb
 - spec/moxml/xpath_spec.rb
 - spec/moxml/xsd_builtin_lexical_spec.rb
+- spec/moxml/xsd_conformance_cases_spec.rb
 - spec/moxml/xsd_engine_meta_spec.rb
 - spec/moxml/xsd_error_contract_spec.rb
 - spec/moxml/xsd_handle_mismatch_spec.rb