moxml 0.1.26 → 0.3.0

data/spec/performance/benchmark_spec.rb

@@ -23,14 +23,21 @@ RSpec.shared_examples "Performance Examples" do
     context "measures performance" do
       let(:doc) { context.parse(large_xml) }
 
+      # Absolute ips varies more than 2x across machines and even between
+      # runs on the same machine (observed: rexml serializer 1.5-2.9 ips on
+      # one machine). These thresholds sit at roughly half of typical
+      # modern-runner throughput so they only trip on asymptotic
+      # regressions (accidental O(n^2), broken caching), not on hardware
+      # variance or a few percent of legitimate overhead.
       let(:thresholds) do
         {
-          nokogiri: { parser: 15, serializer: 1000 },
-          oga: { parser: 10, serializer: 100 },
-          rexml: { parser: 0, serializer: 5 },
-          ox: { parser: 2, serializer: 1000 },
-          headed_ox: { parser: 2, serializer: 1000 },
-          libxml: { parser: 500, serializer: 60 },
+          nokogiri: { parser: 8, serializer: 500 },
+          oga: { parser: 5, serializer: 50 },
+          rexml: { parser: 0, serializer: 1 },
+          ox: { parser: 1, serializer: 400 },
+          headed_ox: { parser: 1, serializer: 400 },
+          libxml: { parser: 200, serializer: 30 },
+          leptris: { parser: 1300, serializer: 600 },
         }
       end
 

data/spec/performance/thread_safety_spec.rb

@@ -20,10 +20,6 @@ RSpec.shared_examples "Thread Safety Examples" do
     end
 
     it "handles concurrent processing" do
-      if Moxml.new.config.adapter_name == :ox
-        skip "Ox doesn't have a native XPath"
-      end
-
       processor = processor_class.new
       threads = []
       results = Queue.new

data/spec/spec_helper.rb

@@ -101,7 +101,11 @@ RSpec.configure do |config|
 end
 
 Moxml.configure do |config|
-  config.adapter = RUBY_ENGINE == "opal" ? Moxml::Config::OPAL_DEFAULT_ADAPTER : Moxml::Config::DEFAULT_ADAPTER
+  # Pin the suite baseline to the fallback adapter so results are
+  # reproducible regardless of whether the local environment has a
+  # leptris new enough to win the runtime-default probe. Leptris gets
+  # full coverage through its explicit adapter/integration matrices.
+  config.adapter = RUBY_ENGINE == "opal" ? Moxml::Config::OPAL_DEFAULT_ADAPTER : Moxml::Config::FALLBACK_ADAPTER
   config.strict_parsing = true
   config.default_encoding = "UTF-8"
   config.entity_load_mode = :optional if RUBY_ENGINE == "opal"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: moxml
 version: !ruby/object:Gem::Version
-  version: 0.1.26
+  version: 0.3.0
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-07-22 00:00:00.000000000 Z
+date: 2026-08-26 00:00:00.000000000 Z
 dependencies: []
 description: |
   Moxml is a unified XML manipulation library that provides a common API
@@ -27,6 +27,7 @@ files:
 - ".github/workflows/rake.yml"
 - ".github/workflows/release.yml"
 - ".github/workflows/round-trip.yml"
+- ".github/workflows/windows.yml"
 - ".gitignore"
 - ".gitmodules"
 - ".rspec"
@@ -119,6 +120,11 @@ files:
 - lib/moxml.rb
 - lib/moxml/adapter.rb
 - lib/moxml/adapter/base.rb
+- lib/moxml/adapter/customized_leptris.rb
+- lib/moxml/adapter/customized_leptris/declaration.rb
+- lib/moxml/adapter/customized_leptris/doctype.rb
+- lib/moxml/adapter/customized_leptris/entity_reference.rb
+- lib/moxml/adapter/customized_leptris/text_segment.rb
 - lib/moxml/adapter/customized_libxml.rb
 - lib/moxml/adapter/customized_libxml/cdata.rb
 - lib/moxml/adapter/customized_libxml/comment.rb
@@ -129,6 +135,8 @@ files:
 - lib/moxml/adapter/customized_libxml/processing_instruction.rb
 - lib/moxml/adapter/customized_libxml/text.rb
 - lib/moxml/adapter/customized_oga.rb
+- lib/moxml/adapter/customized_oga/entity_decoder.rb
+- lib/moxml/adapter/customized_oga/raw_value_override.rb
 - lib/moxml/adapter/customized_oga/xml_declaration.rb
 - lib/moxml/adapter/customized_oga/xml_generator.rb
 - lib/moxml/adapter/customized_ox.rb
@@ -140,14 +148,15 @@ files:
 - lib/moxml/adapter/customized_rexml/entity_reference.rb
 - lib/moxml/adapter/customized_rexml/formatter.rb
 - lib/moxml/adapter/headed_ox.rb
+- lib/moxml/adapter/leptris.rb
 - lib/moxml/adapter/libxml.rb
 - lib/moxml/adapter/libxml/entity_ref_registry.rb
-- lib/moxml/adapter/libxml/entity_restorer.rb
 - lib/moxml/adapter/nokogiri.rb
 - lib/moxml/adapter/oga.rb
 - lib/moxml/adapter/ox.rb
 - lib/moxml/adapter/rexml.rb
 - lib/moxml/attribute.rb
+- lib/moxml/attribute_resolver.rb
 - lib/moxml/builder.rb
 - lib/moxml/c14n.rb
 - lib/moxml/c14n/attribute_handler.rb
@@ -177,8 +186,10 @@ files:
 - lib/moxml/declaration.rb
 - lib/moxml/doctype.rb
 - lib/moxml/document.rb
-- lib/moxml/document_builder.rb
 - lib/moxml/element.rb
+- lib/moxml/entity.rb
+- lib/moxml/entity/reference.rb
+- lib/moxml/entity/restorer.rb
 - lib/moxml/entity_reference.rb
 - lib/moxml/entity_registry.rb
 - lib/moxml/entity_registry_opal_data.rb
@@ -246,6 +257,7 @@ files:
 - lib/moxml/signature/verifier.rb
 - lib/moxml/text.rb
 - lib/moxml/version.rb
+- lib/moxml/xml_emitter.rb
 - lib/moxml/xml_utils.rb
 - lib/moxml/xml_utils/encoder.rb
 - lib/moxml/xpath.rb
@@ -371,12 +383,13 @@ files:
 - spec/integration/README.md
 - spec/integration/all_adapters_spec.rb
 - spec/integration/headed_ox_integration_spec.rb
+- spec/integration/sax_parity_spec.rb
 - spec/integration/shared_examples/edge_cases.rb
 - spec/integration/shared_examples/entity_reference_whitespace.rb
 - spec/integration/shared_examples/high_level/.gitkeep
 - spec/integration/shared_examples/high_level/builder_behavior.rb
 - spec/integration/shared_examples/high_level/context_behavior.rb
-- spec/integration/shared_examples/high_level/document_builder_behavior.rb
+- spec/integration/shared_examples/high_level/parse_behavior.rb
 - spec/integration/shared_examples/integration_workflows.rb
 - spec/integration/shared_examples/line_ending_behavior.rb
 - spec/integration/shared_examples/node_wrappers/.gitkeep
@@ -401,6 +414,7 @@ files:
 - spec/moxml/adapter/base_spec.rb
 - spec/moxml/adapter/entity_restoration_spec.rb
 - spec/moxml/adapter/headed_ox_spec.rb
+- spec/moxml/adapter/leptris_spec.rb
 - spec/moxml/adapter/libxml_internals_spec.rb
 - spec/moxml/adapter/libxml_spec.rb
 - spec/moxml/adapter/nokogiri_spec.rb
@@ -413,6 +427,7 @@ files:
 - spec/moxml/adapter_spec.rb
 - spec/moxml/allocation_benchmark_spec.rb
 - spec/moxml/allocation_guard_spec.rb
+- spec/moxml/attribute_resolver_spec.rb
 - spec/moxml/attribute_spec.rb
 - spec/moxml/builder_spec.rb
 - spec/moxml/c14n/api_spec.rb
@@ -428,12 +443,12 @@ files:
 - spec/moxml/declaration_preservation_spec.rb
 - spec/moxml/declaration_spec.rb
 - spec/moxml/doctype_spec.rb
-- spec/moxml/document_builder_spec.rb
 - spec/moxml/document_spec.rb
 - spec/moxml/element_spec.rb
 - spec/moxml/entity_preservation_spec.rb
 - spec/moxml/entity_reference_spec.rb
 - spec/moxml/entity_registry_spec.rb
+- spec/moxml/entity_spec.rb
 - spec/moxml/error_spec.rb
 - spec/moxml/lazy_parse_spec.rb
 - spec/moxml/moxml_spec.rb
@@ -455,6 +470,7 @@ files:
 - spec/moxml/opal_smoke_spec.rb
 - spec/moxml/processing_instruction_spec.rb
 - spec/moxml/sax/namespace_splitter_spec.rb
+- spec/moxml/sax_entity_parity_spec.rb
 - spec/moxml/sax_spec.rb
 - spec/moxml/signature/algorithms/base64_transform_spec.rb
 - spec/moxml/signature/algorithms/digest_base_spec.rb
@@ -472,6 +488,7 @@ files:
 - spec/moxml/signature/round_trip_spec.rb
 - spec/moxml/text_spec.rb
 - spec/moxml/version_spec.rb
+- spec/moxml/xml_emitter_spec.rb
 - spec/moxml/xml_utils/.gitkeep
 - spec/moxml/xml_utils/encoder_spec.rb
 - spec/moxml/xml_utils_spec.rb

data/lib/moxml/adapter/libxml/entity_restorer.rb

@@ -1,94 +0,0 @@
-# frozen_string_literal: true
-
-module Moxml
-  module Adapter
-    class Libxml < Base
-      # Restores configured character entities into explicit Moxml
-      # EntityReference nodes after LibXML has parsed the native tree.
-      class EntityRestorer
-        def initialize(doc, adapter: Libxml)
-          @doc = doc
-          @ctx = doc.context
-          @registry = @ctx.entity_registry
-          @config = @ctx.config
-          @adapter = adapter
-        end
-
-        def run
-          return unless @registry && @doc.root
-
-          walk(@doc.root)
-        end
-
-        private
-
-        def walk(element)
-          # Snapshot because we may add/remove siblings during the walk.
-          element.children.to_a.each do |child|
-            if child.is_a?(::Moxml::Text)
-              restore_text_node(child)
-            elsif child.is_a?(::Moxml::Element)
-              walk(child)
-            end
-          end
-        end
-
-        # Matches DocumentBuilder's previous behavior, including the libxml
-        # limitation that adjacent native text nodes get merged.
-        def restore_text_node(text_node)
-          content = text_node.content
-          return unless content
-
-          chunks = chunk_text(content)
-          return if chunks.size == 1 && chunks.first.first == :text
-
-          parent = text_node.parent
-          return unless parent
-
-          text_node.remove
-          chunks.each { |type, payload| append_chunk(parent, type, payload) }
-        end
-
-        def chunk_text(content)
-          chunks = []
-          buffer = +""
-          restorable = @registry.restorable_codepoints
-
-          content.each_char do |char|
-            cp = char.ord
-            if restorable.include?(cp) &&
-                (name = @registry.primary_name_for_codepoint(cp)) &&
-                @registry.should_restore?(cp, config: @config)
-              unless buffer.empty?
-                chunks << [:text, buffer.dup]
-                buffer.clear
-              end
-              chunks << [:eref, name]
-            else
-              buffer << char
-            end
-          end
-
-          chunks << [:text, buffer.dup] unless buffer.empty?
-          chunks
-        end
-
-        def append_chunk(parent, type, payload)
-          case type
-          when :text
-            parent.add_child(::Moxml::Text.new(
-                               @adapter.create_native_text(payload), @ctx
-                             ))
-          when :eref
-            parent.add_child(
-              ::Moxml::EntityReference.new(
-                @adapter.create_native_entity_reference(payload),
-                @ctx,
-              ),
-            )
-          end
-        end
-      end
-    end
-  end
-end

data/lib/moxml/document_builder.rb

@@ -1,167 +0,0 @@
-# frozen_string_literal: true
-
-module Moxml
-  class DocumentBuilder
-    attr_reader :context
-
-    def initialize(context)
-      @context = context
-      @node_stack = []
-    end
-
-    def build(native_doc)
-      @current_doc = context.create_document(native_doc)
-
-      # Transfer has_declaration flag if present in attachments
-      if adapter.attachments.key?(native_doc, :has_declaration)
-        has_declaration = adapter.attachments.get(native_doc, :has_declaration)
-        @current_doc.has_xml_declaration = has_declaration
-      end
-
-      # Transfer DOCTYPE from parsed document if it exists in attachments
-      if adapter.attachments.key?(native_doc, :doctype)
-        doctype = adapter.attachments.get(native_doc, :doctype)
-        if doctype
-          adapter.attachments.set(@current_doc.native, :doctype, doctype)
-        end
-      end
-
-      visit_node(native_doc)
-      @current_doc
-    end
-
-    private
-
-    def visit_node(node)
-      case node_type(node)
-      when :document then visit_document(node)
-      when :element then visit_element(node)
-      when :text then visit_text(node)
-      when :cdata then visit_cdata(node)
-      when :comment then visit_comment(node)
-      when :processing_instruction then visit_processing_instruction(node)
-      when :doctype then visit_doctype(node)
-      when :entity_reference then visit_entity_reference(node)
-      end
-    end
-
-    def visit_document(doc)
-      @node_stack.push(@current_doc)
-      visit_children(doc)
-      @node_stack.clear
-    end
-
-    def visit_element(node)
-      childless_node = adapter.duplicate_node(node)
-      adapter.replace_children(childless_node, [])
-      # Prepare node for new document (LibXML needs this)
-      childless_node = adapter.prepare_for_new_document(
-        childless_node,
-        @current_doc.native,
-      )
-      element = Element.new(childless_node, context)
-      @node_stack.last.add_child(element)
-
-      @node_stack.push(element) # add a parent for its children
-      visit_children(node)
-      @node_stack.pop # remove the parent
-    end
-
-    def visit_text(node)
-      # Prepare node for new document before wrapping
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      content = adapter.text_content(node)
-
-      # Check if we should restore entity references for this text
-      if context.config.restore_entities && text_has_restorable_entities?(content)
-        restore_entities_in_text(content)
-      else
-        @node_stack.last&.add_child(Text.new(prepared, context))
-      end
-    end
-
-    def text_has_restorable_entities?(content)
-      return false unless content
-
-      registry = context.entity_registry
-      codepoints = registry.restorable_codepoints
-      content.each_char do |char|
-        return true if codepoints.include?(char.ord)
-      end
-      false
-    end
-
-    def restore_entities_in_text(content)
-      parent = @node_stack.last
-      return unless parent
-
-      registry = context.entity_registry
-      config = context.config
-      buffer = +""
-
-      content.to_s.each_char do |char|
-        codepoint = char.ord
-        name = registry.primary_name_for_codepoint(codepoint)
-
-        if name && registry.should_restore?(codepoint, config: config)
-          # Flush buffered text before the entity
-          unless buffer.empty?
-            parent.add_child(Text.new(adapter.create_text(buffer), context))
-            buffer.clear
-          end
-          parent.add_child(
-            EntityReference.new(adapter.create_entity_reference(name), context),
-          )
-        else
-          buffer << char
-        end
-      end
-
-      # Flush remaining buffer
-      unless buffer.empty?
-        parent.add_child(Text.new(adapter.create_text(buffer), context))
-      end
-    end
-
-    def visit_cdata(node)
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      @node_stack.last&.add_child(Cdata.new(prepared, context))
-    end
-
-    def visit_comment(node)
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      @node_stack.last&.add_child(Comment.new(prepared, context))
-    end
-
-    def visit_processing_instruction(node)
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      @node_stack.last&.add_child(ProcessingInstruction.new(prepared, context))
-    end
-
-    def visit_doctype(node)
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      @node_stack.last&.add_child(Doctype.new(prepared, context))
-    end
-
-    def visit_entity_reference(node)
-      prepared = adapter.prepare_for_new_document(node, @current_doc.native)
-      @node_stack.last&.add_child(EntityReference.new(prepared, context))
-    end
-
-    def visit_children(node)
-      children(node).each { |child| visit_node(child) }
-    end
-
-    def node_type(node)
-      context.config.adapter.node_type(node)
-    end
-
-    def children(node)
-      context.config.adapter.children(node)
-    end
-
-    def adapter
-      context.config.adapter
-    end
-  end
-end

data/spec/moxml/document_builder_spec.rb

@@ -1,30 +0,0 @@
-# frozen_string_literal: true
-
-require "spec_helper"
-
-# DocumentBuilder is an internal class for parsing
-# These tests verify it correctly builds Moxml document from native documents
-RSpec.describe Moxml::DocumentBuilder do
-  let(:context) { Moxml.new }
-
-  describe "#build" do
-    it "builds a Moxml document from native document" do
-      native_doc = context.parse("<root><child>content</child></root>").native
-      builder = described_class.new(context)
-      doc = builder.build(native_doc)
-
-      expect(doc).to be_a(Moxml::Document)
-      expect(doc.root.name).to eq("root")
-    end
-  end
-
-  describe "element handling" do
-    it "handles nested elements" do
-      native_doc = context.parse("<parent><child1/><child2/></parent>").native
-      builder = described_class.new(context)
-      doc = builder.build(native_doc)
-
-      expect(doc.root.children.length).to eq(2)
-    end
-  end
-end