moxml 0.1.14 → 0.1.16

data/lib/moxml/adapter/nokogiri.rb

@@ -7,6 +7,10 @@ module Moxml
   module Adapter
     class Nokogiri < Base
       class << self
+        def attachments
+          @attachments ||= Moxml::NativeAttachment.new
+        end
+
         def set_root(doc, element)
           doc.root = element
         end
@@ -31,7 +35,7 @@ module Moxml
 
           # Use provided context if available, otherwise create new one
           ctx = _context || Context.new(:nokogiri)
-          DocumentBuilder.new(ctx).build(native_doc)
+          Document.new(native_doc, ctx)
         end
 
         # SAX parsing implementation for Nokogiri
@@ -47,7 +51,7 @@ module Moxml
           parser = ::Nokogiri::XML::SAX::Parser.new(bridge)
 
           # Parse
-          if xml.respond_to?(:read)
+          if xml.is_a?(IO) || xml.is_a?(StringIO)
             parser.parse(xml)
           else
             parser.parse(xml.to_s)
@@ -202,7 +206,7 @@ module Moxml
         end
 
         def root(document)
-          document.respond_to?(:root) ? document.root : document.children.first
+          document.is_a?(::Nokogiri::XML::Document) ? document.root : document.children.first
         end
 
         def attribute_element(attr)
@@ -241,8 +245,8 @@ module Moxml
             encoding = declaration_attribute(child, "encoding")
             standalone = declaration_attribute(child, "standalone")
 
-            # Nokogiri's xml_decl can only be set via instance variable
-            element.instance_variable_set(:@xml_decl, {
+            # Store declaration state in attachment map
+            attachments.set(element, :xml_decl, {
               version: version,
               encoding: encoding,
               standalone: standalone,
@@ -273,7 +277,7 @@ module Moxml
               node.name == "xml" &&
               node.parent.is_a?(::Nokogiri::XML::Document)
             # Clear document's xml_decl when removing declaration
-            node.parent.instance_variable_set(:@xml_decl, nil)
+            attachments.set(node.parent, :xml_decl, nil)
           end
 
           node.remove
@@ -387,13 +391,12 @@ module Moxml
           # Handle declaration option
           # Priority:
           # 1. Explicit no_declaration option
-          # 2. Check Nokogiri's internal @xml_decl (when remove is called, this becomes nil)
+          # 2. Check attachment-stored xml_decl (when remove is called, this becomes nil)
           if options.key?(:no_declaration)
             save_options |= ::Nokogiri::XML::Node::SaveOptions::NO_DECLARATION if options[:no_declaration]
-          elsif node.respond_to?(:instance_variable_get) &&
-              node.instance_variable_defined?(:@xml_decl)
-            # Nokogiri's internal state - if nil, declaration was removed
-            xml_decl = node.instance_variable_get(:@xml_decl)
+          elsif attachments.key?(node, :xml_decl)
+            # State stored in attachment - if nil, declaration was removed
+            xml_decl = attachments.get(node, :xml_decl)
             save_options |= ::Nokogiri::XML::Node::SaveOptions::NO_DECLARATION if xml_decl.nil?
           end
 
@@ -404,6 +407,14 @@ module Moxml
           )
         end
 
+        def has_declaration?(native_doc, wrapper)
+          if attachments.key?(native_doc, :xml_decl)
+            !attachments.get(native_doc, :xml_decl).nil?
+          else
+            wrapper.has_xml_declaration
+          end
+        end
+
         private
 
         def build_declaration_attrs(version, encoding, standalone)

data/lib/moxml/adapter/oga.rb

@@ -1,8 +1,7 @@
 # frozen_string_literal: true
 
 require_relative "base"
-require_relative "customized_oga/xml_generator"
-require_relative "customized_oga/xml_declaration"
+require_relative "customized_oga"
 require "oga"
 
 module Moxml
@@ -12,6 +11,10 @@ module Moxml
         # Standard XML entities handled natively by parsers
         STANDARD_XML_ENTITIES = %w[amp lt gt quot apos].freeze
 
+        def attachments
+          @attachments ||= Moxml::NativeAttachment.new
+        end
+
         def set_root(doc, element)
           # Clear existing root element if any - Oga's NodeSet needs special handling
           # We need to manually remove elements since NodeSet doesn't support clear or delete_if
@@ -46,7 +49,7 @@ module Moxml
         def sax_parse(xml, handler)
           bridge = OgaSAXBridge.new(handler)
 
-          xml_string = xml.respond_to?(:read) ? xml.read : xml.to_s
+          xml_string = xml.is_a?(IO) || xml.is_a?(StringIO) ? xml.read : xml.to_s
 
           # Manually call start_document (Oga doesn't)
           handler.on_start_document
@@ -72,6 +75,17 @@ module Moxml
           ::Oga::XML::Text.new(text: encode_entity_markers(content))
         end
 
+        def create_native_entity_reference(name)
+          text = ::Oga::XML::Text.new
+          text.text = "#{ENTITY_MARKER}#{name};"
+          attachments.set(text, :entity_name, name)
+          text
+        end
+
+        def entity_reference_name(node)
+          attachments.get(node, :entity_name)
+        end
+
         def create_native_cdata(content, _owner_doc = nil)
           ::Oga::XML::Cdata.new(text: content)
         end
@@ -132,10 +146,9 @@ module Moxml
         end
 
         def namespace(element)
-          if element.respond_to?(:namespace)
+          case element
+          when ::Oga::XML::Element, ::Oga::XML::Attribute
             element.namespace
-          elsif element.respond_to?(:namespaces)
-            element.namespaces.values.last
           end
         rescue NoMethodError
           # Oga attributes fail with NoMethodError:
@@ -150,7 +163,12 @@ module Moxml
         def node_type(node)
           case node
           when ::Oga::XML::Element then :element
-          when ::Oga::XML::Text then :text
+          when ::Oga::XML::Text
+            if attachments.key?(node, :entity_name)
+              :entity_reference
+            else
+              :text
+            end
           when ::Oga::XML::Cdata then :cdata
           when ::Oga::XML::Comment then :comment
           when ::Oga::XML::Attribute then :attribute
@@ -178,7 +196,7 @@ module Moxml
                              node.doctype].compact
           end
 
-          return all_children unless node.respond_to?(:children)
+          return all_children unless node.is_a?(::Oga::XML::Node) || node.is_a?(::Oga::XML::Document)
 
           all_children + node.children.reject do |child|
             child.is_a?(::Oga::XML::Text) &&
@@ -188,7 +206,7 @@ module Moxml
         end
 
         def parent(node)
-          node.parent if node.respond_to?(:parent)
+          node.parent if node.is_a?(::Oga::XML::Node)
         end
 
         def next_sibling(node)
@@ -215,7 +233,7 @@ module Moxml
         end
 
         def attributes(element)
-          return [] unless element.respond_to?(:attributes)
+          return [] unless element.is_a?(::Oga::XML::Element)
 
           # remove attributes-namespaces
           element.attributes.reject do |attr|
@@ -262,8 +280,8 @@ module Moxml
           # Special handling for declarations on Oga documents
           if element.is_a?(::Oga::XML::Document) &&
               child.is_a?(::Oga::XML::XmlDeclaration)
-            # Set as document's xml_declaration
-            element.instance_variable_set(:@xml_declaration, child)
+            # Track declaration state in attachment map
+            attachments.set(element, :xml_declaration, child)
           end
 
           element.children << child
@@ -295,8 +313,8 @@ module Moxml
           # Special handling for declarations on Oga documents
           if node.is_a?(::Oga::XML::XmlDeclaration) &&
               node.parent.is_a?(::Oga::XML::Document)
-            # Clear document's xml_declaration when removing declaration
-            node.parent.instance_variable_set(:@xml_declaration, nil)
+            # Clear declaration state in attachment map
+            attachments.set(node.parent, :xml_declaration, nil)
           end
 
           node.remove
@@ -316,10 +334,9 @@ module Moxml
         end
 
         def inner_text(node)
-          text = if node.respond_to?(:inner_text)
+          text = if node.is_a?(::Oga::XML::Element)
                    node.inner_text
                  else
-                   # Oga::XML::Text node for example
                    node.text
                  end
           restore_entity_markers(text)
@@ -327,7 +344,7 @@ module Moxml
 
         def set_text_content(node, content)
           encoded = encode_entity_markers(content)
-          if node.respond_to?(:inner_text=)
+          if node.is_a?(::Oga::XML::Element)
             node.inner_text = encoded
           else
             node.text = encoded
@@ -370,22 +387,32 @@ module Moxml
         end
 
         def namespace_definitions(node)
-          return [] unless node.respond_to?(:namespaces)
+          return [] unless node.is_a?(::Oga::XML::Element)
 
           node.namespaces.values
         end
 
         # Doctype accessor methods
+        # Note: Oga stores SYSTEM identifier in public_id for SYSTEM doctypes.
+        # See: Oga::XML::Doctype puts SYSTEM dtd in public_id, system_id is nil.
         def doctype_name(native)
           native.name
         end
 
         def doctype_external_id(native)
-          native.public_id
+          if native.type == "SYSTEM"
+            nil
+          else
+            native.public_id
+          end
         end
 
         def doctype_system_id(native)
-          native.system_id
+          if native.type == "SYSTEM"
+            native.public_id
+          else
+            native.system_id
+          end
         end
 
         def xpath(node, expression, namespaces = nil)
@@ -430,6 +457,16 @@ module Moxml
         # Simple entity-only regex with no nested quantifiers
         ENTITY_REF_REGEX = /&#{ENTITY_PATTERN};/
 
+        def has_declaration?(native_doc, _wrapper)
+          decl = attachments.get(native_doc, :xml_declaration)
+          if decl.nil? && !attachments.key?(native_doc, :xml_declaration)
+            # No attachment entry - check native doc (for parsed documents)
+            native_doc.respond_to?(:xml_declaration) && !native_doc.xml_declaration.nil?
+          else
+            !decl.nil?
+          end
+        end
+
         private
 
         # Convert &entity; back to \x01entity; for Oga text storage.
@@ -463,21 +500,22 @@ module Moxml
           # We need to handle declaration options ourselves for Document nodes
           if node.is_a?(::Oga::XML::Document)
             # Check if we should include declaration
-            # Priority: explicit option > existence of xml_declaration node
+            # Priority: explicit option > existence of xml_declaration (native or attachment)
+            effective_xml_declaration = node.xml_declaration || attachments.get(node, :xml_declaration)
             should_include_decl = if options.key?(:no_declaration)
                                     !options[:no_declaration]
                                   elsif options.key?(:declaration)
                                     options[:declaration]
                                   else
-                                    # Default: include if document has xml_declaration node
-                                    node.xml_declaration ? true : false
+                                    # Default: include if document has xml_declaration
+                                    effective_xml_declaration ? true : false
                                   end
 
             # Fix: Check if declaration already exists in children
             # This prevents duplicate declarations when document already has one
             has_existing_declaration = node.children.any?(::Oga::XML::XmlDeclaration)
 
-            if should_include_decl && !node.xml_declaration && !has_existing_declaration
+            if should_include_decl && !effective_xml_declaration && !has_existing_declaration
               # Need to add declaration - create default one
               output = []
               output << '<?xml version="1.0" encoding="UTF-8"?>'
@@ -512,7 +550,8 @@ module Moxml
 
           # Default: use XmlGenerator
           # But first check if we need to handle declaration specially
-          if node.is_a?(::Oga::XML::Document) && node.xml_declaration
+          effective_xml_declaration = node.is_a?(::Oga::XML::Document) && (node.xml_declaration || attachments.get(node, :xml_declaration))
+          if node.is_a?(::Oga::XML::Document) && effective_xml_declaration
             # Document has declaration - use custom handling to avoid duplicates
             output = []
             xml_declaration_serialized = false