oga 0.3.1-java → 0.3.2-java

data/ext/java/org/liboga/xml/Lexer.rl

@@ -89,6 +89,8 @@ public class Lexer extends RubyObject
     @JRubyMethod
     public IRubyObject advance_native(ThreadContext context, RubyString rb_str)
     {
+        Boolean html_p = this.callMethod(context, "html").isTrue();
+
         Encoding encoding = rb_str.getEncoding();
 
         byte[] data = rb_str.getBytes();
@@ -104,8 +106,12 @@ public class Lexer extends RubyObject
         String id_advance_line        = "advance_line";
         String id_on_attribute        = "on_attribute";
         String id_on_attribute_ns     = "on_attribute_ns";
-        String id_on_cdata            = "on_cdata";
-        String id_on_comment          = "on_comment";
+        String id_on_cdata_start      = "on_cdata_start";
+        String id_on_cdata_body       = "on_cdata_body";
+        String id_on_cdata_end        = "on_cdata_end";
+        String id_on_comment_start    = "on_comment_start";
+        String id_on_comment_body     = "on_comment_body";
+        String id_on_comment_end      = "on_comment_end";
         String id_on_doctype_end      = "on_doctype_end";
         String id_on_doctype_inline   = "on_doctype_inline";
         String id_on_doctype_name     = "on_doctype_name";
@@ -119,6 +125,7 @@ public class Lexer extends RubyObject
         String id_on_proc_ins_end     = "on_proc_ins_end";
         String id_on_proc_ins_name    = "on_proc_ins_name";
         String id_on_proc_ins_start   = "on_proc_ins_start";
+        String id_on_proc_ins_body    = "on_proc_ins_body";
         String id_on_string_body      = "on_string_body";
         String id_on_string_dquote    = "on_string_dquote";
         String id_on_string_squote    = "on_string_squote";

data/ext/ragel/base_lexer.rl

@@ -67,12 +67,35 @@
 
     comment_start = '<!--';
     comment_end   = '-->';
-    comment       = comment_start (any* -- comment_end) comment_end;
+
+    # Everything except "-" OR a single "-"
+    comment_allowed = (^'-'+ | '-') $count_newlines;
 
     action start_comment {
-        callback(id_on_comment, data, encoding, ts + 4, te - 3);
+        callback_simple(id_on_comment_start);
+
+        fnext comment_body;
     }
 
+    comment_body := |*
+        comment_allowed => {
+            callback(id_on_comment_body, data, encoding, ts, te);
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+        };
+
+        comment_end => {
+            callback_simple(id_on_comment_end);
+
+            fnext main;
+        };
+    *|;
+
     # CDATA
     #
     # http://www.w3.org/TR/html-markup/syntax.html#cdata-sections
@@ -83,12 +106,35 @@
 
     cdata_start = '<![CDATA[';
     cdata_end   = ']]>';
-    cdata       = cdata_start (any* -- cdata_end) cdata_end;
+
+    # Everything except "]" OR a single "]"
+    cdata_allowed = (^']'+ | ']') $count_newlines;
 
     action start_cdata {
-        callback(id_on_cdata, data, encoding, ts + 9, te - 3);
+        callback_simple(id_on_cdata_start);
+
+        fnext cdata_body;
     }
 
+    cdata_body := |*
+        cdata_allowed => {
+            callback(id_on_cdata_body, data, encoding, ts, te);
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+        };
+
+        cdata_end => {
+            callback_simple(id_on_cdata_end);
+
+            fnext main;
+        };
+    *|;
+
     # Processing Instructions
     #
     # http://www.w3.org/TR/xpath/#section-Processing-Instruction-Nodes
@@ -103,26 +149,33 @@
     proc_ins_start = '<?' identifier;
     proc_ins_end   = '?>';
 
+    # Everything except "?" OR a single "?"
+    proc_ins_allowed = (^'?'+ | '?') $count_newlines;
+
     action start_proc_ins {
         callback_simple(id_on_proc_ins_start);
         callback(id_on_proc_ins_name, data, encoding, ts + 2, te);
 
-        mark = te;
-
         fnext proc_ins_body;
     }
 
     proc_ins_body := |*
+        proc_ins_allowed => {
+            callback(id_on_proc_ins_body, data, encoding, ts, te);
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+        };
+
         proc_ins_end => {
-            callback(id_on_text, data, encoding, mark, ts);
             callback_simple(id_on_proc_ins_end);
 
-            mark = 0;
-
             fnext main;
         };
-
-        any;
     *|;
 
     # Strings
@@ -253,19 +306,34 @@
     # Machine that processes the contents of an XML declaration tag.
     xml_decl := |*
         xml_decl_end => {
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+
             callback_simple(id_on_xml_decl_end);
+
             fnext main;
         };
 
         # Attributes and their values (e.g. version="1.0").
         identifier => {
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+
             callback(id_on_attribute, data, encoding, ts, te);
         };
 
         squote => start_string_squote;
         dquote => start_string_dquote;
 
-        any;
+        any $count_newlines;
     *|;
 
     # Elements
@@ -302,10 +370,49 @@
         };
     *|;
 
+    action hold_start_element_head {
+        fhold;
+        fnext element_head;
+    }
+
+    # Characters that can be used for unquoted HTML attribute values.
+    # See https://html.spec.whatwg.org/multipage/introduction.html#intro-early-example
+    # for more info.
+    html_unquoted_value = ^(
+        squote | dquote | '`' | '=' | '<' | '>' | whitespace | newline
+    )+;
+
+    # Machine used for processing HTML attribute values.
+    html_attribute_value := |*
+        squote | dquote => {
+            fhold;
+            fnext xml_attribute_value;
+        };
+
+        # Unquoted attribute values are lexed as if they were single quoted
+        # strings.
+        html_unquoted_value => {
+            callback_simple(id_on_string_squote);
+
+            callback(id_on_string_body, data, encoding, ts, te);
+
+            callback_simple(id_on_string_squote);
+        };
+
+        any => hold_start_element_head;
+    *|;
+
+    # Machine used for processing XML attribute values.
+    xml_attribute_value := |*
+        squote => start_string_squote;
+        dquote => start_string_dquote;
+        any    => hold_start_element_head;
+    *|;
+
     # Machine used for processing the contents of an element's starting tag.
     # This includes the name, namespace and attributes.
     element_head := |*
-        whitespace | '=';
+        whitespace;
 
         newline => {
             callback_simple(id_advance_line);
@@ -321,8 +428,16 @@
         };
 
         # Attribute values.
-        squote => start_string_squote;
-        dquote => start_string_dquote;
+        '=' => {
+            if ( html_p )
+            {
+                fnext html_attribute_value;
+            }
+            else
+            {
+                fnext xml_attribute_value;
+            }
+        };
 
         # We're done with the open tag of the element.
         '>' => {
@@ -438,8 +553,8 @@
     main := |*
         doctype_start  => start_doctype;
         xml_decl_start => start_xml_decl;
-        comment        => start_comment;
-        cdata          => start_cdata;
+        comment_start  => start_comment;
+        cdata_start    => start_cdata;
         proc_ins_start => start_proc_ins;
         element_start  => start_element;
         element_end    => close_element;

data/lib/oga/version.rb

@@ -1,3 +1,3 @@
 module Oga
-  VERSION = '0.3.1'
+  VERSION = '0.3.2'
 end # Oga

data/lib/oga/xml/document.rb

@@ -12,7 +12,7 @@ module Oga
     #  The XML declaration of the document.
     #  @return [Oga::XML::XmlDeclaration]
     #
-    # @!attribute [rw] type
+    # @!attribute [r] type
     #  The document type, either `:xml` or `:html`.
     #  @return [Symbol]
     #
@@ -20,7 +20,9 @@ module Oga
       include Querying
       include Traversal
 
-      attr_accessor :doctype, :xml_declaration, :type
+      attr_accessor :doctype, :xml_declaration
+
+      attr_reader :type
 
       ##
       # @param [Hash] options

data/lib/oga/xml/element.rb

@@ -8,7 +8,7 @@ module Oga
     #  The name of the element.
     #  @return [String]
     #
-    # @!attribute [ww] namespace_name
+    # @!attribute [r] namespace_name
     #  The name of the namespace.
     #  @return [String]
     #
@@ -23,7 +23,9 @@ module Oga
     class Element < Node
       include Querying
 
-      attr_accessor :name, :namespace_name, :attributes
+      attr_reader :namespace_name
+
+      attr_accessor :name, :attributes
 
       attr_writer :namespaces
 
@@ -56,6 +58,14 @@ module Oga
         register_namespaces_from_attributes
       end
 
+      ##
+      # @param [String] name
+      #
+      def namespace_name=(name)
+        @namespace_name = name
+        @namespace      = nil
+      end
+
       ##
       # Returns an attribute matching the given name (with or without the
       # namespace).
@@ -289,14 +299,17 @@ module Oga
       #
       # @param [String] name
       # @param [String] uri
+      # @param [TrueClass|FalseClass] flush
       # @see [Oga::XML::Namespace#initialize]
       #
-      def register_namespace(name, uri)
+      def register_namespace(name, uri, flush = true)
         if namespaces[name]
           raise ArgumentError, "The namespace #{name.inspect} already exists"
         end
 
         namespaces[name] = Namespace.new(:name => name, :uri => uri)
+
+        flush_namespaces_cache if flush
       end
 
       ##
@@ -306,20 +319,25 @@ module Oga
       # @return [Hash]
       #
       def available_namespaces
-        return {} if html? # HTML(5) completely ignores namespaces
-
-        merged = namespaces.dup
-        node   = parent
-
-        while node && node.respond_to?(:namespaces)
-          node.namespaces.each do |prefix, ns|
-            merged[prefix] = ns unless merged[prefix]
+        # HTML(5) completely ignores namespaces
+        if html?
+          return @available_namespaces ||= {}
+        elsif !@available_namespaces
+          merged = namespaces.dup
+          node   = parent
+
+          while node && node.respond_to?(:namespaces)
+            node.namespaces.each do |prefix, ns|
+              merged[prefix] = ns unless merged[prefix]
+            end
+
+            node = node.parent
           end
 
-          node = node.parent
+          @available_namespaces = merged
         end
 
-        return merged
+        return @available_namespaces
       end
 
       ##
@@ -339,19 +357,40 @@ module Oga
         return self_closing
       end
 
+      ##
+      # Flushes the namespaces cache of the current element and all its child
+      # elements.
+      #
+      def flush_namespaces_cache
+        @available_namespaces = nil
+        @namespace            = nil
+
+        children.each do |child|
+          child.flush_namespaces_cache if child.is_a?(Element)
+        end
+      end
+
       private
 
       ##
       # Registers namespaces based on any "xmlns" attributes.
       #
       def register_namespaces_from_attributes
+        flush = false
+
         attributes.each do |attr|
           # We're using `namespace_name` opposed to `namespace.name` as "xmlns"
           # is not a registered namespace.
           if attr.name == XMLNS_PREFIX or attr.namespace_name == XMLNS_PREFIX
-            register_namespace(attr.name, attr.value)
+            flush = true
+
+            # Ensures we only flush the cache once instead of flushing it on
+            # every register_namespace call.
+            register_namespace(attr.name, attr.value, false)
           end
         end
+
+        flush_namespaces_cache if flush
       end
 
       ##

data/lib/oga/xml/lexer.rb

@@ -262,10 +262,40 @@ module Oga
       end
 
       ##
-      # Called on a CDATA tag.
+      # Called on the open CDATA tag.
       #
-      def on_cdata(value)
-        add_token(:T_CDATA, value)
+      def on_cdata_start
+        add_token(:T_CDATA_START)
+      end
+
+      ##
+      # Called on the closing CDATA tag.
+      #
+      def on_cdata_end
+        add_token(:T_CDATA_END)
+      end
+
+      ##
+      # Called for the body of a CDATA tag.
+      #
+      # @param [String] value
+      #
+      def on_cdata_body(value)
+        add_token(:T_CDATA_BODY, value)
+      end
+
+      ##
+      # Called on the open comment tag.
+      #
+      def on_comment_start
+        add_token(:T_COMMENT_START)
+      end
+
+      ##
+      # Called on the closing comment tag.
+      #
+      def on_comment_end
+        add_token(:T_COMMENT_END)
       end
 
       ##
@@ -273,8 +303,8 @@ module Oga
       #
       # @param [String] value
       #
-      def on_comment(value)
-        add_token(:T_COMMENT, value)
+      def on_comment_body(value)
+        add_token(:T_COMMENT_BODY, value)
       end
 
       ##
@@ -314,6 +344,15 @@ module Oga
         add_token(:T_PROC_INS_NAME, value)
       end
 
+      ##
+      # Called on the body of a processing instruction.
+      #
+      # @param [String] value
+      #
+      def on_proc_ins_body(value)
+        add_token(:T_PROC_INS_BODY, value)
+      end
+
       ##
       # Called on the end of a processing instruction.
       #