oga 0.2.0-java → 0.2.1-java

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

@@ -101,6 +101,31 @@ public class Lexer extends RubyObject
         int pe    = data.length;
         int eof   = data.length;
 
+        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_doctype_end      = "on_doctype_end";
+        String id_on_doctype_inline   = "on_doctype_inline";
+        String id_on_doctype_name     = "on_doctype_name";
+        String id_on_doctype_start    = "on_doctype_start";
+        String id_on_doctype_type     = "on_doctype_type";
+        String id_on_element_end      = "on_element_end";
+        String id_on_element_name     = "on_element_name";
+        String id_on_element_ns       = "on_element_ns";
+        String id_on_element_open_end = "on_element_open_end";
+        String id_on_element_start    = "on_element_start";
+        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_string_body      = "on_string_body";
+        String id_on_string_dquote    = "on_string_dquote";
+        String id_on_string_squote    = "on_string_squote";
+        String id_on_text             = "on_text";
+        String id_on_xml_decl_end     = "on_xml_decl_end";
+        String id_on_xml_decl_start   = "on_xml_decl_start";
+
         %% write exec;
 
         this.lines = lines;
@@ -160,6 +185,17 @@ public class Lexer extends RubyObject
 
         this.callMethod(context, "advance_line", lines);
     }
+
+    /**
+     * Returns true if we're in an HTML script tag. See
+     * Oga::XML::Lexer#inside_html_script? for more information.
+     */
+    public Boolean inside_html_script_p()
+    {
+        ThreadContext context = this.runtime.getCurrentContext();
+
+        return this.callMethod(context, "inside_html_script?").isTrue();
+    }
 }
 
 %%{

data/ext/ragel/base_lexer.rl

@@ -28,6 +28,17 @@
     # When you call a method in Ruby make sure that said method is defined as
     # an instance method in the `Oga::XML::Lexer` class.
     #
+    # The name of the callback to invoke should be an identifier starting with
+    # "id_". The identifier should be defined in the associated C and Java code.
+    # In case of C code its value should be a Symbol as a ID object, for Java
+    # it should be a String. For example:
+    #
+    #     ID id_foo = rb_intern("foo");
+    #
+    # And for Java:
+    #
+    #     String id_foo = "foo";
+    #
     # ## Machine Transitions
     #
     # To transition from one machine to another always use `fnext` instead of
@@ -59,7 +70,7 @@
     comment       = comment_start (any* -- comment_end) comment_end;
 
     action start_comment {
-        callback("on_comment", data, encoding, ts + 4, te - 3);
+        callback(id_on_comment, data, encoding, ts + 4, te - 3);
     }
 
     # CDATA
@@ -75,7 +86,7 @@
     cdata       = cdata_start (any* -- cdata_end) cdata_end;
 
     action start_cdata {
-        callback("on_cdata", data, encoding, ts + 9, te - 3);
+        callback(id_on_cdata, data, encoding, ts + 9, te - 3);
     }
 
     # Processing Instructions
@@ -93,8 +104,8 @@
     proc_ins_end   = '?>';
 
     action start_proc_ins {
-        callback_simple("on_proc_ins_start");
-        callback("on_proc_ins_name", data, encoding, ts + 2, te);
+        callback_simple(id_on_proc_ins_start);
+        callback(id_on_proc_ins_name, data, encoding, ts + 2, te);
 
         mark = te;
 
@@ -103,8 +114,8 @@
 
     proc_ins_body := |*
         proc_ins_end => {
-            callback("on_text", data, encoding, mark, ts);
-            callback_simple("on_proc_ins_end");
+            callback(id_on_text, data, encoding, mark, ts);
+            callback_simple(id_on_proc_ins_end);
 
             mark = 0;
 
@@ -124,7 +135,7 @@
     squote = "'";
 
     action emit_string {
-        callback("on_string_body", data, encoding, ts, te);
+        callback(id_on_string_body, data, encoding, ts, te);
 
         if ( lines > 0 )
         {
@@ -135,13 +146,13 @@
     }
 
     action start_string_squote {
-        callback_simple("on_string_squote");
+        callback_simple(id_on_string_squote);
 
         fcall string_squote;
     }
 
     action start_string_dquote {
-        callback_simple("on_string_dquote");
+        callback_simple(id_on_string_dquote);
 
         fcall string_dquote;
     }
@@ -150,7 +161,7 @@
         ^squote* $count_newlines => emit_string;
 
         squote => {
-            callback_simple("on_string_squote");
+            callback_simple(id_on_string_squote);
 
             fret;
         };
@@ -160,7 +171,7 @@
         ^dquote* $count_newlines => emit_string;
 
         dquote => {
-            callback_simple("on_string_dquote");
+            callback_simple(id_on_string_dquote);
 
             fret;
         };
@@ -179,22 +190,35 @@
     doctype_start = '<!DOCTYPE'i whitespace+;
 
     action start_doctype {
-        callback_simple("on_doctype_start");
+        callback_simple(id_on_doctype_start);
         fnext doctype;
     }
 
+    # Machine for processing inline rules of a doctype.
+    doctype_inline := |*
+        ^']'* $count_newlines => {
+            callback(id_on_doctype_inline, data, encoding, ts, te);
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+        };
+
+        ']' => { fnext doctype; };
+    *|;
+
     # Machine for processing doctypes. Doctype values such as the public
     # and system IDs are treated as T_STRING tokens.
     doctype := |*
         'PUBLIC' | 'SYSTEM' => {
-            callback("on_doctype_type", data, encoding, ts, te);
+            callback(id_on_doctype_type, data, encoding, ts, te);
         };
 
-        # Consumes everything between the [ and ]. Due to the use of :> the ]
-        # is not consumed by any+.
-        '[' any+ :> ']' => {
-            callback("on_doctype_inline", data, encoding, ts + 1, te - 1);
-        };
+        # Starts a set of inline doctype rules.
+        '[' => { fnext doctype_inline; };
 
         # Lex the public/system IDs as regular strings.
         squote => start_string_squote;
@@ -205,11 +229,11 @@
         whitespace;
 
         identifier => {
-            callback("on_doctype_name", data, encoding, ts, te);
+            callback(id_on_doctype_name, data, encoding, ts, te);
         };
 
         '>' => {
-            callback_simple("on_doctype_end");
+            callback_simple(id_on_doctype_end);
             fnext main;
         };
     *|;
@@ -222,20 +246,20 @@
     xml_decl_end   = '?>';
 
     action start_xml_decl {
-        callback_simple("on_xml_decl_start");
+        callback_simple(id_on_xml_decl_start);
         fnext xml_decl;
     }
 
     # Machine that processes the contents of an XML declaration tag.
     xml_decl := |*
         xml_decl_end => {
-            callback_simple("on_xml_decl_end");
+            callback_simple(id_on_xml_decl_end);
             fnext main;
         };
 
         # Attributes and their values (e.g. version="1.0").
         identifier => {
-            callback("on_attribute", data, encoding, ts, te);
+            callback(id_on_attribute, data, encoding, ts, te);
         };
 
         squote => start_string_squote;
@@ -257,23 +281,23 @@
     element_end   = '</' identifier (':' identifier)* '>';
 
     action start_element {
-        callback_simple("on_element_start");
+        callback_simple(id_on_element_start);
         fhold;
         fnext element_name;
     }
 
     action close_element {
-        callback_simple("on_element_end");
+        callback_simple(id_on_element_end);
     }
 
     # Machine used for lexing the name/namespace of an element.
     element_name := |*
         identifier ':' => {
-            callback("on_element_ns", data, encoding, ts, te - 1);
+            callback(id_on_element_ns, data, encoding, ts, te - 1);
         };
 
         identifier => {
-            callback("on_element_name", data, encoding, ts, te);
+            callback(id_on_element_name, data, encoding, ts, te);
             fnext element_head;
         };
     *|;
@@ -284,16 +308,16 @@
         whitespace | '=';
 
         newline => {
-            callback_simple("advance_line");
+            callback_simple(id_advance_line);
         };
 
         # Attribute names and namespaces.
         identifier ':' => {
-            callback("on_attribute_ns", data, encoding, ts, te - 1);
+            callback(id_on_attribute_ns, data, encoding, ts, te - 1);
         };
 
         identifier => {
-            callback("on_attribute", data, encoding, ts, te);
+            callback(id_on_attribute, data, encoding, ts, te);
         };
 
         # Attribute values.
@@ -302,13 +326,23 @@
 
         # We're done with the open tag of the element.
         '>' => {
-            callback_simple("on_element_open_end");
-            fnext main;
+            callback_simple(id_on_element_open_end);
+
+            if ( inside_html_script_p() )
+            {
+                mark = ts + 1;
+
+                fnext script_text;
+            }
+            else
+            {
+                fnext main;
+            }
         };
 
         # Self closing tags.
         '/>' => {
-            callback_simple("on_element_end");
+            callback_simple(id_on_element_end);
             fnext main;
         };
     *|;
@@ -337,7 +371,7 @@
 
     text := |*
         terminate_text | allowed_text => {
-            callback("on_text", data, encoding, ts, te);
+            callback(id_on_text, data, encoding, ts, te);
 
             if ( lines > 0 )
             {
@@ -351,7 +385,7 @@
 
         # Text followed by a special tag, such as "foo<!--"
         allowed_text %{ mark = p; } terminate_text => {
-            callback("on_text", data, encoding, ts, mark);
+            callback(id_on_text, data, encoding, ts, mark);
 
             p    = mark - 1;
             mark = 0;
@@ -367,6 +401,30 @@
         };
     *|;
 
+    # <script> tags in HTML can contain basically anything except for the
+    # literal "</script>". As a result of this we can't use the regular text
+    # machine.
+    script_text := |*
+        '</script>' => {
+            callback(id_on_text, data, encoding, mark, ts);
+
+            mark = 0;
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+
+            callback_simple(id_on_element_end);
+
+            fnext main;
+        };
+
+        any $count_newlines;
+    *|;
+
     # The main machine aka the entry point of Ragel.
     main := |*
         doctype_start  => start_doctype;

data/lib/oga/version.rb

@@ -1,3 +1,3 @@
 module Oga
-  VERSION = '0.2.0'
+  VERSION = '0.2.1'
 end # Oga

data/lib/oga/xml/document.rb

@@ -77,6 +77,13 @@ module Oga
         return xml
       end
 
+      ##
+      # @return [TrueClass|FalseClass]
+      #
+      def html?
+        return type == :html
+      end
+
       ##
       # Inspects the document and its child nodes. Child nodes are indented for
       # each nesting level.

data/lib/oga/xml/element.rb

@@ -212,11 +212,8 @@ module Oga
       # @param [String] text
       #
       def inner_text=(text)
-        children.each do |child|
-          child.remove if child.is_a?(Text)
-        end
-
-        children << XML::Text.new(:text => text)
+        text_node = XML::Text.new(:text => text)
+        @children = NodeSet.new([text_node])
       end
 
       ##
@@ -310,7 +307,7 @@ module Oga
         self_closing = children.empty?
         root         = root_node
 
-        if root.is_a?(Document) and root.type == :html \
+        if root.is_a?(Document) and root.html? \
         and !HTML_VOID_ELEMENTS.include?(name)
           self_closing = false
         end

data/lib/oga/xml/entities.rb

@@ -10,9 +10,16 @@ module Oga
       # @return [Hash]
       #
       DECODE_MAPPING = {
-        '&lt;'  => '<',
-        '&gt;'  => '>',
-        '&amp;' => '&'
+        '&lt;'   => '<',
+        '&#60;'  => '<',
+        '&gt;'   => '>',
+        '&#62;'  => '>',
+        '&apos;' => "'",
+        '&#39;'  => "'",
+        '&quot;' => '"',
+        '&#34;'  => '"',
+        '&amp;'  => '&',
+        '&#38;'  => '&',
       }
 
       ##
@@ -22,8 +29,10 @@ module Oga
       #
       ENCODE_MAPPING = {
         '&' => '&amp;',
+        '"' => '&quot;',
+        "'" => '&apos;',
         '>' => '&gt;',
-        '<' => '&lt;'
+        '<' => '&lt;',
       }
 
       ##

data/lib/oga/xml/lexer.rb

@@ -40,6 +40,14 @@ module Oga
     class Lexer
       attr_reader :html
 
+      ##
+      # Element name used to determine if a tag being processed is a Javascript
+      # tag.
+      #
+      # @return [String]
+      #
+      SCRIPT_TAG = 'script'.freeze
+
       ##
       # @param [String|IO] data The data to lex. This can either be a String or
       #  an IO instance.
@@ -181,6 +189,15 @@ module Oga
         return @elements.last
       end
 
+      ##
+      # Returns true if the current element is the HTML `<script>` element.
+      #
+      # @return [TrueClass|FalseClass]
+      #
+      def inside_html_script?
+        return html? && current_element == SCRIPT_TAG
+      end
+
       ##
       # Called when processing a single quote.
       #