http://codereview.appspot.com/2006042/diff/37001/38005 File java/gadgets/src/main/java/org/apache/shindig/gadgets/parse/caja/VanillaCajaHtmlParser.java (right):
http://codereview.appspot.com/2006042/diff/37001/38005#newcode70 java/gadgets/src/main/java/org/apache/shindig/gadgets/parse/caja/VanillaCajaHtmlParser.java:70: DomParser parser = getDomParser(source + '\n', mq); I don't think you need the extraneous EOL whitespace at the any more. http://codereview.appspot.com/2006042/diff/37001/38002 File java/gadgets/src/test/java/org/apache/shindig/gadgets/parse/caja/VanillaCajaHtmlParserTest.java (right): http://codereview.appspot.com/2006042/diff/37001/38002#newcode102 java/gadgets/src/test/java/org/apache/shindig/gadgets/parse/caja/VanillaCajaHtmlParserTest.java:102: public void testBadTagBalancing() throws Exception { This is not correct both according to spec and in practice on common browsers. The html5 spec suggests that the open script tag will consume till the next close script tag or the end of file. 4/4 modern browsers also consume tokens till the next close script tag and do not execute the script at all if a close script is not found. http://codereview.appspot.com/2006042/
