parsing of Microsoft Word doc with style "Heading X" where X>6 creates invalid HTML with tags <h7>,<h8> etc

org.apache.tika.parser.microsoft.WordExtractor will translate heading styles to "h" tags with a level greater than 6 which means the xhtml is invalid. The xhtml DTD only defines header elements 1 to 6:
<!ENTITY % heading "h1|h2|h3|h4|h5|h6">

changing line 380 from:
tag = "h"+num;
to
tag = "h"+Math.min(num, 6);

will resolve this.