Transformation Question

I am using xsltproc (the default install in Fedora) to transform my docbook article to HTML. I noticed in the source code of the HTML file it puts  and what appears to be a pipe in some places.

 

For example I am using the <literallayout> element as shown below:

<literallayout>&amp;lt;hotel-room code="NSQ" occupancy="5" max-occupancy="5" non-smoking="yes" suite="yes"&gt;
&nbsp;   ; &nbsp; &nbsp; &lt;bed type="queen" count="1"/&gt;
&nbsp; &nbsp;   ; &nbsp; &lt;bed type="twin" count="3"/&gt;<;/literallayout>

 

The source code of the HTML file for this section after the transformation:

&lt;div class="literallayout"><p&gt;&lt;hotel-roomÂ| code="NSQ"Â| occupancy="5"Â| max-occupancy="5"Â| non-smoking="yes"Â| suite="yes"&gt;&lt;br />
Â| Â| Â| Â| Â| Â| Â| Â| Â| &lt;bedÂ| type="queen"Â| count="1"/&gt;<;br />
Â| Â| Â| Â| Â| Â| Â| Â| Â| &lt;bedÂ| type="twin"Â| count="3"/&gt;<;/p></div>

 

Then of course, in a browser it would display as it looks below:

<hotel-room code="NSQ"Â&nbsp;occupancy="5"Â&nbsp;max-occupancy="5"Â&nbsp;non-smoking="yes" suite="yes">
 Â&nbsp;  Â&nbsp;  Â&nbsp; &lt;bed ;type="queen" count="1"/>;
 Â&nbsp; ; Â&nbsp;  Â&nbsp; &lt;bedÂ&nbsp;type="twin" count="3"/>;

 

I get this same behavior in other places as well (e.g. table titles), but this example has the most occurences.

 

At this point I am not sure it this is caused by xsltproc, the encoding, or some other factor that I am overlooking. Has anyone run across this before? Any ideas on how to fix this?

&nbsp;

Thanks in advance.

 

Andy