fix NPE in solr rss where snippet contains only the title text

and adjusted xslt, for solr snippets (&hl=true) to decode the xml encoded html <b> tag by adding disable-output-escaping
(still open item description may be double as dc: tag and rss.description tag)

htroot/yacysearch.xsl

@@ -28,7 +28,7 @@
 	<xsl:template match='item'>
 		<div class="searchresults">
 			<h4 class="linktitle"><a href="{link}" ><xsl:value-of select='title'/></a></h4>
-			<p class="snippet"><span class="snippetLoaded"><xsl:value-of select='description'/></span></p>
+			<p class="snippet"><span class="snippetLoaded"><xsl:value-of select='description' disable-output-escaping="yes"/></span></p>
 			<p class="url"><a href="{link}" ><xsl:value-of select='link' /></a></p>
 			<p class="urlinfo"><xsl:value-of select='pubDate' /></p>
 		</div>

pom.xml

@@ -34,7 +34,7 @@
         <maven.compiler.source>1.7</maven.compiler.source>
         <maven.compiler.target>1.7</maven.compiler.target>
         <!-- the Solr version used in dependency section for all related dependencies -->
-        <solr.version>4.8.1</solr.version> 
+        <solr.version>4.9.0</solr.version> 
         
         <!-- properties used for filtering yacyBuildProperties.java -->       
         <REPL_DATE>${DSTAMP}</REPL_DATE>

source/net/yacy/cora/federate/solr/responsewriter/OpensearchResponseWriter.java

@@ -239,16 +239,14 @@ public class OpensearchResponseWriter implements QueryResponseWriter {
             LinkedHashSet<String> snippet = urlhash == null ? null : snippets.get(urlhash);
             String tagname = RSSMessage.Token.description.name();
             if (snippet == null || snippet.size() == 0) {
+                writer.write("<"); writer.write(tagname); writer.write('>');
                 for (String d: descriptions) {
-                    writer.write("<"); writer.write(tagname); writer.write('>');
                     XML.escapeCharData(d, writer);
-                    writer.write("</"); writer.write(tagname); writer.write(">\n");
                 }
-            } else {
-                OpensearchResponseWriter.removeSubsumedTitle(snippet, title);
-                writer.write("<"); writer.write(tagname); writer.write('>');
-                XML.escapeCharData(OpensearchResponseWriter.getLargestSnippet(snippet), writer);
                 writer.write("</"); writer.write(tagname); writer.write(">\n");
+            } else {
+                removeSubsumedTitle(snippet, title);
+                solitaireTag(writer, tagname, getLargestSnippet(snippet)); // snippet may be size=0
             }
             // open: where do we get the subject?
             //solitaireTag(writer, DublinCore.Subject.getURIref(), ""); // TODO: fill with actual data