Howdy, all --<div><br></div><div>If I run the following query against a large, full-text-indexed store (output from &quot;svn log --xml&quot; on a repository with several hundred thousand entries), I get results back in well under a second:</div>
<div><br></div><div>count(/log/logentry[./paths/path/text() contains text &quot;trunk&quot;])</div><div><br></div><div>However, the following variant takes multiple minutes to execute -- actually, has never yet completed before I&#39;ve lost patience and cancelled it:</div>
<div><br></div><div>count(/log/logentry[</div><div>  ./paths/path/text() contains text &quot;trunk&quot;</div><div>  and not(</div><div>    (./paths/path/text() contains text &quot;tags&quot;) or</div><div>    (./paths/path/text() contains text &quot;branches&quot;)</div>
<div>  )</div><div>])</div><div><br></div><div>Is there a way to get reasonable(ish) performance for the more complex query?</div>