<html xmlns="http://www.w3.org/1999/xhtml">
    <body>
        <h1>Free text operator</h1>
        <h2>Indexing</h2>
        <p>By default, eXist creates free-text indexes on all elements as the documented is loaded, which are updated when updates on. 
This behaviourcan be supressed using the system configuration file...</p>
        <h2>Using the free text operator</h2>
        <p>XQuery in exist extends the XQuery language with the &amp;= operator. For example <br/>

 doc('/db/advice/matchexample.xml')//p[ . &amp;= 'how']   
<br/>
would find all paragraphs containing the word 'how'. 
This match is case-independent. You will notice that the matches have been tagged with <br/>

&lt;exist:match xmlns:exist=http://exist.sourceforge.net/NS/exist"&gt;how&lt;/exist:match&gt; 

</p>
        <h2>Removing the matching tag</h2>
        <p>Use the serialise option highlight-matches=none  to omit this tag from the output.</p>
        <h2>Highlighting the matches</h2>
        <p>It would be handy to be able to use a style sheet to highlight matches by for example, setting the font to red.  
Until CSS supports multiple namespaces, this must be done using XSLT to transform this tag into an HTML tag</p>
    </body>
</html>
