Anyone have an example on how to convert HTML code back to charatcter in an XSL?

For example:

changing: "capitol'
into: "capitol'

or

changing:
<p>Page 12
<p>
<p> CONGRESS KNEW PAROCHIALISM
would trump national security in deciding

into:
Page 12

CONGRESS KNEW PAROCHIALISM
would trump national security in deciding

I want to automate data entry into my Filemaker Pro 7 database, and I can get all the data, but it still has HTML code leftover. Any help is greatly appreciated! Here's what I have so far:

<?xml version="1.0" encoding="UTF-8" ?>
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
    <xsl:output indent="yes" method="xml"/>
    <xsl:template match="newsclips">
        <FMPXMLRESULT xmlns="http:www.filemaker.com/fmpxmlresult">
            <ERRORCODE>0</ERRORCODE>
            <PRODUCT BUILD="11-05-2003" NAME="Filemaker Pro" VERSION="7.0v1"/>
            <DATABASE DATEFORMAT="M/d/yyyy" LAYOUT="" NAME="test.fp7" RECORDS="{@count}" TIMEFORMAT="h:mm:ss a"/>

            <METADATA>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="clipid" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="headline" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="link" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="registration" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="clipdate" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="cliptime" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="cliprating" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="source" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="sourceid" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="state" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="country" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="circulation" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="byline" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="sourcelanguage" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="keywordfound" TYPE="TEXT"/>
                <FIELD EMPTYOK="YES" MAXREPEAT="1" NAME="abstract" TYPE="TEXT"/>
            </METADATA>

            <RESULTSET FOUND="{@count}">
                <xsl:for-each select="newsclip">
                    <ROW MODID="0" RECORDID="{position()}">
                        <COL><DATA><xsl:value-of select="clipid"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="headline"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="link"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="registration"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="clipdate"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="cliptime"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="cliprating"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="source"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="sourceid"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="state"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="country"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="circulation"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="byline"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="sourcelanguage"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="keywordfound"/></DATA></COL>
                        <COL><DATA><xsl:value-of select="abstract"/></DATA></COL>

                    </ROW>
                </xsl:for-each>
            </RESULTSET>
        </FMPXMLRESULT>
    </xsl:template>
</xsl:stylesheet>

Thanks,

-Teo

I suspect that your XML source contains escaped HTML markup. This is always a bad idea. Because the markup is disguised by escaping it, it won't be recognized by the XML parser: you have to extract it as text and put it back through another stage of parsing, e.g. using an extension like saxon:parse().

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

Thanks Michael,

The XML feed is custom produced by an agency for the people I work for. Should I request the XML be cleaned up at the source if I can? If not, how would I make use of saxon:parse? Do I need to download the opensource version of saxon from sourceforge? I was looking at w3 and there was a section on disabling output escaping, is this a bad idea?

Thanks again!

-Teo