Hi Folks,

I'm baffled.

Here is a sample input line:

5.2 Record Type (S/T)

The first part -- 5.2 -- is the 'section number'.

The second part -- Record Type -- is the 'section name'.

The third part -- (S/T) -- is the 'abbreviated name'.

I want the XSLT to generate this:

<header>
    <section-num>5.2</section-num>
    <section-name>Record Type</section-name>
    <abbreviated-name>(S/T)</abbreviated-name>
</header>

I created several ENTITIES for use by my XSLT:

<!ENTITY section-num "[0-9]+(\.[0-9]+)+">
<!ENTITY section-name "[A-Z][a-zA-Z /]+">
<!ENTITY paren "\([^\)]+\)">

I use 'analyze-string' to break apart the input:

<xsl:analyze-string select="$line" regex="^(&section-num;) (&section-name;)
(&paren;)$">
    <xsl:matching-substring>
        <header>
            <section-num><xsl:value-of
select="regex-group(1)"/></section-num>
            <section-name><xsl:value-of
select="regex-group(2)"/></section-name>
            <abbreviated-name><xsl:value-of
select="regex-group(3)"/></abbreviated-name>
        </header>
    </xsl:matching-substring>
</xsl:analyze-string>

That produces this erroneous output:

<header>
    <section-num>5.2</section-num>
    <section-name>.2</section-name>
    <abbreviated-name>Record Type</abbreviated-name>
</header>

I am baffled why I am getting that output.

What am I doing wrong, please?

/Roger