Parsing XML with LotusScript and Microsoft's XML object (Page 3)

Search DominoPower's 11,425 Lotus-related article archive

Home
EasyPrint
News details

Articles-only

Twitter Feed

Parsing XML with LotusScript and Microsoft's XML object (continued)

The second XMLStyleSheet agent
XSL style sheets are really a subject by themselves but, as always, you can start with something simple. A good starting point is using HTML elements with XSL statements to select fields from a source XML document. An analogy would be a mail merge.

For this agent, I created a new TableTest document to hold the style sheet and used it against the XML field in my News document which was populated when I ran the XMLHeadlines agent above. Using existing XML data from this field allowed me to try out the loadXML function which loads a string rather than a file. With the XML and style sheet data loaded into two separate objects, it was a simple matter of calling transformNode() and putting the returned string of HTML in a new document. Below is the style sheet I used in the agent.


<?xml version='1.0'?>

<xsl:stylesheet xmlns:xsl="http://www.w3.org/TR/WD-xsl";>

  <xsl:template match="/">

    <HTML>

      <BODY>

        <TABLE BORDER="2">

          <TR>

            <TD><H3><xsl:value-of select="rdf:RDF/channel/title"/></H3></TD>

          </TR>

          <xsl:for-each select="rdf:RDF/item">

            <TR>

              <TD><xsl:value-of select="title"/></TD>

            </TR>

          </xsl:for-each>

        </TABLE>

      </BODY>

    </HTML>

  </xsl:template>

</xsl:stylesheet>

The "template match" line says to start at the root of the XML document. The "value-of select" statement picks up a particular field value as in the first agent (in this case, the title of the channel). The "for-each" statement repeats a block of HTML for each matching element (in this case, each item). Within the block I get the particular title value. The following is a partial output:


<HTML>

<BODY>

<TABLE BORDER="2">

<TR>

<TD><H3>DominoPower News Center</H3></TD>

</TR>

<TR>

<TD>Battle against piracy</TD>

</TR>

<TR>

<TD>SmartSuite review</TD>

</TR>

The beauty of this approach is that you can use the same agent with different style sheets, depending on what you want as output, a bit like using sub-forms in a form. The style sheets can be very complex and perform selection, ordering, pattern matching, and can have embedded JavaScript in as well. Hopefully this little example will whet your appetite for further experimentation as it did mine.

Product availability and resources
Internet Explorer 5 is freely available from http://www.microsoft.com.

The example agents database was created using Domino 4.63 and is at http://www.electricmemo.com/articler.nsf.

You can also download the sample database from ftp://ftp.dominopower.com/dominopower/outgoing/issues/dp0899-xml.zip. It is not normally our practice at DominoPower to provide download samples. However, at the request of the author, we have posted this file. If you have problems using this file, please contact the author directly or post a message on the PowerBoards. The editorial staff at DominoPower is not able to provide technical support.

The MSXML on-line resources are at http://msdn.microsoft.com/xml/reference/scriptref/XMLDOM_Objects.asp.

For reference material, check out XML in Action by William J. Pardi ISBN 0-7356-0562-9.

For a list of published channels download My.UserLand's service list from http://my.userland.com/xml/servicelist.xml. If you're using Internet Explorer 5, you'll be able to directly see the XML code when clicking on this link. Other browsers may attempt to download this list as a file.

For a general list of who's publishing XML content try XMLTree visit http://www.xmltree.com.