Scanning For FME Schema Features

The XML Reader defaults to scanning the FME features constructed through the xfMap when it is requested for schema features. This can be explicitly stated in the xfMap by including a <scan> child element in the optional <schema-type> element.

The following xfMap sample explicitly instructs the XML Reader to scan the FME features it constructs for the generation of the schema features (however, it is not necessary, since this is the default mode).

<?xml version="1.0"?>
<!DOCTYPE xfMap SYSTEM "xfMap.dtd">
<xfMap>
	
	<schema-type>
		<scan/>
	</schema-type>
	...					 
</xfMap>

If possible, the <generate> or <inline> element should be used instead. This avoids the double read the XML Reader would perform on an XML input dataset when the FME uses it for translation: once for returning schema features, and once for returning the actual FME data features.

The XML Reader will by default skip attributes with empty values when scanning the data features. To scan the attributes with empty values the optional scan-empty-attributes attributes in the <scan> element should be set true. The attribute type for an empty value is scanned as xml_char(1).

<schema-type>
   <scan scan-empty-attributes=”true”/>
 </schema-type>