creating a publication (or book) from multiple XML files

twfurst · June 24th, 2010, 07:57 PM

I have a question regarding how to combine multiple files into a single publication, or book.

To start with, I will have five different XSL-FO styles sheets (right now three are complete) for these XML files (of varying numbers), depending on which schema validates them. Now, these files when combined will comprise a document. They are referenced in what is a "publication module". This is just another XML file that gives the identifying information (not including file extension) for each of the XML files that make up that publication.

I am not sure how I should go about using this "publication module" to assemble a book. As it is now, each stylesheet for each different schema type has it's own headers and footers, which need to remain as is, but further information will need to be added, such as the publication number and so on.

I have done some reading regarding the use of the document() function, but am not sure if this the correct route to take. Should I create a stylesheet that includes the other stylesheets and uses this function?

If anyone has any ideas, or has come into a similar situation, the help will be very appreciated.

Thank you

bonekrusher · June 26th, 2010, 06:42 AM

If you are using a "publication module" like in s1000d than the document() function works great. I use a "Pipe Line" methodology. I write my xslt to process all my files into a temporary document or multiply temporary documents, then covert them to xsl-fo, then send them through a xsl-fo processor.

Something like this:

Code:

<xsl:variable name="tempDoc">
	<xsl:for-each select="document('some.xml')">
		<xsl:for-each select="root">
			<xsl:element name="newRoot">
				<xsl:copy>
					<xsl:copy-of select="@*"/>
					<xsl:apply-templates/>
				</xsl:copy>
			</xsl:element>
		</xsl:for-each>
	</xsl:for-each>
</xsl:variable>
<xsl:value-of select="$tempDoc/newRoot"/>

twfurst · June 26th, 2010, 12:33 PM

S1000D is exactly what I am working with. So, to make sure I followed what you are saying. You read each XML file, based on the publication module, into a new document, and ten convert that to FO and then process?

bonekrusher · June 26th, 2010, 01:36 PM

Exactly. You will have some challenges processing the <refdm>s that are not required in your publication module. Also, resolving xrefs across data modules when the author uses duplicate @ids.

Fun stuff... Email me directly with any S1000D question.

Regards,

Phil