saxon.character.representation

["Bill Burns" <bburns@verbumcomm.com>]

Hello,

 

I'm working with a customization layer based on a dated version of DocBook XSL (HTML  1.41?) and using the 1.74 release HTML style sheets for everything else.

 

This is for a home-grown WebHelp solution that I'm in the process of internationalizing (mostly complete, except for the Lucene search pieces). In our customization, we set the following parameters:

 

<xsl:param name="chunker.output.encoding" select="'UTF-8'"/>

 

<xsl:param name="saxon.character.representation" select="'entity;decimal'"/>

 

My assumption would be that the UTF-8 output encoding would make the saxon.character.representation value mostly unnecessary. However, all non-ascii characters are being converted to numeric entities. Changing the saxon.character.representation value to native converts to native non-UTF-8 charsets, so that doesn't help matters.

 

Any suggestions on where to look for a fix? Is there some other place in the default style sheets (besides chunker.xsl) where the encoding is (or could be) overridden?

 

Thanks,

 

Bill Burns

Verbum Communications, Inc.

+1.208.336.6081

bburns@verbumcomm.com

http://www.verbumcomm.com