Back to Search Back to Search Results

DITA XSLT Conventions

The built-in DITA XSLT modules follow a number of common conventions.
  • The @class attribute in HTML output elements is comprised of:
    • Every tag name in the DITA specialization hierarchy for the source element.
    • The value of the @outputclass attribute, if any.
    For example, since <b> is a specialization of <ph>, <b outputclass="superBold"> would become <b class="ph b superBold">. This allows you to do most of your DITA styling using CSS instead of XSLT.
  • For the most part, the templates are configured in both the #default and html modes. The exception is <map> and <topicref> processing, which is generally processed using mode="toc", switching for mode="html" for visible content, like <navtitle> content.
  • Conditional processing attributes - @product, @platform, @audience, @otherprops, and any @props specializations identified in the @domains attribute of the root element - will be preserved in HTML output, prefixed with @data- (for example, @data-audience).
  • All templates in these modules have a negative @priority, so that they can be easily overridden by stylesheets that reference them.
  • @id attributes are converted into named anchors of the form <a name="{@id}" class="xmlId"/>
  • The @xtrc and @xtrf attributes, generated during processing to identify the source of a given XML element, are represented as @data-xtrc and @data-xtrf in the HTML output.
    Note: The Comment Manager uses @data-xtrc and @data-xtrf to trace comment locations back to source XML.
  • @xml:lang attributes are converted to @lang, and @dir is preserved.

Related information:

Comment Management with the Titania Annotator Plugin
Parent topic: Writing XSLT for DITA