On Wed, 9 Jun 2004, David Sewell wrote:
> I see two strategies for dealing with this:
> 1. Treat <app> strictly as a mixed-content-containing element and allow
> linebreaks within it only where they correspond to word boundaries.
> Be very careful not to reformat the source XML.
> 2. Use the "type" attribute on <rdg> to distinguish between whole- and
> partial-word readings, and break lines with wild abandon.
> Thoughts on which is better practice?
Is wrapping the whole word in a <w> element and then
normalize-space() or strip-space of that element not
an option as well? I've only used <app><rdg> with the
smallest granularity of a single word, so I'm also
interested in what others thing. I would prefer 2) if
one of the above. But you could use a combination of
normalize-space and translate to just cut out any line
breaks in a single-word app, as well?
Dr James Cummings, Oxford Text Archive, University of Oxford
James dot Cummings at ota dot ahds dot ac dot uk