Or you could just add the arbitrary attribute as an
attribute in a customization, Eliot, then maintain a transform to turn
it into a data element with a "name" attribute of the same name if you
want to produce valid DITA for export.
That's what I'm doing in a customized subset of DITA I'm implementing -
plus changing a few element names to reflect local locutions, and
maintaining transforms for them as well.
It seems such a process should be generalizable in future for adding
arbitrary attributes, generalizing them as data elements, then
respecializing them as attributes - perhaps by adding some flags to the
data element to govern the process.
--Dana
W. Eliot Kimber wrote:
Michael
Priestley wrote:
> Or have I missed something?
<data> might be a better choice than <ph> but otherwise I
think you are correct.
Data, yes, that might be better.
Cheers,
E.
