Robert Whalen wrote:
> How might I include in my P5 DTD the Unicode entity references from my P4
> DTD? Unlike the old Pizza Chef, Roma does not seem to have a mechanism for
> including these. And because my project requires numerous entries of long
> esses, ligatures, carets, etc., I'd prefer not having to type a numerical
> code point every time. Is there a simple way to convert the old entity refs
> and either insert them into the P5 DTD or create an extensions file
There is no mechanism provided, because it would be DTD-specific.
The recommendation could be to expand the Unicode entities
in the source text and work in future with a UTF-8 capable
Alternatively, you can generate the DTD and then
add the extensions file in by hand, or include
it in each instance file.
a) any XML processing you do on your
P4 texts will almost certainly expand your entities anyway.
If you want them preserved, you'll need to do
b) XML processors are not obliged to read the external DTD,
so having entities in the DTD only can cause your documents
to be invalid
c) the schema languages (W3C Schema and RELAXNG) don't
support these character entities, so you won't be able to take
advantage of them
All in all, annoying as it is, I suggest you bite the bullet
and drop character entities entirely. You can always
use the new <g> element for obscure stuff.
Personally, I use Emacs most often, with an addon
which expands — as I type it into the right
Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431
OSS Watch: JISC Open Source Advisory Service