Re: [xml] What is the best way to implement semantics?

John J. Boyer wrote:
In my xmll to braille program I have libxml2 create a parse tree of the document. I have written a basic tree-traversal routine. What is the best way to implement semantics, that is, to interpret elements and their attributes? Is there code or documentation that I should be looking at?

I am thinking of imllementing Docbook and Daisy directly. How can I tell which one I have?

from the Doctype if present; I guess that's getIntSubset(doc)
(I use the perl bindings, myself: $doc->internalSubset; )

Other xml formats will be converted to one of the above two using xslt. Is this realistic?

Noting that I know next-to-nothing about braille and how you can control it
to convey information (taller dots for emphasis?  Prefixing "The title is: "...?)
-- and this isn't really the forum to go into it.

Anyway, a quick google suggests that Daisy is designed exactly for this,
but I'm not sure what implementing Docbook directly buys you.
Docbook is a good model for some kinds of documents but a bad match to others,
so you'd end up with a questionable translation to Docbook.
Why not transform all document types, including Docbook, to Daisy and working
with that?

bruce miller nist gov

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]