XML Memory Requirements (was Re: Feeling good about SML)

[David Megginson]

"Michael Champion" <Mike.Champion at softwareag-usa.com> writes:

> Again, this is the voice of inexperience speaking so be merciful if I'm
> missing something obvious ... but in an "SML" without entities, attributes,
> PIs, and comments, and without any concern for maintaining SGML
> compatibility,  I can't think of too many printable characters besides "<"
> ">" and maybe " " that *have* to be illegal in element names. I can think of
> a lot of characters that sensible people probably would not put in their
> markup, but that's another matter, and arguably not the job of the SML
> parser to enforce.

I think that this is a lot more on the money than most of the SML
discussion -- if we want to be able to build really, really small XML
processors, we're going to have to lighten the error-reporting
requirements quite a bit, somehow.

Actually, it's not that expensive to be picky about characters 0x00 to
0xff, since the tables will be small; above 0xff, things become
trickier if you're trying to keep your parser's total footprint small
enough for embedded systems (say, <20K including both code, stack, and
dynamic memory, which is otherwise quite doable with XML 1.0).  Unless
embedded systems get a lot bigger, the error-reporting requirement is
going to be the real hurdle to implementation.


All the best,


David

-- 
David Megginson                 david at megginson.com
           http://www.megginson.com/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)