ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steve Loughran <ste...@apache.org>
Subject Re: task namespaces
Date Mon, 12 Sep 2005 15:08:24 GMT
jfuller@ruminate.co.uk wrote:

> there are many ways to skin this cat....and I have done em all. I think we will all wish
in the near future that anything that consumes XML has such a default ignore unknown xml handling,
just makes life a little easier when interoperating.

maybe, maybe not.

the original HTML browsers @ cern used to ignore unknown elements, which 
caused trouble when <i> came along, as something like

<p>
The neutrino flux generated by the WA141 experiment is <i>not</i> 
considered a health risk to people in the Jura mountains.
</p>

Would appear in the wrong browsers, without the "not", leading to a 
complete inversion of meaning. Hence the current default of 'silently 
ignore unknown tags in HTML". Which is why all things like inline 
javascript have to comment themselves out, so the browser would ignore 
them.


Returning to Ant, now that it has namespaces, you can in theory mix it 
with other stuff, be it nest it in a SOAP payload, a CDDLM deployment 
descriptor request sent over SOAP (that is somewhere at the far end of 
my todo list), or with some RDF metadata. The latter is an interesting 
use, and if we do some kind of per-namespace element handoff, 
permittable without too much trauma.

Actually, still on the subject of namespaces, I think we need better 
diagnostics there.
1. something to recognise that there is a type declared of that name in 
a different namespace, to catch namespace errors
2. <diagnostics>/-diagnostics to list types declared by namespace.

-steve

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org
For additional commands, e-mail: dev-help@ant.apache.org


Mime
View raw message