Edwin Goei wrote:
> The DocumentBuilderFactory#[gs]etAttribute() methods are implementation
> dependent so you can define them to do whatever you want them to do.
Then I'm in favor of mapping this method to the underlying
get/setFeature/Property methods in the parsers. Shouldn't
be too much work, even with the Object mapping to a boolean:
if (value instanceof Boolean) {
parser.setFeature(name, ((Boolean)value).booleanValue());
}
else {
parser.setProperty(name, value);
}
It isn't *that* easy, though, because the attribute methods
are on the factory and not on the parser instance interface.
But I still think it's not too bad.
> However, I think we should try to avoid possible "name" argument clashes
> with other implementations, which should easily be possible.
This isn't really a problem, I think, because the JAXP
spec suggests using Java package/class naming for new
attributes which are completely different from the URIs
used by the rest of the world. (Can't seem to find the
text at the moment but I remember reading it.)
Anybody want to do this work? If not, I'll do it.
--
Andy Clark * IBM, TRL - Japan * [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
