Re: New DOM Level 2 Working Draft

Re isIgnorableWhitespace depending on DTD Information not currently modelled in
the DOM:

>The same is true for the "isSpecified" property of an attribute, which
>is already part of DOM

Granted. In my implementation, I extended the DOM model to handle this.

>Why just "parsers" ?? [setting a flag]

Not just parsers, agreed. But setting that flag shouldn't be part of the DOM API
per se, since it really should be derived information. The DOM API _isn't_
testable in its own right at this time; there are a couple of other DTD-related
corners where a value is set without any explicit reference to how it gets set,
or how an object is created.

Re the readonly flag: As far as I know, the only readonly nodes are children of
an EntityReference node. Whether that's set by a parser or not depends on
whether your DOM leaves the construction of those children as a parser task or
fills them in automagically when the EntityReference is accessed.

>a set of inconsistent judgements

I wasn't around at the time, but I'd guess the problem is that some of the past
items -- such as isSpecified -- are left over from the period when DOM Level 1's
working drafts had more DTD support than wound up in the final Recommendation,
and that the inconsistancy was in trying to hold onto them in the absence of a
DTD context.

______________________________________
Joe Kesselman  / IBM Research

Received on Tuesday, 28 September 1999 13:12:28 UTC