W3C home > Mailing lists > Public > www-dom@w3.org > April to June 2001

Re: expandEntityReferences

From: Joseph Kesselman <keshlam@us.ibm.com>
Date: Thu, 24 May 2001 11:19:46 -0400
To: <www-dom@w3.org>
Message-ID: <OFEACC3460.F42FF53F-ON85256A56.0053090B@pok.ibm.com>

>>  If false, those children  and their descendants will be rejected."
>I could be nitpicking here, but really says that all decendants of entity
>references should be FILTER_REJECTed.

Clearer... though since "Rejected" and "skipped" are defined terms in this
chapter,  I'm not sure  that's a problem.

> If false, the entity reference will appear to have no children."

... Hm. You're right;  as with other filtering it happens at the point
where navigation actually occurs, and if you explicitly do a setLocation
you may jump past that. But I really do want to use the terminology we've
already established in the chapter, so folks can see how
expandEntityReferences compares to and interacts with other kinds of
filtering. So I'd prefer:
     If false, those children will be rejected.
or, if it really needs the detail:
     If false, those children will not be visible (as if blocked by
FILTER_REJECT).
But note that the NodeIterator's documentation does use the terms "skip"
and "reject" later in the same paragraph, so there's already a presumption
that readers understand what they mean.

______________________________________
Joe Kesselman  / IBM Research
Received on Thursday, 24 May 2001 11:20:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:48 GMT