XOM API for Processing XML with Java

["Regier Avery J" <RegierAveryJ AT JohnDeere.com>]

You can first check to see if 'a' has *any* children.  If it doesn't then you 
don't need to climb back up to the root of 'c' to look for 'a'.  In cases 
where you are building from the bottom up, you won't have to climb back to 
the root for any append.  You'll only do the check if 'a' already has 
children.

-Avery

-----Original Message-----
From: John Cowan [mailto:jcowan AT reutershealth.com]
Sent: Friday, September 12, 2003 4:43 PM
To: Elliotte Rusty Harold
Cc: xom-interest AT lists.ibiblio.org
Subject: Re: [XOM-interest] Node hierarchy does not prevent cycles


Elliotte Rusty Harold scripsit:

> Is there any way to do this without appending the node itself? Hmm, 
> seems like there might be:
> 
> a.appendChild(b);
> b.appendChild(c);
> c.appendChild(a);
> 
> This is a problem. It seems like it could require insertChild descend 
> the entire subtree to  verify the insertion is OK. Ugly. Does anyone 
> see a simple, fast way to fix this?

"Piffle.  A trifle."
        --the incredibly arrogant Eastern Empire wizard from Mercedes Lackey 
books

Ascend the "this" object's parent chain for the proposed child object.
If it's there, throw.  If you reach the Document object safely, all is well.
The first two calls above will be fine because b is not an ancestor of a
and c is not an ancestor of b.  But since a *is* an ancestor of c, 
sproinggg....

-- 
"But the next day there came no dawn,           John Cowan
and the Grey Company passed on into the         jcowan AT reutershealth.com
darkness of the Storm of Mordor and were        http://www.ccil.org/~cowan
lost to mortal sight; but the Dead              http://reutershealth.com
followed them.          --"The Passing of the Grey Company"
_______________________________________________
XOM-interest mailing list
XOM-interest AT lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/xom-interest