[jdom-interest] CDATA sections?

[Wesley Biggs]

More control to the user.  More predictable output.

I agree with you that there's nothing about the nature of the XML document
that is being thwarted if there is no CDATA support.  It's just a measure of
convenience to the programmer.

Besides, if it's not in the core API some folk (you know who you are) are
just going to subclass Element anyway.

And last but not least, nothing starts a war like trying to determine the
metrics for "sufficiently ugly". :-)

CDATA, CDATA run..

Wes

-----Original Message-----
From: Jason Hunter
To: Wesley Biggs
Cc: jdom-interest at jdom.org
Sent: 6/1/00 7:13 PM
Subject: Re: [jdom-interest] CDATA sections?

> While I'm still not sure everyone is talking about the same thing, 
> I agree that it would be a powerful feature if, as Kevin suggests, a
> hand-built JDOM tree could choose, on an element by element basis,
> whether or not it should be output as CDATA.

Why is this better than having an Outputter with logic that if the
content is sufficiently "ugly" it will use a CDATA section instead of
escapes?