Yes, that sounds correct. This isn't a fancy Java-aware parser, just a
simple utility for flexibility.
We are considering moving to use Apache Commons EL for these sorts of
strings, but it wouldn't be totally backward compatible so it is
something tabled for the future when we want to put in the effort and
after a point where a release is complete or something so that it can
be part of an upgrade path.
There are lots of ways to code the same thing in Java, and usually an
alternative can be found. If not then it is always possible to write a
static Java method to take care of the details.
Please remember that this "feature" is meant for limited use
convenience and it is not really meant to be the way data is prepared
or gathered.
-David
On Jan 30, 2008, at 7:24 PM, X Gylee wrote:
Thx for your response Dimitri. However, as I stated before in my
question, I
have tried replacing the curlies with { and } which are
the same
as { and }. And they didn't work.
In the meantime, I have looked at the source code for the string
parser. Up
to my understanding, it seems that the firstly-found closing-curly is
determined as the closing of the whole ${bsh: ... }, and this makes
any
statement with curlies fail. Could one of the developers verify this?
From: "Dimitri Unruh" <[EMAIL PROTECTED]>
To: [email protected]
Date: Wed, 30 Jan 2008 09:58:13 +0100
Subject: Re: Curly braces in ${bsh; ...}
I guess you should try this one:
{ for {
and
} for }
