jakarta-taglibs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael Nascimento Santos" <mister...@hotmail.com>
Subject Re: Another bug, spec issue
Date Fri, 19 Dec 2003 18:20:27 GMT
In this blog entry (http://weblogs.java.net/pub/wlg/730 ) I explain why it
is correct and cannot be changed. Code that relies on spec being right is
simply compliant and we cannot make it not work just because it feels
natural do so. As you said, JSP 2.0 corrects the issue. So, you have two
possibilities, that being:

- Working with JSTL 1.0, take the EL interpreter source code, change it so
it works the way you want and compile your own version of it;
- Upgrade to a J2EE 1.4 container

My point about it is: we cannot force users who actually read the specs and
whose code relies in this (unnatural, but specified) behaviour to rewrite it
just because the mandated behaviour seems bad. Specs are there, to make good
and bad things, but they at least make things standardized. As it is an
open-source implementation, you can always roll you own (uncomplaint)
version of it.

[]s
Michael Nascimento Santos
JSR 207 Expert Group Member
http://today.java.net/pub/au/80
Sun Certified Programmer for the Java 2 Platform
Sun Certified Web Component Developer for J2EE
Moderador SouJava - http://www.soujava.org.br

----- Original Message ----- 
From: "Serge Knystautas" <sergek@lokitech.com>
To: "Tag Libraries Developers List" <taglibs-dev@jakarta.apache.org>
Sent: Friday, December 19, 2003 12:49 PM
Subject: Another bug, spec issue


> It looks like others have realized this based on the changes in EL (as
> defined by JSP 2.0), but see what you guys think of the short-term
> solution...
>
> Right now in JSTL 1.0, the EL operator "empty" is not defined for
> Collection.  It is defined for Map and List, but nothing specified for
> collection, so it falls to the default result of false.  This is
> corrected in the JSP 2.0 spec.
>
> I understand it could be viewed as a spec violation to have a JSTL 1.0
> implementation properly apply the empty operator to Set (or other
> collection).  My only response is that anyone who has code that applies
> empty to a Set and expects this to be always false (intentional or
> otherwise) has a bug (intentional or otherwise) and could benefit from
> this change.
>
> So, I propose changing the JSTL 1.0 implementation to have EL's empty
> operator treat Collections the same way it does Lists.  Thoughts?
>
> -- 
> Serge Knystautas
> President
> Lokitech >> software . strategy . design >> http://www.lokitech.com
> p. 301.656.5501
> e. sergek@lokitech.com
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: taglibs-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: taglibs-dev-help@jakarta.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: taglibs-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: taglibs-dev-help@jakarta.apache.org


Mime
View raw message