Paulo Gaspar wrote:
[SNIP]
> But it does not work because of that wrapper object. I only went
> as far as to find out that it was some private hard-to-identify
> object - I could not see its members in the debugger- but I
> still did not trace the problem to its foundation. That is what
> I want to do now.
I know the problem - it was a decision that I had to make a while ago,
and did to ensure backwards compatibility. As time has shown, it's more
important to go the other way. We've been through this before.
Later today, I'll post a summary and change the code in HEAD, so you can
give it a whirl.
>
> BTW, I tried to see the wrapper members because I was trying to
> find how to get the wrapped object. Since the debugger could not
> even get the class from the object, I kept not knowing where to
> start. Is it some anonymous inner class or something like that?
>
> Now, I think that
> 1. The #if() directive should resolve a reference that returns
> null to as being false;
> 2. Macros should unwrap arguments before passing them to methods
> called inside themselves.
>
> While 2 looks clear to me, but maybe you think different about 1.
> Anyway, I think 1 is the most intuitive way to go.
>
The issue with 1) is that #if() should behave correctly in cases like
this. 2), no, because no one should know.
geir
--
Geir Magnusson Jr. [EMAIL PROTECTED]
System and Software Consulting
Developing for the web? See http://jakarta.apache.org/velocity/
You have a genius for suggesting things I've come a cropper with!
