Re: [sv-ec] "this" as default argument

Hi folks,

Having this sorted out would be useful to us - is anyone willing to take 
a look at this and compose an erratum?

Thanks,
Mike

Rich, Dave wrote:
> I believe the answer to your question has to be yes, but the LRM does
> not support that answer. 'this' is not defined in the scope containing
> the declaration of the method. That rule was created to keep you from
> doing things like:
> 
> function void foo(int A=x, int B=A);
>    int x = bar();
> endfunction
> 
> The rules for determining the order of evaluation would be too complex
> at the very least. However, if you have:
> 
> class A;
> 
> int x;
> function void foo(int x=x);
> ...
> endfunction
> 
> It's semantically clear that the x on the RHS refers to this.x,
> regardless of what identifier is on the LHS because the RHS is evaluated
> in the context of the scope containing the declaration of foo, not
> inside of foo.
> 
> There is an erratum somewhere here.
> 
> 
> Dave
> 
> 
> 
>>-----Original Message-----
>>From: owner-sv-ec@eda.org [mailto:owner-sv-ec@eda.org] On Behalf Of
>>Michael Burns
>>Sent: Thursday, March 30, 2006 3:53 PM
>>To: sv-ec@eda.org
>>Subject: [sv-ec] "this" as default argument
>>
>>
>>Hi folks,
>>
>>Is it the intent of the committee that "this" should be a valid
> 
> default
> 
>>argument value for class methods? For example,
>>
>>class Foo;
>>...
>>
>>  task mytask(Foo f = this);
>>     ...
>>  endtask
>>endclass
>>
>>The 1800 standard seems to suggest this (12.4.3, "The default_value is
>>an expression. The expression is evaluated in the scope containing the
>>subroutine declaration each time a call using the default is made"),
> 
> but
> 
>>section 7.10 makes it a little less clear ("The this keyword shall
> 
> only
> 
>>be used within nonstatic class methods") - does use as a default value
>>in the method's arg list count as being "within" the method?
>>
>>--Mike Burns
>>Freescale Semiconductor