🚀 go-pugleaf

In article <49lidc$bjc@natasha.rmii.com>,
Jerry Coffin <jcoffin@rmii.com> wrote:
>fjh@munta.cs.mu.OZ.AU (Fergus Henderson) wrote:
>>>IMHO the most important reason to disallow RT-sized arrays
>>>is that then sizeof also becomes a RT-size.
>
>>A fair point -- but would that really cause any problems?
>
>Unless I'm missing something, a run-time version of sizeof would likely
>have to evaluate its argument, so you'd basically end up with two
>different sizeof's - one that happened at compile time, didn't evaluate
>its argument and allowed use of expressions that can't be evaluated
>legally.

While the essence here is true, I don't think it is a real problem.  The
compiler must be able to detect a variable of type dynamic array (as
I'm late to this discussion, I am assuming dynamic arrays are arrays
allocated on the stack of variable size) and thus infer that it must 
compute the sizeof said array at run-time.  Which basically, it has to
do anyway under these assumptions.

--
Eric



---
[ comp.std.c++ is moderated.  Submission address: std-c++@ncar.ucar.edu.
  Contact address: std-c++-request@ncar.ucar.edu.  The moderation policy
  is summarized in http://dogbert.lbl.gov/~matt/std-c++/policy.html. ]