[fortran, patch] Fix PR15959, character array initialization.

[Feng Wang]

--- Steven Bosscher <stevenb@suse.de> wrote:
> On Apr 01, 2005 11:45 AM, Feng Wang <wf_cs@yahoo.com> wrote:
> > This patch fix pr15959. If we resolve a constant character array, update
> the
> > expression's character length with the maximun of its constructor. And set
> the
> > initializer's length to symbol.
> > 
> > Tested with no regression on i686. Ok for mainline and gcc40?
> 
> Mostly looks OK.  A few points:
> 
> ! 	  max_length = (p->expr->value.character.length > max_length) ?
> ! 		p->expr->value.character.length : max_length;
> 
> I believe we have a MAX macro for this.

Yes. Changed.

> 
> ! 	else
> ! 	  return;
> 
> Why return here, can you add a comment?  Can this happen at all,
> that we have a non-EXPR_CONSTANT in an character array initializer?

This is not only for initializer. We may have variable character array
constructor. I will add comment for this. Example from
gfortran.dg/string_ctor_1.f90:
subroutine test (n, t, u)
  integer n
  character(len=n) :: s(2)
  character(len=*) :: t
  character(len=*) :: u
  ! A variable array constructor.
  s = (/t, u/)
end subroutine

> 
> 
> ! 	      p->expr->ts.cl->length = gfc_int_expr (max_length);
> 
> Maybe a really silly question, but I suppose we pad the strings
> that are shorter?  Could you add a comment here about where that
> happens?
> 
Yes, it is a really question. In fact we now don't pad parameter string. But I
think it is another bug. Example follows:

character(len = 10),parameter:: a = 'hello'
print *, a, "world"
end

And I report this bug as PR20713.



Best Regards,
Feng Wang

_________________________________________________________
Do You Yahoo!?
150ÍòÇúMP3·è¿ñËÑ£¬´øÄú´³ÈëÒôÀÖµîÌÃ
http://music.yisou.com/
ÃÀÅ®Ã÷ÐÇÓ¦Óо¡ÓУ¬ËѱéÃÀͼ¡¢ÑÞͼºÍ¿áͼ
http://image.yisou.com
1G¾ÍÊÇ1000Õ×£¬ÑÅ»¢µçÓÊ×ÔÖúÀ©ÈÝ£¡
http://cn.rd.yahoo.com/mail_cn/tag/1g/*http://cn.mail.yahoo.com/event/mail_1g/