Re: gfortran patch for PR28974 Slow compilation of data statements.

[Jerry DeLisle <jvdelisle at verizon dot net>]

Bud Davis wrote:
> --- Jerry DeLisle <jvdelisle@verizon.net> wrote:
>  
> > While working pr24978, 
>
> jerry,
>
> that number sounds familiar.  i looked at that a while
> back.  if i remember correctly, it turned out to be
> initializing the same element more than once....which
> already is accepted for some kinds of data
> types....and not for others.
>
> to sum it up, i gave up when i saw how much work it
> would be.
>
> my hat is off to you !!! 
>
>
> --bud
Thats exactly it.  There are several places that have to be checked and the 
right error/warning emitted.  I have managed to catch my 6 test cases so far, 
but regressed a couple of places where we error when we shouldn't.  I just need 
to keep at it.  One case thats tricky is an array of user defined type that has 
an array in it.  The same element of the first array is repeated but the 
sublevel is not.  Its interesting.

Jerry