On Wed, Apr 15, 2020 at 1:47 PM Thomas Koenig wrote:
>
> Hi Fritz,
>
> > While you're touching the code anyway, how would you feel about
> > replacing the nearby "goto done"s with a chain of "else if"? There's
> > really no reason I can see for goto here, since the block following
> > the conditio
Hi Fritz,
While you're touching the code anyway, how would you feel about
replacing the nearby "goto done"s with a chain of "else if"? There's
really no reason I can see for goto here, since the block following
the conditions is already "done".
I think this would really be pushing things at st
> Yes. Looking back at the code, I think it can also be cleaned up
> a little - turning the error to warnings is only needed on that
> particular branch, and resetting it to the default can also
> happen there, and at the target of a goto statement.
>
> So, here's an updated patch. OK for trunk?
Am 15.04.20 um 12:33 schrieb Tobias Burnus:
On 4/15/20 12:26 PM, Thomas Koenig via Fortran wrote:
+ /* Turn erros into warnings with -std=gnu and -std=legacy,
Only glanced at it – but can you also fix the old* typo "erro(r)s"?
Yes, I think I can manage that :-)
Regards
Thomas
On 4/15/20 12:26 PM, Thomas Koenig via Fortran wrote:
+ /* Turn erros into warnings with -std=gnu and -std=legacy,
Only glanced at it – but can you also fix the old* typo "erro(r)s"?
Tobias
(*old as the comment block has been moved around)
-
Mentor Graphics (Deutschland
Hi Fritz,
I wonder: could you simply replace the gfc_error_opt(0, ...) call with
gfc_error?
Yes. Looking back at the code, I think it can also be cleaned up
a little - turning the error to warnings is only needed on that
particular branch, and resetting it to the default can also
happen there
On Mon, Apr 13, 2020 at 10:20 AM Thomas Koenig via Fortran
wrote:
>
> Hello world,
>
> the attached patch fixes an ICE on invalid: When the return type of
> a function was misdeclared with a wrong rank, we issued a warning,
> but not an error (unless with -pedantic); later on, an ICE ensued.
>
> N
