This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: PR 60414: Patch proposal
- From: dominiq at lps dot ens dot fr (Dominique Dhumieres)
- To: fortran at gcc dot gnu dot org
- Cc: vehre at gmx dot de
- Date: Mon, 21 Jul 2014 09:26:05 +0200
- Subject: Re: PR 60414: Patch proposal
- Authentication-results: sourceware.org; auth=none
> I was contracted to fix some issues listed in the bugtracker for fortran.
GREAT! For how long? What will be your strategy to pick bugs among the
765 ones?
> I have to apologize for the incomplete patch: The testcase
> unlimited_polymorphism_18.f90 was missing. My fault. Please find the corrected
> patch attached.
While your test is suitable for visual inspection, it is not for the automated
test suite: the results should be compared to the expected ones and the executable
should abort if there is any mismatch. For instance
write (*, *) "???"
should be replaced with
call abort
(or abort(), abort (), ..., there are variations in the test suite).
Replacing the other 'write (*,...' will be slightly more difficult.
For 'FWriteArr' this can be done with something such as
if(any(X /= (/ 0, 1, 2, 3, 4, 5 /)) call abort
For 'FWrite' it will probably needed to pass the expected value of X.
Although I have tested your patch, I did not try to fix the test (so
the above is untested).
Last point, the second line of your comment in fortran/interface.c
could be improved: I have read it several time and still does not
fully understand it.
Cheers,
Dominique