This is the mail archive of the
mailing list for the GCC project.
Re: [patch,fortran] Allow namelist with internal file (PR 28224)
- From: Tobias Burnus <burnus at net-b dot de>
- To: Paul Thomas <paulthomas2 at wanadoo dot fr>
- Cc: gcc-patches at gcc dot gnu dot org, fortran at gcc dot gnu dot org
- Date: Sat, 28 Oct 2006 18:36:11 +0200
- Subject: Re: [patch,fortran] Allow namelist with internal file (PR 28224)
- References: <email@example.com> <45437C57.firstname.lastname@example.org>
Paul Thomas schrieb:
> Might it not be better to use -std=f95 for this test....
>> * gfortran.dg/namelist_internal.f90: New namelist test.
>> * gfortran.dg/namelist_internal2.f90: New namelist test.
> ...and to drop this latter?
Well, the first test test whether internal file read/write of namelists
The second test tests whether it is still forbidden for -std=f95.
I can drop either test, but I cannot simply change -std=f2003 to
-std=f95 in the first test; doing all the modifications needed, I will
end up with the second test.
That is anyhow the question: How much should one test? Only a minimal
set, which will get easily broken or also more extensively the behaviour
of a function, even if it is unlikely that a regression occurs?
The former prevents bloating, the latter reduces the chance of more
In this regard, shall I check in namelist_internal.f90 or
namelist_internal2.f90 or both?
(If only namelist_internal2.f90, I will rename it to namelist_internal.f90.)