From kde-core-devel  Mon Jul 04 17:31:09 2011
From: Michael Pyne <mpyne () kde ! org>
Date: Mon, 04 Jul 2011 17:31:09 +0000
To: kde-core-devel
Subject: Re: -Wunused-but-set-variable warnings
Message-Id: <2673077.CkjUNdZycd () midna>
X-MARC-Message: https://marc.info/?l=kde-core-devel&m=130980070810783
MIME-Version: 1
Content-Type: multipart/mixed; boundary="--nextPart2933431.jescoY5SxN"


--nextPart2933431.jescoY5SxN
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="utf-8"

On Monday, July 04, 2011 12:02:47 Dawit A wrote:
> The following files all contain set but unused variables:
>=20
> kdelibs/kdecore/util/kshareddatacache.cpp:751:14: warning: variable
> =E2=80=98freedPagesRequired=E2=80=99 set but not used [-Wunused-but-s=
et-variable]

Thanks for the heads-up on this one!

> Unlike the -Wunused-parameter fixing this warning messages requires
> context because the variable may be set and unused due to a mistake
> that can potentially be causing a bug.

In this case it was not, but as you point out that was only due to othe=
r=20
things going on afterwards.

> As such can kdelibs cmake file
> be changed to error out, -Werror=3Dunused-but-set-variable, for such
> warnings ?

I really kind of abhor -Werror in all of its incarnations. It's like ho=
lding a=20
whole village hostage until one man gives themselves up to the posse. I=
n this=20
case it would affect /anyone/ trying to simply build KDE (even if they =
have=20
more advanced compilers with more warnings generated than the developer=
 who=20
would need to make the fix).

Being unable even to build kdelibs, kde-workspace, etc. would make it v=
ery=20
difficult for other developers to properly build and *test* the code th=
at they=20
are working on elsewhere. So while we should certainly apply social pre=
ssure=20
to compile without warnings and ensure bug-free code, don't overdo it! =
The=20
likely thing to come from such a decision is that the warnings are "fix=
ed" by=20
hastily-applied patches that either do not fix the problem, or possibly=
 even=20
introduce new ones.

I almost broke KSharedDataCache in a way I wouldn't have been able to e=
asily=20
test until I took a bit more time to think about what the code was actu=
ally=20
doing now, and needed to be doing.

Regards,
 - Michael Pyne
--nextPart2933431.jescoY5SxN
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (GNU/Linux)
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=+0fJ
-----END PGP SIGNATURE-----

--nextPart2933431.jescoY5SxN--