--===============0223996616==
Content-Type: multipart/signed;
boundary="nextPart1775399.FXSF6THoVQ";
protocol="application/pgp-signature";
micalg=pgp-sha1
Content-Transfer-Encoding: 7bit

--nextPart1775399.FXSF6THoVQ
Content-Type: text/plain;
charset="iso-8859-6"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Am Freitag, 24. Februar 2006 19:23 schrieb Kevin Krammer:
> On Friday 24 February 2006 18:37, Arnold Krille wrote:
> > But the problem is that A's interface should be compatible in the Qt-way
> > (thats why the slots and signals are setValue(...) and valueChanged(...=

))
> > and that B is a Qt-class (QSpinBox, QLineEdit, QDoubleSpinBox to be
> > precise).

> If the value received in the slot is the same the object already has, why
> would it emit the signal if not change happend?


Because sender and receiver are in different threads and when QSpinBox gets=
=20
the former new value back it is already at the next value because the user =
is=20
spinning the mouse-wheel and the connection is not a direct one...

> Qt classes don't do this, you can for example connect a slider and a
> spinbox to each other and get no loop at all


At least in Qt3 this wasn't true, now with Qt4 I always protect my own clas=
ses=20
against direct loops. And now I am searching for doing this in a=20
multi-thread-safe way...

Arnold

=2D-=20
Get my public-key from pgp.mit.edu or pgp.uni-mainz.de

--nextPart1775399.FXSF6THoVQ
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQBEAJYMuYLL1cDjHx0RAmInAJ4vgwfeUr2P6wWenVU3rF megIqNAQCggSf0
dSc43ynvAquPBDyUGD/JMwM=
=WI6O
-----END PGP SIGNATURE-----

--nextPart1775399.FXSF6THoVQ--

--===============0223996616==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline


>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<


--===============0223996616==--