No subject
Wed Oct 7 22:37:18 EDT 2009
contemporary distros would it make any sense at all to perhaps maintain=20
backward compatibility for some small window of time? eg: if in early 2009=20
qt 4.5 is commonly used by the KDE distro's then for all of 2010 paraview=20
supports at a minimum qt4.5.
If it were just for developers I'd say no, but since anyone who wants to=20
take advantage of parallel pv has to compile. this type of thing has=20
potential to affect a bunch of users as well.=20
Utkarsh Ayachit wrote:=20
Folks,
As we are approaching 3.8 branch, what does everyone think of making
Qt 4.6 the officially supported version? Quite a few of us have been
using Qt 4.6 for quite some time now and don't think there are any
issues.
Utkarsh
_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org <mailto:Paraview-developers at paraview.org>=20
http://public.kitware.com/mailman/listinfo/paraview-developers <
http://public.kitware.com/mailman/listinfo/paraview-developers>=20
=20
_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org <mailto:Paraview-developers at paraview.org>=20
http://public.kitware.com/mailman/listinfo/paraview-developers <
http://public.kitware.com/mailman/listinfo/paraview-developers>=20
_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org
http://public.kitware.com/mailman/listinfo/paraview-developers
Ce message et toutes les pi=E8ces jointes (ci-apr=E8s le 'Message') sont=20
=E9tablis =E0 l'intention exclusive des destinataires et les informations q=
ui=20
y figurent sont strictement confidentielles. Toute utilisation de ce=20
Message non conforme =E0 sa destination, toute diffusion ou toute=20
publication totale ou partielle, est interdite sauf autorisation expresse.
Si vous n'=EAtes pas le destinataire de ce Message, il vous est interdit de=20
le copier, de le faire suivre, de le divulguer ou d'en utiliser tout ou=20
partie. Si vous avez re=E7u ce Message par erreur, merci de le supprimer de=20
votre syst=E8me, ainsi que toutes ses copies, et de n'en garder aucune trac=
e=20
sur quelque support que ce soit. Nous vous remercions =E9galement d'en=20
avertir imm=E9diatement l'exp=E9diteur par retour du message.
Il est impossible de garantir que les communications par messagerie=20
=E9lectronique arrivent en temps utile, sont s=E9curis=E9es ou d=E9nu=E9es =
de toute=20
erreur ou virus.
____________________________________________________
This message and any attachments (the 'Message') are intended solely for=20
the addressees. The information contained in this Message is confidential.=20
Any use of information contained in this Message not in accord with its=20
purpose, any dissemination or disclosure, either whole or partial, is=20
prohibited except formal approval.
If you are not the addressee, you may not copy, forward, disclose or use=20
any part of it. If you have received this message in error, please delete=20
it and all copies from your system and notify the sender immediately by=20
return message.
E-mail communication cannot be guaranteed to be timely secure, error or=20
virus-free.
_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org
http://public.kitware.com/mailman/listinfo/paraview-developers
**** Kenneth Moreland
*** Sandia National Laboratories
***********=20
*** *** *** email: kmorel at sandia.gov
** *** ** phone: (505) 844-8919
*** web: http://www.cs.unm.edu/~kmorel
_______________________________________________
Paraview-developers mailing list
Paraview-developers at paraview.org
http://public.kitware.com/mailman/listinfo/paraview-developers
Ce message et toutes les pi=E8ces jointes (ci-apr=E8s le 'Message') sont =
=E9tablis =E0 l'intention exclusive des destinataires et les informations q=
ui y figurent sont strictement confidentielles. Toute utilisation de ce Mes=
sage non conforme =E0 sa destination, toute diffusion ou toute publication =
totale ou partielle, est interdite sauf autorisation expresse.
Si vous n'=EAtes pas le destinataire de ce Message, il vous est interdit de=
le copier, de le faire suivre, de le divulguer ou d'en utiliser tout ou pa=
rtie. Si vous avez re=E7u ce Message par erreur, merci de le supprimer de v=
otre syst=E8me, ainsi que toutes ses copies, et de n'en garder aucune trace=
sur quelque support que ce soit. Nous vous remercions =E9galement d'en ave=
rtir imm=E9diatement l'exp=E9diteur par retour du message.
Il est impossible de garantir que les communications par messagerie =E9lect=
ronique arrivent en temps utile, sont s=E9curis=E9es ou d=E9nu=E9es de tout=
e erreur ou virus.
____________________________________________________
This message and any attachments (the 'Message') are intended solely for th=
e addressees. The information contained in this Message is confidential. An=
y use of information contained in this Message not in accord with its purpo=
se, any dissemination or disclosure, either whole or partial, is prohibited=
except formal approval.
If you are not the addressee, you may not copy, forward, disclose or use an=
y part of it. If you have received this message in error, please delete it =
and all copies from your system and notify the sender immediately by return=
message.
E-mail communication cannot be guaranteed to be timely secure, error or vir=
us-free.
--=_alternative 005D36AAC12576C4_=
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="iso-8859-1"
<br><font size=3D2 face=3D"sans-serif">I don't know about Qt forward compat=
ibility
issues, but the point Utkarsh made is that PV will support the same version
of Qt for two releases, which is a clear enough policy. If Qt is good at
keeping forward compatibility, then this will extend the life of the "=
minimum"
Qt version, but those that use a version < official know that they have
a version that can be dropped anytime. The minimum is only here as an infor=
mation,
and the developer that makes the change that increment the minimum should
also modify the cmake command accordingly. </font>
<br><font size=3D2 face=3D"sans-serif">Utkarsh, to help the developers in t=
his
task, could you also always provide a dashboard on the "minimum"
version, so that changes that increment it will be discovered as soon as
possible?</font>
<br>
<br><font size=3D2 face=3D"sans-serif">By the way, do paraview have an offi=
cial
policy on the interval between two releases?</font>
<br>
<br><font size=3D2 face=3D"sans-serif">Stephane<br>
</font>
<br>
<br>
<br>
<table width=3D100%>
<tr valign=3Dtop>
<td width=3D40%><font size=3D1 face=3D"sans-serif"><b>kmorel at sandia.gov</b>=
</font>
<br><font size=3D1 face=3D"sans-serif">Envoy=E9 par : paraview-developers-b=
ounces at paraview.org</font>
<p><font size=3D1 face=3D"sans-serif">08/02/2010 17:32</font>
<td width=3D59%>
<table width=3D100%>
<tr valign=3Dtop>
<td>
<div align=3Dright><font size=3D1 face=3D"sans-serif">A</font></div>
<td><font size=3D1 face=3D"sans-serif">utkarsh.ayachit at kitware.com, paravie=
w-developers at paraview.org</font>
<tr valign=3Dtop>
<td>
<div align=3Dright><font size=3D1 face=3D"sans-serif">cc</font></div>
<td>
<tr valign=3Dtop>
<td>
<div align=3Dright><font size=3D1 face=3D"sans-serif">Objet</font></div>
<td><font size=3D1 face=3D"sans-serif">Re: [Paraview-developers] Requiring
Qt 4.6 for ParaView 3.8</font></table>
<br>
<table>
<tr valign=3Dtop>
<td>
<td></table>
<br></table>
<br>
<br>
<br><font size=3D2 face=3D"Calibri">How good is Qt at maintaining forward c=
ompatibility
with, for example, its ui file format. If all the developers are
on 4.6 and someone makes a change with Qt Designer to one of the ui files,
will the change break all the 4.5 builds?<br>
<br>
-Ken<br>
<br>
<br>
On 2/8/10 8:21 AM, "Utkarsh Ayachit" <</font><a href=3Dutkarsh=
.ayachit at kitware.com><font size=3D2 color=3Dblue face=3D"Calibri"><u>utkars=
h.ayachit at kitware.com</u></font></a><font size=3D2 face=3D"Calibri">>
wrote:<br>
</font>
<br><font size=3D2 face=3D"Calibri">Those are indeed good and valid points.
We were discussing a possible policy and here's what we have. What does
everybody think?<br>
<br>
* There will 1 officially supported Qt version. This is the Qt version
our binaries will be released with. The officially supported Qt version
will remain fixed for 2 revisions e.g. 3.8 and 3.10 both will officially
support the same Qt version. There will be a minimum Qt version below which
it cannot compile (for 3.8 this will be Qt 4.5). However the minimum version
may go higher in every release (but <=3D the official Qt version) depend=
ing
on whether we end up using some latest Qt functionality provided
by the officially supported release.<br>
<br>
* Let Qt 4.6 be an exception, since Qt 4.6 is required to build on Snow
Leopard and it's a nightmare to support different versions on different
platforms. So 3.8 and 3.10, will be requiring Qt 4.6 (with min. Qt 4.5)<br>
<br>
* CMake will flag an error when the Qt version is below the minimum; a
warning when the Qt version is below the official (but above the
min.) and be silent with the Qt version is same as the official or above.<b=
r>
<br>
* Any user-interface related issues for any Qt version, but the official
version may not be addressed.<br>
<br>
* We'll always have a dashboard machine compiling with the latest Qt so
that there will be no surprises when we switch to the latest Qt.<br>
<br>
* After two consecutive major releases (not patch releases) have stuck
with the same QT version and it's time to upgrade to a newer Qt version,
we'll pick the latest Qt version released at least 3 months prior to the
release.<br>
<br>
How does that sound? If it's acceptable, I can post it to the Wiki as the
official policy for Qt version updating.<br>
<br>
Utkarsh<br>
<br>
<br>
On Mon, Feb 8, 2010 at 3:32 AM, Stephane PLOIX <</font><a href=3Dstephan=
e.ploix at edf.fr><font size=3D2 color=3Dblue face=3D"Calibri"><u>stephane.plo=
ix at edf.fr</u></font></a><font size=3D2 face=3D"Calibri">>
wrote:</font>
<br><font size=3D2 face=3D"Calibri"><br>
Hi, <br>
<br>
I think that we have to be carefull : with the branding refactoring making
it easy to build applications on top of ParaView components, and the PV_INS=
TALL_DEVELOPMENT
options, ParaView should now be considered more like an SDK than a stand-al=
one
application. <br>
<br>
As more and more applications are built on top of PV, changing the version
of Qt will impact all those applications, you should then have a clear
policy regarding those versions, and give a advanced warning of any changed
to come. <br>
<br>
Some possible policy would be : <br>
at each release of a new stable version of Qt, PV will support both the
new and the old versions for 1 year, and drop the previous version after
that <br>
or <br>
each stable version of ParaView will support both the lastest stable version
of Qt and the previous one <br>
<br>
or whatever, but please make it a public policy so that we know what to
expect. <br>
<br>
Best <br>
Stephane<br>
<br>
<br>
<br>
</font><font size=3D2 color=3Dblue face=3D"Calibri"><b><u><br>
</u></b></font><a href=3Dberk.geveci at kitware.com><font size=3D2 color=3Dblu=
e face=3D"Calibri"><b><u>berk.geveci at kitware.com</u></b></font></a><font si=
ze=3D2 face=3D"Calibri">
</font><font size=3D2 face=3D"Calibri"><br>
Envoy=E9 par : </font><a href=3D"paraview-developers-bounces at paraview.org">=
<font size=3D2 color=3Dblue face=3D"Calibri"><u>paraview-developers-bounces=
@paraview.org</u></font></a><font size=3D2 face=3D"Calibri">
</font><font size=3D2 face=3D"Calibri">07/02/2010 21:48</font><font size=3D=
2 face=3D"Calibri">
</font>
<div align=3Dright>
<p><font size=3D2 face=3D"Calibri">A </font></div>
<p><a href=3Dburlen.loring at gmail.com><font size=3D2 color=3Dblue face=3D"Ca=
libri"><u>burlen.loring at gmail.com</u></font></a><font size=3D2 face=3D"Cali=
bri">
</font>
<div align=3Dright>
<p><font size=3D2 face=3D"Calibri">cc </font></div>
<p><a href=3D"paraview-developers at paraview.org"><font size=3D2 color=3Dblue=
face=3D"Calibri"><u>paraview-developers at paraview.org</u></font></a><font s=
ize=3D2 face=3D"Calibri">
</font>
<div align=3Dright>
<p><font size=3D2 face=3D"Calibri">Objet </font></div>
<p><font size=3D2 face=3D"Calibri">Re: [Paraview-developers] Requiring Qt 4=
.6
for ParaView 3.8</font><font size=3D2 face=3D"Calibri"> <br>
<br>
<br>
<br>
</font><font size=3D4 face=3D"Calibri"><br>
I wouldn't object to supporting 4.5 and 4.6. I think setting up 1-2 dashboa=
rds
to verify building with 4.5 would be enough. We don't run into Qt specific
bugs too often anyway. I defer to Utkarsh though.<br>
<br>
Also, if there is an attractive feature on the latest version of Qt that
will make our lives much easier, we are likely to move to it and not support
the previous version. So, I don't want to have an official rule about suppo=
rting
more than 1 version.<br>
<br>
-berk<br>
</font><font size=3D2 face=3D"Calibri"><br>
</font><font size=3D4 face=3D"Calibri"><br>
On Sun, Feb 7, 2010 at 3:08 PM, burlen <</font><a href=3Dburlen.loring at g=
mail.com><font size=3D4 color=3Dblue face=3D"Calibri"><u>burlen.loring at gmai=
l.com</u></font></a><font size=3D2 face=3D"Calibri">
<</font><a href=3Dmailto:burlen.loring at gmail.com><font size=3D2 color=3D=
blue face=3D"Calibri"><u>mailto:burlen.loring at gmail.com</u></font></a><font=
size=3D2 face=3D"Calibri">>
</font><font size=3D4 face=3D"Calibri">> wrote:</font><font size=3D2 fac=
e=3D"Calibri">
</font><font size=3D4 face=3D"Calibri"><br>
Will this mean that PV won't build with any thing less than Qt 4.6?<br>
I wonder if it would make sense to have a small range of supported versions
rather than a single one?<br>
<br>
It's currently the case that with qt anything less than 4.5 pv fails to
compile (even if you remove the version test in cmakelists) because pv
uses some qt classes that were only added to 4.5. In my opinion it would
be a burden to pv users to get too far ahead of the various KDE supported
distro's packages. eg. Kubuntu 9.10 reports qt version 4.5.2. KUbuntu 8.04.4
LTS reports qt version 4.3.4.<br>
<br>
More information about the Paraview-developers
mailing list