From kde-core-devel Wed Jan 30 09:18:37 2008 From: Lubos Lunak Date: Wed, 30 Jan 2008 09:18:37 +0000 To: kde-core-devel Subject: Re: qt-copy patch 0180 Message-Id: <200801301018.37647.l.lunak () suse ! cz> X-MARC-Message: https://marc.info/?l=kde-core-devel&m=120168476826132 On st 30. ledna 2008, Richard Moore wrote: > On 1/29/08, Lubos Lunak wrote: > > right) any use of scripting in Plasma will make it abort when any of > > x,y,z in > > I don't follow you - please explain how this affects scripting so I > can code around it. Are you simply saying that this bug in QUiLoader > introduces the problem? QUiLoader uses QObjectPrivate and it is in a static library, so it gets pulled into libplasma binary. Qt changes, QObjectPrivate changes, it doesn't match what's in libplasma -> abort. I'm not familiar with that part of Plasma at all, but since QUiLoader is for loading UI from .ui files, I assume it's for scripted plasmoids. > If so I guess the fix is to make our own > version of that class. I take back that part about it being hard to fix. I was already too tired in the evening and after trying to find a hack for the #0180 patch without changing QObjectPrivate I failed to see the simple solution when it turned out that QObjectPrivate is still meant to be private. I'll make a qt-copy patch for removing the QUiLoader dependency on QObjectPrivate. That also means that there is no problem with patch #0180 and has never been. If it causes problems again somewhen, it'll mean something else in Qt is broken. One obviously cannot mix different parts of Qt from different builds, but that's a bad idea anyway. -- Lubos Lunak KDE developer -------------------------------------------------------------- SUSE LINUX, s.r.o. e-mail: l.lunak@suse.cz , l.lunak@kde.org Lihovarska 1060/12 tel: +420 284 028 972 190 00 Prague 9 fax: +420 284 028 951 Czech Republic http//www.suse.cz