KDE 4.4 Qt version

John Tapsell johnflux at gmail.com
Wed Jan 13 10:02:02 GMT 2010

2010/1/13 David Faure <faure at kde.org>:
> On Tuesday 12 January 2010, Tom Albers wrote:
>> Op Tuesday 12 January 2010 19:20 schreef u:
>> > > So yeah, I think requiring Qt 4.6.1 would be a good idea, imho.
>> >
>> > I agree, if we know that version x.y.z of package Foo makes our software
>> > crash, we should require a version != x.y.z of Foo.
>> > Even if that means to delay the release.
>> > What's a release worth if it crashes all the time...
>> Yes, I agree too. Since it is runtime, can't we detect at runtime and print
>> a nice error message and refuse to start kde and the applications?
>> Distro's which do basic testing will catch and fix it for their users,
>> users which build themselves are warned that way too...
> If you just detect the qt version and decide to error out based on that,
> then it will break for the distros or people who have backported the fix
> to that version. The only correct way would be a runtime-check for the actual
> bug, which is almost impossible (e.g. let's start a simple kde app and ask it
> to quit via dbus and let's see if it crashes.... ;)

If this really is getting into 4.6.1 then I think we don't have to
worry too much about this bug other than putting a note about it in
whatever file distros read.  Do we have file for notes to packagers?


More information about the kde-core-devel mailing list