Re: Proposal: Archive all unported Qt5 projects

Am 04.06.26 um 00:57 schrieb Jaroslaw Staniek:
> On Thu, Jun 4, 2026 at 12:04 AM Luigi Toscano 
> <[email protected]> wrote:
>
>     Nicolas Fella ha scritto:
>     > Hi,
>     >
>     > Qt6/KF6 has been around long enough that most KDE projects have
>     been ported to
>     > it. Qt5 is increasing degrees of dead, and any project not
>     ported to Qt6 is
>     > likely equally dead.
>     >
>     > With that in mind I propose that we declare unmaintained and
>     archive all KDE
>     > projects that still don't build with Qt6 (unless there's signs
>     of active
>     > development or they are otherwise relevant).
>     >
>     > In particular this would apply to the following projects:
>     >
>     > [...]
>
>     > No release in a long time:
>     > [...]
>     > kdb
>     > kproperty
>     > kreport
>     > kexi
>
>     Does anyone know what happened with kexi and its libraries? Quite
>     recently
>     (ok, it was more than one year ago) a new website was created
>     (with some
>     questionable images, but that's another story) which seemed to
>     imply that some
>     update was close, but then nothing else.
>     (cc JJaroslaw Staniek).
>
>
> Thank you for the CC.
>
> Indeed, the update was quite close to release, but it did not bring 
> any significant new features. Then, in early 2023, the AI revolution 
> arrived and fundamentally changed the landscape, including the area in 
> which applications like KEXI operate. Portability between backends was 
> not as crucial question as integration with external tools (scripting, 
> and soon, AI models).
>
> Since our last releases in 2019, there have been numerous 
> discussions-many of them documented-about the future direction of the 
> project. One option being considered is to simplify and narrow the 
> scope towards self-contained database applications based exclusively 
> on SQLite.
>
> There are many technical and product-level considerations involved, 
> and any such transition would require careful evaluation to minimize 
> disruption and loss of functionality. At least we are not dependent on 
> QtSql, which has seen relatively few changes in Qt 6 compared to Qt 5, 
> aside from improvements related to QML integration.
>
> Porting to Qt 6 and a release in 2026 is of interest to me.

A good first step would be making sure that the current CI builds for 
Kexi and dependencies work, they have been failing for quite some time now.