Challenge for real experts: why does cmake 3.1.3 not use automoc for a certain file?

[Friedrich W. H. Kossebau]

Am Dienstag, 23. Juni 2015, 21:31:31 schrieb Alex Merry:
> On Wednesday 17 June 2015 23:11:33 Friedrich W. H. Kossebau wrote:
> > Am Dienstag, 16. Juni 2015, 19:16:11 schrieb Friedrich W. H. Kossebau:
> > > Interesting... seems renaming the target "KisAsyncMergerTest" to
> > > something
> > > else, e.g. "KisAsyncMergerTest_" unbreaks this.
> > > Any explanation? There is no other target with that name
> > > "KisAsyncMergerTest".
> > > 
> > > Having done that workaround, now another target is hit with the same
> > > problem, KisBookmarkedConfigurationManagerTest. Got the same treatment
> > > with
> > > _ appended, proper automoc targets now created for me locally.
> > > 
> > > Currently waiting if CI finds the next problem now (not enough disk
> > > space
> > > left right now locally for giving a full test myself with the other
> > > cmake
> > > version, Calligra/Krita is too large :), so partially relying on
> > > build.kde.org).
> > > 
> > > Could there be some hash collision or something else? There are a lot of
> > > targets in that CMakeLists.txt, perhaps more than someone expected...
> > > http://quickgit.kde.org/?p=calligra.git&a=blob&h=082747066eb6910d29a7fd3
> > > a5
> > > 9b 6b531af529e95&f=krita%2Fimage%2Ftests%2FCMakeLists.txt&o=plain
> > 
> > So I renamed three test build targets in a row, and everytime another test
> > target started failing, with automoc no longer being run. With current
> > "frameworks" branch build now fails on KisFilterMaskTest, due to no
> > automoc
> > run.
> > 
> > Just tried cmake-3.3.0-rc2, and things are okay there again, builds fine
> > with automoc run on all files where needed.
> > 
> > So... what do you recommend to do? Blacklist some cmake versions and not
> > build the tests there? Any other workarounds possible?
> > 
> > Do I better contact the cmake guys directly, or is there still some KDE
> > contact man/woman on this list?
> 
> Blacklisting might be your best bet. If it works with 3.3.0, the CMake folks
> are, honestly, unlikely to be willing to do anything - patch releases are
> generally not made after a new feature release is out.
> 
> As for other workarounds, that would be a case of either trying things out
> blindly or trying to track down exactly why this is happening. I thought I
> had tried and failed to reproduce the problem with CMake 3.1.3, but I can't
> find any evidence of that, so I may be misremembering. If I can, I'll spend
> a little time trying to track down the issue, but I can't promise that I'll
> get anywhere with that.

For the record, finally reported as http://www.cmake.org/Bug/view.php?id=15647

CMake versions affected so far seem at least: 3.1.3 - 3.2.3

Cheers
Friedrich