Hi Steven,
On Thu, 16 Sep 2021 08:15:31 +0200 Timo Röhling wrote:
One alternative solution would be to patch the upstream CMake files
and export GMock independently with their own GMockConfig.cmake and
GMockTargets.cmake. Then, for backwards compatibility, you could
add something like
fi
* Steven Robbins [2021-09-16 07:58]:
So it looks like gtest 1.11.0 may well have fixed it. There's a passing test
now: https://ci.debian.net/data/autopkgtest/testing/amd64/f/fastcdr/15289175/
log.gz
Unfortunately, that is a false negative because autopkgtest uses
testing versions for dependenc
Forgot the URL. Sorry for all the noise. I promise to stop sending
further emails now ;)
https://github.com/google/googletest/pull/3376
--
⢀⣴⠾⠻⢶⣦⠀ ╭╮
⣾⠁⢠⠒⠀⣿⡁ │ Timo Röhling │
⢿⡄⠘⠷⠚⠋⠀ │ 9B03 EBB9 8300 D
On Thursday, September 16, 2021 1:07:19 A.M. CDT you wrote:
> Sorry for the confussion, that also refers to fastcdr, which has a
> failing autopkgtest here:
> https://ci.debian.net/data/autopkgtest/testing/amd64/f/fastcdr/15272003/log.
> gz
So it looks like gtest 1.11.0 may well have fixed it. T
On Wed, 15 Sep 2021 18:17:33 -0500 Steven Robbins wrote:
Intriguing. Note that gtest is intended as the base layer (gmock is built on
top of gtest), so it would be a loss to have libgtest-dev pull in gmock.
Hopefully there's an alternative fix.
I have submitted a rework of the CMake files to
On Thursday, September 16, 2021 1:07:19 A.M. CDT you wrote:
> * Steven Robbins [2021-09-15 20:52]:
> >I thought this tag was for the case that the package fails to build from
> >source. ?? https://www.debian.org/Bugs/Developer
>
> I thought this is also transitively for FTBFS caused in other
> p
On Thu, Sep 16, 2021 at 08:07:19AM +0200, Timo Röhling wrote:
> * Steven Robbins [2021-09-15 20:52]:
> > I thought this tag was for the case that the package fails to build from
> > source. ?? https://www.debian.org/Bugs/Developer
> I thought this is also transitively for FTBFS caused in other
>
* Steven Robbins [2021-09-15 20:05]:
Do you have a minimal reproduction by any chance? I'd like to test out 1.11
-- just in case it fixes the problem.
See attached CMakeLists.txt
BTW, I isolated the CMake upstream change that exposes the bug:
https://gitlab.kitware.com/cmake/cmake/-/commit/23
* Steven Robbins [2021-09-15 18:17]:
In case you're wondering, apparently this bug has been exposed by
the changes in the new CMake version that has hit unstable, so this
will possibly affect more packages.
Intriguing. Note that gtest is intended as the base layer (gmock is built on
top of gte
* Steven Robbins [2021-09-15 20:52]:
I thought this tag was for the case that the package fails to build from
source. ?? https://www.debian.org/Bugs/Developer
I thought this is also transitively for FTBFS caused in other
packages, but I might be wrong on this. Feel free to adjust the tags
if y
On Wednesday, September 15, 2021 2:48:01 P.M. CDT Timo Röhling wrote:
> Control: tags 994419 + ftbfs
I thought this tag was for the case that the package fails to build from
source. ?? https://www.debian.org/Bugs/Developer
> I just noticed that this bug not only breaks autopkgtest
What auto
On Wednesday, September 15, 2021 2:25:49 P.M. CDT Timo Röhling wrote:
> Package: libgtest-dev
> Version: 1.10.0.20201025-1.1
I've discovered that upstream has released 1.11 -- which I'll package up.
> the GTestTargets.cmake that is shipped in libgtest-dev also exports
> the GTest::gmock and GTe
On Wednesday, September 15, 2021 2:48:01 P.M. CDT Timo Röhling wrote:
> > Please add Depends: libgmock-dev to libgtest-dev.
>
> I just noticed that this bug not only breaks autopkgtest but also
> causes fastcdr to FTBFS.
>
> In case you're wondering, apparently this bug has been exposed by
> the
13 matches
Mail list logo
