The log showcasing the issue is obtained by emerging sci-libs/itk::science , and it might be an issue with that package, though I'm fairly sure it isn't. The relevant snippet is: ``` CMake Error in Modules/IO/HDF5/src/CMakeLists.txt: Imported target "hdf5::hdf5_cpp" includes non-existent path "/var/tmp/portage/sci-libs/hdf5-1.14.0/work/hdf5-1.14.0/src/H5FDsubfiling" ``` Note that for some reason sci-libs/itk is looking for ITK in its WORKDIR. This is nowhere to be found in the ebuild and it goes away if =sci-libs/hdf5-1.12.2-r6 is installed instead. Which, obviously, does not provide the WORKDIR file either. Best I could come up with is that maybe hdf5 reports its path... somehow 🤔 and the new one reports the WORKDIR path instead of the actual path under which it's installed... Reproducible: Always
Created attachment 860419 [details] build log of program trying to link hdf5
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=995c8aefa871da5ad197b29b782ff9e2d1e46e83 commit 995c8aefa871da5ad197b29b782ff9e2d1e46e83 Author: Sam James <sam@gentoo.org> AuthorDate: 2023-04-20 19:34:45 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-04-20 19:36:51 +0000 sci-libs/hdf5: fix installed CMake files (exclude temporary include dir for build) Closes: https://bugs.gentoo.org/904515 Closes: https://bugs.gentoo.org/904691 Signed-off-by: Sam James <sam@gentoo.org> .../hdf5/files/hdf5-1.14.0-fix-include-path.patch | 34 ++++++++++++++++++++++ .../{hdf5-1.14.0.ebuild => hdf5-1.14.0-r1.ebuild} | 1 + 2 files changed, 35 insertions(+)
