[checkmk-commits] [tribe29/checkmk] 732283: Bumped googletest to latest version. Create alread...

Branch: refs/heads/master Home: https://github.com/tribe29/checkmk Commit: 7322836ddeddaa328b8ac58635c659a2094aab4e https://github.com/tribe29/checkmk/commit/7322836ddeddaa328b8ac58635c659a20… Author: Sven Panne <sven.panne(a)tribe29.com> Date: 2023-01-03 (Tue, 03 Jan 2023) Changed paths: M agents/wnx/scripts/unpack_all.cmd M agents/wnx/scripts/unpack_package.cmd M livestatus/src/.gitignore M livestatus/src/Makefile.am M omd/Licenses.csv R omd/packages/googletest/README A omd/packages/googletest/create-archive A omd/packages/googletest/googletest-71140c3ca7-patched.tar.gz R omd/packages/googletest/googletest-9c332145b7.tar.gz M omd/packages/googletest/googletest.diff M packages/livestatus/test/CMakeLists.txt Log Message: ----------- Bumped googletest to latest version. Create already patched tar file. Patching sources in CMake seems to be a mess, see e.g. the discussion at https://gitlab.kitware.com/cmake/cmake/-/issues/17287. The underlying problem is: If the unpatched sources and/or the set of patches change, things should be re-patched from scratch, but only then. This common problems seems to be ignore by CMake, which is a pity. Bazel reportedly has similar problems, although this sounds strange: The patched sources are a function of the sources + patches (see above), so Bazel should handle this just fine. Anyway: Battling the build system seems to be worse than having this intermediate step. If we ever learn how to do this in a cleaner way *within* the build system, we can easily change our minds again later. Change-Id: I86c4fe6574c0fde77d177aac71f5da76de8d5f67