Closed MarcelKoch closed 5 months ago
I've added now a GH action to check for clang-tidy warnings.
Using it as part of our git-hooks is not a good idea IMO. clang-tidy requires the path to the compilation database, but determining that when running the git-hooks seems generally not possible. So far, I've only come up with some hacks that create a file which stores the build directory of the last cmake call. But I don't think that is a reasonable solution. The GH action should suffice.
In general I prefer snake_case for variables and functions because it makes it easier to see if something is a type or a variable. @HenningScheufler what is your opinion?
Ok lets go with camelCase everywhere.
Does this need to wait on anything? I would suggest applying the fixes in another PR.
@greole can we merge?
This PR adds a clang-tidy check to enforce a consistent naming scheme. Currently the scheme is defined as
Other ideas are up to discussion. The clang-tidy check can be adjusted quite a lot, for example different case for const varibles, global variables, abstract classes, etc. The full list can be found here: https://clang.llvm.org/extra/clang-tidy/checks/readability/identifier-naming.html#cmdoption-arg-AbstractClassCase
Right now this check is not enforced. It needs either IDE support (e.g. CLion has this out-of-the-box), or manual invocation. To run it manually do
It requires to have the
compile_commands.json
available, which can be generated by CMake, by configuring it with-DCMAKE_EXPORT_COMPILE_COMMANDS=ON
. IMO we should just always set this option, there is no downside to it.I will also try to make it part of our git hooks.