What would be the gain of such a thing? It would be extremely slow to run clang-format every single time you ran CMake, and also having a run of CMake generate changes to your source tree seems like using the wrong tool for the job. I think we should follow whatever LLVM does, which in this case is that it is each developers' responsibility to run clang-format before submitting.
There may be a case to be made for presubmit hooks that prevent you from checking in code which doesn't pass clang-format, but that discussion should happen on llvm-dev@. On Tue, Sep 6, 2016 at 4:04 PM Eugene Zelenko via lldb-dev < lldb-dev@lists.llvm.org> wrote: > Hi! > > Since LLDB code is freshly formatted, it may make sense to run > Clang-format automatically via CMake. If not as part of regular build, > but may be subtarget of install or check targets as it done in Polly. > > Eugene. > _______________________________________________ > lldb-dev mailing list > lldb-dev@lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev >
_______________________________________________ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
