On Mon, Oct 14, 2024 at 04:06:44PM +0200, Patrick Steinhardt wrote: > The compilation of clar-based unit tests is broken because we do not > add the binary directory into which we generate the "clar-decls.h" and > "clar.suite" files as include directories. Instead, we accidentally set > up the source directory as include directory. I am confused. What is the difference between CMAKE_SOURCE_DIR and CMAKE_BINARY_DIR here, and why does the difference between the two matter? > diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt > index 62af7b33d2f..093852ad9d6 100644 > --- a/contrib/buildsystems/CMakeLists.txt > +++ b/contrib/buildsystems/CMakeLists.txt > @@ -1042,7 +1042,7 @@ file(WRITE "${CMAKE_BINARY_DIR}/t/unit-tests/clar.suite" "${clar_decls}" "${clar > list(TRANSFORM clar_test_SUITES PREPEND "${CMAKE_SOURCE_DIR}/t/unit-tests/") > list(TRANSFORM clar_test_SUITES APPEND ".c") > add_library(unit-tests-lib ${clar_test_SUITES} "${CMAKE_SOURCE_DIR}/t/unit-tests/clar/clar.c") > -target_include_directories(unit-tests-lib PRIVATE "${CMAKE_SOURCE_DIR}/t/unit-tests") > +target_include_directories(unit-tests-lib PUBLIC "${CMAKE_BINARY_DIR}/t/unit-tests") This also changes the 'scope' parameter of 'target_include_directories' from PRIVATE to PUBLIC, but the commit message doesn't mention such a change. Is it intentional? If so, can the commit message be updated to explain why this is done? If not, is this a stray change that snuck in? (If all of this is obvious to you, I apologize for the confusion on my end. I'm not at all familiar with our CMake bits, so the extra explanation would help me quite a bit in making sense of this.) Thanks, Taylor