Steve,
Can you explain the need for 953def1e17f3bbba0aa42037ae15ced011d8fd2a:
Fix DAG checker finding cycling dependencies.
Before this patch, the following is reported falsely as a self-reference:
target_link_libraries(empty2 LINK_PUBLIC empty3)
target_link_libraries(empty3 LINK_PUBLIC empty2)
add_custom_target(...
-DINCLUDES=$<TARGET_PROPERTY:empty2,INTERFACE_INCLUDE_DIRECTORIES>
)
The reason is that the existing code assumed that all reading of
include directories would be done through cmTarget::GetIncludeDirectories()
and would therefore be initialized with a DagChecker. That is not the case
if reading the property with an 'external' generator expression.
As an unfortunate side-effect of backward compatibility for the
LINK_INTERFACE_LIBRARIES property, self references can no longer be errors.
+ // It would be better to consider it an error for the foo target
+ // to have a INTERFACE_INCLUDE_DIRECTORIES which depends directly on its
+ // own INTERFACE_INCLUDE_DIRECTORIES property, but as the error of a
+ // target having itself in its own LINK_INTERFACE_LIBRARIES is 'allowed'
+ // and tested, and as the interface includes and defines are now based
+ // on the link interface, it breaks the CMakeOnly.LinkInterfaceLoop test.
? The LinkInterfaceLoop test was added here:
http://cmake.org/gitweb?p=cmake.git;a=commitdiff;h=8e756d2b
to cover a case that was accidentally allowed but that should not be used.
I'd rather not drop this great safety check for backward-compatibility
with bad code.
Can't you instead simply have the usage requirements skip appending
a target's own requirements to itself? That way the resulting
generator expression would still not have a self reference.
-Brad
--
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please keep messages on-topic and check the CMake FAQ at:
http://www.cmake.org/Wiki/CMake_FAQ
Follow this link to subscribe/unsubscribe:
http://public.kitware.com/cgi-bin/mailman/listinfo/cmake-developers
