2018-02-06 13:15 GMT+03:00 Geogin Varghese :
> What is the recommended way of removing transitively inherited
> attributes for a target.
>
> Example:
> target_link_libraries(A PRIVATE B)
> target_compile_options(B PUBLIC -Werror)
>
Shouldn't it be PRIVATE for B?
E.g.
For stuff like this one you could add -Wno-error afterwards to A's
compile_options.
---
Jean-Michaël Celerier
http://www.jcelerier.name
On Tue, Feb 6, 2018 at 11:15 AM, Geogin Varghese wrote:
> What is the recommended way of removing transitively inherited
> attributes
What is the recommended way of removing transitively inherited
attributes for a target.
Example:
target_link_libraries(A PRIVATE B)
target_compile_options(B PUBLIC -Werror)
How should the -Werror option be removed from target A.
--
Powered by www.kitware.com
Please keep messages on-topic and
Kitware is offering a free course on VTK/ParaView/CMake March 13-15,
2018 at Kitware's Clifton Park, NY, USA. The signup can be found here:
https://docs.google.com/forms/d/e/1FAIpQLSeee7oTBzapITmark0dIX0E86zEyyPmh-LEbecz2zm_822WFw/viewform
Hope to meet some of you in person.
-Bill
--
Yes, it should be configurable via one option flag.
The cmake user can control it but it is not well documented and different to
configure depending on compiler and generator used:
#for ninja generator:
set(CMAKE_DEPFILE_FLAGS_CXX "-MMD -MT -MF " CACHE STRING
"dependency flag" FORCE)
# for
On 02/06/2018 03:19 AM, Claus Klein wrote:
> IMHO: This make ninja slower without much user benefit!
For reference, the change was made for CMake 3.6 here:
https://gitlab.kitware.com/cmake/cmake/commit/6d74e7870b8804a5af0bc395a9fbb45c1b3d26a4
The benefit is that sources recompile when system
Kitware is offering a free course on VTK/ParaView/CMake March 13-15,
2018 at Kitware's Clifton Park, NY, USA. The signup can be found here:
https://docs.google.com/forms/d/e/1FAIpQLSeee7oTBzapITmark0dIX0E86zEyyPmh-LEbecz2zm_822WFw/viewform
Hope to meet some of you in person.
-Bill
--
On 02/06/2018 10:03 AM, Claus Klein wrote:
> why different default behaviors in Nina and Makefile generators?
The Makefile dependency generation works totally differently and
can't do system headers well at all. One day perhaps they should
be ported to use a depfile approach too and then a
VERSION_MINOR 11)
-set(CMake_VERSION_PATCH 20180206)
+set(CMake_VERSION_PATCH 20180207)
#set(CMake_VERSION_RC 1)
---
Summary of changes:
Source/CMakeVersion.cmake |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
hooks/
I’m wondering about the different dependency strategies between ‚Unix Makefiles’ and Nina generator.IMHO: This make ninja slower without much user benefit!-Clausfind . -name depend.make | xargs cat# CMAKE generated file: DO NOT EDIT!# Generated by "Unix Makefiles" Generator, CMake Version
10 matches
Mail list logo