a5d45e685f Tests: Add case for ENVIRONMENT_MODIFICATION property OP=reset behavior
e2854b4fa2 cmCTestRunTest: Implement the ENVIRONMENT test property with EnvDiff too
bfa1c5285b cmSystemTools: Add EnvDiff class to hold ENVIRONMENT_MODIFICATION logic
a0b1c4ee90 cmCTestRunTest: Simplify by using GetSystemPathlistSeparator
4e6cbb1f13 cmCTestRunTest: Remove unnecessary CMAKE_BOOTSTRAP guard
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !7572
When processing the reset operation in the context of a CTest
`ENVIRONMENT_MODIFICATION` property, the value the variable is reset to
is the one it had after `ENVIRONMENT` was processed, not before.
This was broken once during refactoring and is subtle enough that it
should be tested.
Refactoring in commit 8c65b7042e (cmExportFileGenerator: Simplify
collection of targets missing from export set, 2022-04-11,
v3.24.0-rc1~281^2) accidentally dropped the behavior change from
commit 0ad2a1c181 (Export: Never treat private link libraries as
public package dependencies., 2013-09-24, v3.0.0-rc1~559^2).
Restore the behavior and add a test.
Fixes: #23838
81fd0d6e26 ExternalProject: Fix install on BUILD_ALWAYS+BUILD_BYPRODUCTS with Ninja
d9e88721ad ExternalProject: Add test case for BUILD_ALWAYS
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !7546
In commit 695f0d0d3a (cmFortranParser: Parse keywords as lexical tokens,
2016-09-05, v3.7.0-rc1~150^2) we created keyword-specific variants of
the original `USE WORD other EOSTMT` production, such as
`MODULE WORD other EOSTMT` and `INTERFACE WORD other EOSTMT`. The same
pattern was used by more keyword-specific productions in commit b5ac8b8aa7
(Fortran: Add support for submodule syntax in dependency scanning,
2016-09-05, v3.7.0-rc1~73^2~1).
The postfix part (`other`) of several keyword-specific productions is
not needed to match Fortran syntax. See the Fortran 2018 standard,
para.4.1.4/1 on p.28, para.14.2.1/2 on pp.293-294. The postfix is
needed only for a case of operator 'use':
use <module-name> [, only : <list-of-vars>]
The unnecessary postfix matching from the keyword-specific productions
such as module, submodule, and interface declarations can cause spurious
module dependencies to be detected, so remove it.
Extend the test suite with examples covering the previously-broken
cases.
Fixes: #18427
The `BUILD_BYPRODUCTS` option causes the ExternalProject's `build` step
in `build.ninja` to have `restat = 1`, so its "always out-of-date"
status caused by `BUILD_ALWAYS` does not propagate to the `install`
step. Mark the latter step as explicitly always out-of-date too.
Fixes: #23820
The `PREFIX/(Foo|foo|FOO).*/(cmake|CMake)/(Foo|foo|FOO).*/` search
path is the similar to the one already exists
`PREFIX/(Foo|foo|FOO).*/(lib/ARCH|lib*|share)/cmake/(Foo|foo|FOO).*/`.
c633bd76d8 Add linker prefix for "-stack" flag on MSVC builds
6823db0c5a Enable IPO for IntelLLVM compilers on Linux and Windows
adbffd204d Propagate OpenMP compiler flags to linker for IntelLLVM
3640842df2 Tests: Update ModuleDefinition for IntelLLVM MSVC Frontend Variant
5d5a712303 IntelLLVM: Pass linker flags to the compiler when used as linker
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: buildbot <buildbot@kitware.com>
Merge-request: !7533
The CMAKE_VERIFY_INTERFACE_HEADER_SETS variable is intended to
be under the control of the user. It doesn't discriminate between
header sets defined in the main project and those defined by
dependencies brought into the build directly via FetchContent.
Developers will usually only be interested in verifying the main project's
header sets, not those from dependencies.
Make the variable effectively only enable header set verification of the
main project by turning it off during FetchContent_MakeAvailable() calls.
The user still has variables like CMAKE_PROJECT_INCLUDE and
CMAKE_PROJECT_<projectName>_INCLUDE available to them if they
want to enable verification of all or specific dependencies respectively.
Fixes: #23808
Directory-level rules in `CMakeFiles/Makefile2` were previously
previously written by each directory's local generator using its own
decision for using relative or absolute paths.
Since commit d33b12d84b (Add support for build tree symlink inside
source tree, 2022-02-25, v3.24.0-rc1~583^2), each local generator
explicitly models the relationship between its source and build paths,
and uses this to determine when it is safe to use relative paths.
Because `add_subdirectory` supports arbitrary placement of the source
and build directories, different local generators may have different
relationships between their source and build paths. This can cause
disagreement among rules written to `CMakeFiles/Makefile2`.
Restore consistency by always using the root local generator to write
rules to `CMakeFiles/Makefile2`. Relative paths should always be
expressed w.r.t. the top-level build directory since that is the working
directory in which the `make` tool processing the file will run.
Fixes: #23814
4f95e6b284 ci: test BMI exporting and installation with GCC in CI
c49d5f137b RunCMake/CXXModules: add a "deep-chain" test
297e0f4dce cmCxxModuleMapper: support MSVC module map format
b3c2880cb2 cmCxxModuleMapper: track transitive modules for MSVC
a43713d615 CTestCustom: ignore `cm::optional` uninitialized memory false positive
b90de0b492 RunCMake/CXXModules: support MSVC extensions
a84c186a7d cmScanDepFormat: support the MSVC 17.3 toolchain temporarily
d7f5064ff7 cmScanDepFormat: support P1689R5
...
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !7481
IntelLLVM for MSVC uses the compiler for linking. The /DEF flag needs
to be passed to the linker, and the compiler does not accept it, unless
it is prefixed with `/Qoption,link,`.
Rather than explicitly set the LINK_FLAGS target property, as is done
for other MSVC frontend variants, add the .def file to the sources
for `add_library`. `add_library` will wrap the `/DEF` flag using
`CMAKE_${lang}_LINKER_WRAPPER_FLAG`. Presumably, this should work
for any MSVC front end, but leave the original test to avoid changing
a test that is working.
Signed-off-by: William R. Dieter <william.r.dieter@intel.com>
Some clients have been explicitly checking whether the string specified
after a keyword is empty. Offer them a way to specify that the string
must be non-empty as part of the binding type.
If a target doesn't have any source files, fall back to the global
list of enabled languages to determine the language of the header
file to verify.
Fixes: #23774
