Our custom command generation logic assumes that all command lines have
at least `argv0`. In `add_custom_{command,target}` we already check
that at least a `COMMAND` was given, but using `COMMAND_EXPAND_LISTS` in
combination with a generator expression that expands to an empty string
may produce an empty command line. In this case simply add an empty
string as a command to maintain our internal invariant.
Fixes: #17993
In commit v3.11.0-rc1~433^2~1 (Add support for IMPORTED GLOBAL targets
to be aliased, 2017-09-14) we accidentally dropped the error on calling
`add_library` to alias an imported target that is not globally visible.
The `add_executable` command's equivalent check was properly updated.
Restore the check in `add_library` with the same update. Also fix the
test case accordingly.
Fixes: #17982
Extend the fix in commit v3.9.6~1^2 (Restore exclusion of "gcc_eh" from
implicit link libraries, 2017-11-07) to also exclude `gcc_eh` libraries
referenced by absolute path to the library file.
Issue: #17436
Since commit v3.11.0-rc1~467^2 (VS,Xcode: Add CMakeLists.txt sources
without mutating targets, 2017-10-18) we do not add `CMakeLists.txt` to
target sources but instead generate references to them directly. This
broke projects that explicitly specify their `CMakeLists.txt` file as a
source file because the explicit entry is no longer consolidated with
the generated one.
Teach the relevant generators to avoid duplicating `CMakeLists.txt`
source references and add test cases.
Fixes: #17828
In commit v3.11.0-rc1~68^2 (CPack: accept --trace and --trace-expand,
2017-12-09) a nullptr dereference was added that occurs when
`cpack -G NotAGenerator` is invoked. Add the needed condition.
Fixes: #17900
Some code paths in `ExpandVariablesInString{New,Old}` were not checking
the `filename` parameter for a null pointer, but this can happen when
using the above flags together. Add the checks and a test case.
Fixes: #17896
Revert commit v3.8.0-rc1~305^2 (Remove CTestTestfile.cmake when BUILD_TESTING
is OFF, 2016-11-14) again. We reverted it once in commit v3.8.0-rc3~22^2
(Revert "Remove CTestTestfile.cmake when BUILD_TESTING is OFF", 2017-03-06) but
it was accidentally restored by commit v3.11.0-rc1~387^2 (server: add
"ctestInfo" request to get test info, 2017-10-25), perhaps due to conflict
resolution during rebase.
We cannot remove `CTestTestfile.cmake` when testing is off because it breaks
projects that never enable testing but create their own `CTestTestfile.cmake`
manually instead. Revert the change again and add a test case.
Features recorded by commit v3.6.0-rc1~120^2~5 (Features: Record
standards and features for Intel C++ on UNIX, 2016-04-28) for the Intel
compiler left out initializer list support because our test case in
`Tests/CompileFeatures/cxx_generalized_initializers.cpp` caused an
internal compiler error. It turns out this is because the Intel
compiler asserts the `initializer_list` constructor signatures to verify
that they match its own `<initializer_list>` header. It was our dummy
implementation used to test the language feature without any headers
that caused the ICE. Revise it to use a constructor signature accepted
by the Intel compiler.
Fixes: #17829
In gtest_discover_tests(), the TIMEOUT keyword was making it
impossible to set the TIMEOUT test property via the PROPERTIES
keyword. This would be a frequent case, but it doesn't complain
and instead silently does something different to what would
normally be expected. The TIMEOUT keyword has been renamed
to DISCOVERY_TIMEOUT, thereby removing the clash.
This is a breaking change. 3.10.1 and 3.10.2 were the only versions
that supported the TIMEOUT keyword and uses of it were likely
not working as intended.
Fixes: #17801
When evaluating include directories during export to a `try_compile`
test project, thread the compile language through to the generator
expression evaluator so it can support `$<COMPILE_LANGUAGE:...>`.
Issue: #17811
When evaluating `INTERFACE_SYSTEM_INCLUDE_DIRECTORIES`, or evaluating
`INTERFACE_INCLUDE_DIRECTORIES` on an imported target, thread the
compile language through to the generator expression evaluator so
that it can support `$<COMPILE_LANGUAGE:...>`.
Fixes: #17811
The test was expecting timestamps on day 1 in 1970 to start at 86400 but
they actually started at 0. This worked without `SOURCE_DATE_EPOCH=1`
because after 1972, leap days compensated the offset.
Fixes: #17762
In case CMAKE_XCODE_GENERATE_TOP_LEVEL_PROJECT_ONLY has been enabled
generate only the root-level ZERO_CHECK target so targets in
subdirectories pick up the root generator target of ZERO_CHECK.
For the case that CMAKE_XCODE_GENERATE_TOP_LEVEL_PROJECT_ONLY is not
enabled more investigation and a proper and final fix is still needed.
Issue: 14297
`Modules/CMakeCCompilerId.c.in` will report the C dialect as 11 whenever
`__STDC_VERSION__` indicates *at least* C 11. Make the test consistent
with this. We already do this for the C++ case.
Fixes: #17740
The `LINK_ONLY` generator expression is used to represent private
dependencies of static libraries in their `INTERFACE_LINK_LIBRARIES`
property value. Fix evaluation of generator expressions during export
to support the `LINK_ONLY` genex. Extend the RunCMake.AndroidMK test
with a case for this.
This is to avoid scope issues with CTEST_RUN_CURRENT_SCRIPT. If
ctest_start() is called within a function scope, the value of
CTEST_RUN_CURRENT_SCRIPT that it sets doesn't make it to the global
scope. With this change, ctest_start() no longer sets
CTEST_RUN_CURRENT_SCRIPT, and instead sets a field directly in
cmCTestScriptHandler. The old behavior of CTEST_RUN_CURRENT_SCRIPT
has also been kept for projects and tests that rely on setting it.
The last KDevelop3 release was many years ago, in 2008 I think.
I haven't seen or read about anybody using KDevelop 3 since a
long time, so I think it can safely be removed from CMake.
KDevelop 4 (first released in 2010) has its own proper CMake
support now, independent from this generator.
Alex
Refactoring in commit v3.4.0-rc1~321^2 (cmMakefile: Remove special
handling of INCLUDE_REGULAR_EXPRESSION, 2015-04-04) accidentally broke
propagation of the include regex to subdirectories. Refactoring in
commit v3.5.0-rc1~319^2~1 (cmState: Initialize properties immediately,
2015-10-07) moved maintenance of this value from `cmMakefile` to
`cmStateSnapshot`.
Restore propagation of the `INCLUDE_REGULAR_EXPRESSION` to
subdirectories and add a test to cover it.
Fixes: #17676
b4e6911b Autogen: Tests: Extend AUTOUIC include patterns test
ff91a5d5 Autogen: Tests: Extend AUTOMOC include patterns test
a1d491ca Autogen: AUTOMOC/UIC fix for moc/uic include on the first line
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1670
