Pass CMAKE_BUILD_TYPE into the test on generators that use it so that
the per-config part of the test can activate as needed. Do not make the
per-config part conditional on the Debug configuration because the
generator expressions evaluate to empty in other configurations.
Skip the per-config source case with the Ninja generator because it does
not currently work. cmQtAutoGenerators::InitializeAutogenTarget needs
to know the list of source files on a target, but generator expressions
in the list cannot be evaluated until after CreateGeneratorTargets has
been called. That cannot happen until after Autogen targets have been
generated. It is a chicken-and-egg problem.
07d44d63 cmake: Remove confusing duplication.
ea819b29 cmMakefile: Remove unused method.
6ad86c7f cmMakefile: Remove bad comment.
fca2b542 cmMakefile: Internalize setting of CMakeInstance on Properties.
7bb4e3db cmMakefile: Out-of-line Home directory accessors.
6241253a cmake: Out-of-line Home and Start directory methods.
0ee3ccb3 cmake: Fix variable name bugs.
57dd094e Use vector, not list for cmCommand storage.
6deb43e6 Remove some files which do not need to be in BootstrapCommands.
ecdb1b3b Add some missing includes.
04b307b9 cmake: Simplify CommandExists method.
0f1f324b cmake: Rename oddly named variables.
275185ac cmake: Constify GetCommand method.
c57f086a cmake: Don't lower-case a string needlessly.
23368c9b cmake: Use make_pair instead of Foo::value_type.
14c70b8c cmake: out-of-line try compile state methods.
...
9660a3cc Makefile: Fix multiple custom command outputs with one missing
5c08e255 KWSys SystemTools: Teach Touch with !create to succeed on missing file
The use of "cmake -E touch_nocreate" added in commit v3.2.1~4^2
(Makefile: Fix multiple custom command outputs regression, 2015-03-06)
caused builds to fail when one of the outputs is intentionally not
created. This was fixed by our parent commit by making touch_nocreate
succeed when the file is missing. Add a test case covering it.
For the Watcom WMake generator, check for the SYMBOLIC source file
property separately on each output. The mark is needed on outputs that
are not really created to tell 'wmake' not to complain that it is
missing. The mark is also needed on outputs that are created or 'wmake'
will not consider them out of date when they exist.
Inspired-by: Ben Boeckel <ben.boeckel@kitware.com>
Port most CMakeTestBadCommandLines test cases to RunCMake.CommandLine
and drop the former test. Add validation of expected results, which was
not done by the old test.
882f48e5 Link libraries by full path even in implicit directories
318cd370 Help: Add link target for Find Modules section of cmake-developer.7
1535dcd8 Tests: Teach RunCMake to optionally merge command output to stdout
When CMP0003 was first introduced we wanted to link all libraries by
full path. However, some projects had problems on platforms where
find_library would find /usr/lib/libfoo.so when the project really
wanted to link to /usr/lib/<arch>/libfoo.so and had been working by
accident because pre-CMP0003 behavior used -lfoo to link.
We first tried to address that in commit v2.6.0~440 (Teach find_library
to avoid returning library paths in system directories, 2008-01-23) by
returning just "foo" for libraries in implicit link directories. This
caused problems for projects expecting find_library to always return a
full path. We ended up using the solution in commit v2.6.0~366 (...
switch library paths found in implicit link directories to use -l,
2008-01-31). However, the special case for libraries in implicit link
directories has also proven problematic and confusing.
Introduce policy CMP0060 to switch to linking all libraries by full path
even if they are in implicit link directories. Explain in the policy
documentation the factors that led to the original approach and now to
this approach.
Prefix test commands with the CROSSCOMPILING_EMULATOR property
for target executables. This allows test suites to be run on the host
when crosscompiling.
If the CMAKE_CROSSCOMPILING_EMULATOR variable is defined, and
CMAKE_CROSSCOMPILING is TRUE, then use CMAKE_CROSSCOMPILING_EMULATOR to run
the try_run executables. This prevents the need to populate
TryRunResults.cmake when cross compiling.
Add CROSSCOMPILING_EMULATOR target property for executables. This is used by
subsequent patches to run exectuables created for the target system when
crosscompiling. The property is initialized by the
CMAKE_CROSSCOMPILING_EMULATOR variable when defined.
Some tests may need to read "cmake --build" output passed through from
native build tools and do not know if it will be on stdout or stderr.
Optionally use the same variable for the execute_process output so
that it merges them and we can always match using expected stdout.
VS 2013 originally claimed to support initializer lists but a bug was
found in which it generated bad code silently. For this reason we
previously considered support to not be present. However, Update 3 adds
a hard error on cases that previously generated bad code, so it is now
safe to use initializer lists with VS 2013 Update 3 or greater. At
worst a compiler error will be issued in the cases that do not work, but
that is no different from any other compiler-specific workaround a
project code may need.
The only reason this failed to compile on VS 2013 was because the
compiler uses different initializer_list constructor argument types than
our dummy implementation. The standard does not specify the non-default
constructor argument types for initializer_list. Use a template to
match any two-arg constructor a compiler might select (e.g. begin/end or
begin/len). Use #error to preserve the error on VS 2013.
RPM, DEB and archive packages were not created
in cases when CPACK_<generator>_COMPONENT_INSTALL
was set to ON and a single component existed.
Patch also changes CPackRPM test to implicitly
test for this case.
XAML files are by default of type Page in the vcxproj and can be
overriden by setting the VS_XAML_TYPE property. The .cpp and .h file
of the same name are automatically added as depending on the XAML file.
New VSXaml test builds a basic XAML WindowsStore 8.1 app with VS2013.
Otherwise an #include directive on the first line of a source file is
ignored if the file contains a Byte-Order-Mark.
Suggested-by: Aleksey Konovalov <konovalov.aleks@gmail.com>
Add policy CMP0059 to cover this change. The property has been
deprecated since CMake 2.4 anyway.
This will help clean up cmMakefile -- the DefineFlagsOrig member should
not need to exist.
When the cmake sources are all set to read-only (e.g. after importing
into perforce), the CMake.ELF test fails because the copy of the binary
is also read-only and cannot be modified. Fix this by copying the
binary without source permissions.
When matching ctest command-line output, we must account for the
formatting of times that take 10 seconds or more. The values are
right-justified, so use " +" to match any amount of space before them.
RPM packages can contain symbolic links to relative paths - including
support for multiple relocation paths through generation of post install
relocation scripts. Add basic support with limitations described in
documentation.
Revert commit v3.0.0-rc1~175^2~20 (add_dependencies: Disallow use with
INTERFACE_LIBRARY, 2013-12-25). Teach our dependency analysis to
transitively follow INTERFACE target utility dependencies as was done or
IMPORTED targets in commit v2.8.6~127^2~1 (Allow add_dependencies() on
imported targets, 2010-11-19). Extend the InterfaceLibrary test with a
case to cover header generation for a header-only INTERFACE library via
a custom target.
Extend the ``DATA{Dir/,...}`` syntax with a new ``RECURSE:`` option
to enable recursive matching of associated files. This will allow
an entire directory tree of data to be referenced at once.
bd9c7f9b Ninja: Add policy to require explicit custom command byproducts
ed8e30b0 cmGlobalNinjaGenerator: Optimize handling of known build outputs
ad094f43 cmGlobalNinjaGenerator: Fix spelling of "unknown"
82a37d3c cmGlobalNinjaGenerator: Drop unused member
4178cd88 Help: Add notes for topic 'xcode-xctest'
87a4b858 Tests: Add XCTest example to test Frameworks and Cocoa App Bundles
ba14510b OS X: Add FindXCTest module
3714955b OS X: Add handling for XCTest bundles
54a5cdbb Tests: Compute Xcode version for any generator on OS X
Add policy CMP0058 to avoid generating 'phony' ninja rules for unknown
custom command dependencies. This requires projects to specify their
custom command byproducts explicitly. With this requirement we no
longer have to assume that unknown custom command dependencies are
generated and can instead simply assume they are source files expected
to exist when the build starts. This is particularly important in
in-source builds. It is also helpful for out-of-source builds to allow
Ninja to diagnose missing files before running custom command rules that
depend on them.
GLOB lists directories by default and GLOB_RECURSE does not.
LIST_DIRECTORIES enables user to control the behavior explicitly for
consistently for both GLOB and GLOB_RECURSE.
Store the version in CMake_TEST_XCODE_VERSION for use by tests that work
with any generator on OS X but may depend on the Xcode version providing
the tools.
Component specific attributes test passed
even when attribute was able to leak to
the next component as library package is
the last that is generated. This patch
fixes the test as header package is generated
in the middle so leakage causes the test to
fail.
Any environment vars that were configured for a test via the
ENVIRONMENT property will now be output when the test is run
with verbose logging enabled.
Implement a brand new FindMatlab module:
- Add support for versions and components.
- Find Matlab and its version in a more precise and multiplatform way.
- Add API to create a new mex extension with documentation.
- Add API to add matlab unit tests (with or without the unit test framework).
- Find as much as possible based on a single Matlab_ROOT_DIR cache entry
and allow the user to change it to re-find everything.
Fixed architecture test that was
missing architecture in regular
expressions - bug was detected
on older rpm versions where the
check failed.
Extended architecture test that takes
into account older versions of rpm.
This patch is related to 15442.
232a6883 Help: Add release notes for target-language-genex.
9e168941 File(GENERATE): Process genex evaluation files for each language.
b734fa44 Genex: Allow COMPILE_LANGUAGE when processing include directories.
0b945ea9 Genex: Allow COMPILE_LANGUAGE when processing compile definitions.
5c559f11 Genex: Enable use of COMPILE_LANGUAGE for compile options.
e387ce7d Genex: Add a COMPILE_LANGUAGE generator expression.
4a0128f4 VS6: Compute CMAKE_*_FLAGS and COMPILE_DEFINITIONS* only when needed
Move the test cases from Tests/CMakeTests/ConfigureFileTest.cmake.in
over to use the RunCMake.configure_file infrastructure. This does much
more robust verification of CMake output for each test case, and would
have caught the regression fixed in our parent commit.
Rename the example jacoco.xml file to be jacoco.xml.in to stop CMake
from apptempting to calculate Jacoco Coverage when running over itself.
Enclose a push of -1 to the coverage vector to only happen if there is a
fin to calculate for. This prevents a crash if the target file doesn't
exist.
d1082448 Tests: Extend RunCMake.CommandLine to cover 'cmake --build' for 'ninja -v'
ce935ebe cmake: Teach --build to honor CMAKE_VERBOSE_MAKEFILE for Ninja
A more-specific error message is always displayed earlier in the
output if any real error occurred. This final summary message
is distracting to readers searching through the output for the
word "error". Simply drop it.
Use test infrastructure added by commit 1cbb1562 (Fix handling of
relocation prefix parent directories, 2015-02-26) to cover the same use
cases from tests added by commit 5857ca5e (CPackRPM: Drop explicit
handling of '@' symbols that breaks them, 2015-01-07) and drop the
latter.
In commit 3ec02547 (CPackRPM: Allow multiple path relocation prefixes
for one package, 2015-01-21) a regression was introduced that causes
parent directories of relocation paths to be incorrectly included in
the rpm. Fix this and make the test case more strict to cover it.
Allow URL templates to contain a %(algo:<key>) placeholder that is
replaced by mapping the canonical hash algorithm name through a map
defined by the <key>.
Extend the Module.ExternalData test to cover the behavior.
Extend the RunCMake.ExternalData test to cover error cases.
Everything they cover is now covered by RunCMake.ctest_submit test cases
(except ctest_coverage LABELS which were not actually checked anyway).
Drop these redundant tests.
e2b9e7f7 Tests: Drop unnecessary stderr matching from RunCMake.ctest_memcheck
7ce9f6e2 Tests: Teach RunCMake to tolerate 'Error kstat returned' lines in test output
69de0f7e Tests: Teach RunCMake to tolerate Guard Malloc lines in test output
57f2aa7c Tests: Teach RunCMake to tolerate BullseyeCoverage lines in test output
b04c8ec3 Tests: Match curl error in RunCMake.ctest_submit FailDrop-* cases
1d82105e Tests: Rename RunCMake.{CTestMemcheck => ctest_memcheck}
d7662141 Tests: Rename RunCMake.{CTestSubmit => ctest_submit}
9b50388b Tests: Split RunCTest helper out of RunCMake.CTest(Submit|Memcheck)
76e7c22b Tests: Rename RunCMake.CTestMemcheck internals to match CTestSubmit
The change in commit v2.8.12~218^2 (CPackRPM protect '@' character in
filename processed in the spec file, 2013-07-05) was not necessary after
commit v2.8.12~439^2 (Add support for componentized USER spec file,
2013-04-01). The latter replaced ${VAR} references in the spec file
template string with \@VAR\@ references, thus protecting '@' symbols
automatically. This caused CPackRPM to break paths with @ symbols.
Revert the change to fix the behavior, and add a test case.
Otherwise find_library is unable to lookup the XCTest framework which
is not located in the SDK serach path:
In the 10.10 SDK the SDK frameworks are located here:
$DEVELOPER_DIR/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.10.sdk/System/Library/Frameworks
whereas the Platform SDKs are located here:
$DEVELOPER_DIR/Platforms/MacOSX.platform/Developer/Library/Frameworks
Signed-off-by: Gregor Jasny <gjasny@googlemail.com>
If a feature is added multiple times via ADD_FEATURE_INFO it should
appear only once in FEATURE_SUMMARY.
Signed-off-by: Gregor Jasny <gjasny@googlemail.com>
bb9d71b4 Replace loops with algorithms.
4afe6c26 cmAlgorithms: Add cmReverseRange adaptor.
a3a0a8c2 cmAlgorithms: Add cmFindNot algorithm.
8c74a41f cmRST: Replace two erase with a rotate and larger erase.
61fe1919 cmAlgorithms: Update concept requirement to FowardIterator
09d6125b cmAlgorithms: Move cmRotate out of 'implementation detail' namespace.
8ed6ecac cmRST: Move two algorithms beside each other.
dfe49c20 cmRST: Use std::min where appropriate.
21b0654a cmGlobalGenerator: Convert set insert algorithm to vector algorithms.
416df93a Convert some raw loops to cmWrap.
37b88d34 cmAlgorithms: Add cmWrap.
a2818093 Use cmJoin where possible.
76207b08 cmCacheManager: Replace loop with algorithm.
60c3bb73 cmGlobalGenerator: Replace loop with algorithm.
05fec779 cmTarget: Port loop to algorithm.
9c225767 cmGlobalGenerator: Replace set::insert algorithm with cmRemoveDuplicates.
...
The Dummy*-stderr.txt files were needed only to match platform-specific
memcheck tooling output and verify that stderr is otherwise empty. Now
that the RunCMake infrastructure knows how to strip such lines before
matching, we can simply drop these files and use the default empty
string match.
When RunCMake tests run on Solaris, the output may contain unexpected
lines of the form "Error kstat returned...". These lines are printed by
SystemInformationImplementation::RunProcess when called from
SystemInformationImplementation::ParseValueFromKStat (see issue #12066).
Until someone investigates why kstat returns values outside the range it
documents, simply remove such lines from the actual output before
matching it against the expected output.
When RunCMake tests run under Xcode Guard Malloc, Guard Malloc may add
lines of the form "<tool>(<pid>) malloc:..." to the output. Remove such
lines from the actual output before matching it against the expected
output.
When RunCMake tests run under dynamic analysis, Bullseye may add lines
of the form "BullseyeCoverage..." to the output. Remove such lines from the
actual output before matching it against the expected output.
Since jsoncpp 0.7.0 (2014-11-20) the upstream may provide a CMake
package configuration file such that find_package(jsoncpp) will find a
jsoncppConfig.cmake file. In order to avoid conflicting with this
(especially on case-insensitive filesystems), and since we always prefer
projects to provide package config files (that they maintain), it is
better to not provide FindJsonCpp publicly.
Move FindJsonCpp into a private source directory that is not installed
so that we can still use it for building CMake itself.
Reported-by: Ryan Pavlik <ryan.pavlik@gmail.com>
The find_package command already knows how to compute installation
prefixes from PATH. Use the same approach to establish prefixes for
find_library, find_file, and find_path to use to look in directories
like "<prefix>/lib[/<arch>]" and "<prefix>/include" for libraries and
headers. This will reduce the amount of configuration end users need to
do to establish a work environment rooted under a specific prefix.
Several tests use slight variations of the same logic to enable CPack
RPM tests. Consolidate this logic into one check before any tests are
added. Look for 'rpmbuild' only on Linux and only when the test build
tree does not have spaces in the path. In particular, this will make
the result available in time for the RunCMake.CPackRPM test to be
activated even if CMake is configured exactly once.
This will allow per-config destinations for targets in EXPORT sets.
Using multiple install(TARGETS) with separate CONFIGURATIONS is
rejected as a target appearing more than once in an export set.
Now instead one can write
install(TARGETS foo EXPORT exp DESTINATION lib/$<CONFIG>)
to get a single logical membership of the target in the export set
while still having a per-config destination.
f7e33820 Add release notes for export-interface-source-files.
6da65b39 Allow export of targets with INTERFACE_SOURCES.
736bcb96 Tests: Move IfacePaths test stderr files.
d9f8390d Tests: Run IfacePaths tests with a parameter.
cc152094 Tests: Split part of include_directories test to a generic location.
c0d8e715 RunCMake: Allow specifying the stderr file for a test.
68d29f51 RunCMake: Allow specifying the directory to run tests in.
1c3918ff RunCMake: Remove unneeded files.
0f870234 Merge branch 'backport-no-global-setlocale' into no-global-setlocale
cd408d93 Add setlocale() calls around use of libarchive APIs (#14934, #15377)
87be2e14 Do not call setlocale() globally in CMake applications (#15377)
The moved tests are related to paths in INTERFACE_INCLUDE_DIRECTORIES
in general, and when exporting, and covering cases of the install
location being within the build or source prefix etc. Other build
properties containing paths should have similar tests, so this
allows some parameterization while keeping the preparation of the
various directory structures.
Revert the changes made by commit v3.1.0-rc1~406^2~1 (Encoding: Add
setlocale() to applications, 2014-05-30) and commit v3.1.0-rc1~406^2
(Encoding: Change to only set LC_CTYPE, 2014-06-11), and other setlocale
calls added later in their spirit. CMake has not been taught how to
deal with non-C locales everywhere. We do not define any functionality
for character conversions for non-ASCII strings. Another solution will
be needed to address the original problem motivating addition of
setlocale() calls.
This bug caused c_function_prototypes to not be recorded at configure
time when compiling with -std=gnu99 or similar. In the case of feature
recording, that was not a problem, because the logic in
CMakeDetermineCompileFeatures.cmake currently assumes that a feature
present for an earlier standard is present for a later standard.
However, the detection strings are also used in WriteCompilerDetectionHeader,
so the feature macro has been defined to '0' when using a later language
dialect.
Fix that by not checking the existence of the __STDC_VERSION__ macro at
all when detecting C90 features.
809a5a5e Help: Add notes for topic 'CTestCoverageCollectGCOV-refinements'
03c0812c CTestCoverageCollectGCOV: Fix handling of international characters
8caa4e72 CTestCoverageCollectGCOV: Add test case
5c828cc8 CTestCoverageCollectGCOV: Allow custom flags to gcov
30cb628e CTestCoverageCollectGCOV: Fix handling of large file counts
Define an empty string in CMAKE_<LANG>_STANDARD_DEFAULT to mean that
the toolchain has no notion of lanuage standard levels. In this case
the <LANG>_STANDARD[_REQUIRED] properties will have no effect.
Update the RunCMake.CompileFeatures test to exclude the
LinkImplementationFeatureCycle test when there is no standard default.
It can never fail because no use of specific features will adjust the
CXX_STANDARD level required for any target since the standard levels
have no meaning in this case.
ab9fa54d Xcode: Switch to internal CMAKE_MAKE_PROGRAM lookup by generator (#15324)
11e2e6ca Xcode: Select make program at build time
e4055a61 Xcode: Add internal API to find xcodebuild
The "cmakexbuild" wrapper is not needed for Xcode 4 and above, and the
path to it may change when CMake moves. Avoid storing a specific path
to a build program in CMakeCache.txt and instead compute the value for
CMAKE_MAKE_PROGRAM on demand. However, if a user does set the value
explicitly then honor it.
This does for Xcode what commit v3.0.0-rc1~260^2~4 (VS: Switch to
internal CMAKE_MAKE_PROGRAM lookup by generators, 2013-11-15) did for
Visual Studio generators.
Custom command path normalization added in commit v3.1.0-rc1~471^2
(add_custom_command: Normalize OUTPUT and DEPENDS paths, 2014-05-28)
broke use of OBJECT_DEPENDS to bring in custom commands because the
latter paths were not normalized too. Normalize them and add a test
case.
Reported-by: Daniel v. Gerpen
With the Makefile generators we expect that touching or modifying the
input file of a configure_file, or removing its output, will cause CMake
to re-run on the next build. Extend the RunCMake.configure_file test
with a case covering this. Also check that CMake does not re-run if
nothing has changed.
With the Makefile and Ninja generators we expect that touching the input
source file for a try_compile will cause CMake to re-run on the next
build. Extend the RunCMake.try_compile test with a case covering this.
Also check that CMake does not re-run if nothing has changed.
Whenever feature support is added to a compiler, the CompileFeatures
test needs to be updated to set expected availability of features.
Add #error directives to ensure the test fails if expectations are
not set.
Suggested-by: Stephen Kelly <steveire@gmail.com>
Read file names from the lines of a specified file. Reject input lines
starting in '-' to leave room for option parsing to be added later. Add
just '--add-file=' now to allow files starting in '-' to be specified.
d0adcccb try_run: Add tests for LINK_LIBRARIES with mock libraries.
223c5cb7 try_run: Add test for bad link libraries.
e2b1f058 try_run: Add support for LINK_LIBRARIES option.
Most functionality is already implemented in Source/cmCoreTryCompile.{h,cxx}.
Document and improve argument parsing.
This functionality is already being used by a number of modules, like
CheckCSourceCompiles.cmake, but it is not documented.
Shorten the add_definitions command test directory and target names to
avoid creating really long paths that fail with some tools on Windows.
While at it, remove unnecessary project() command calls.
f3e0b6f1 CTestCoverageCollectGCOV: Add module to run gcov
6dd980e0 ctest_submit: Make CDASH_UPLOAD mode arguments more strict
5dc33f89 ctest_submit: Add CDASH_UPLOAD mode to upload files to CDash
965a50dc Features: Fix GNU 4.4 and 4.5 C standard level flags
7f4154a4 Features: Fix CompileFeatures non-feature tests for space in path
f40c19b5 Features: Fix CompileFeatures test for C non-features
When curl is given a URL of the form '<something>://' it tries to
resolve '<something>' as a host name. If the host happens to exist and
have a server then the drop might actually appear to work. Instead use
an explicit '-no-site-' host to ensure it cannot connect.
Reported-by: Gilles Khouzam <Gilles.Khouzam@microsoft.com>
Fix the <lang>_non_features try_compile calls to work correctly when
there is a space in the path. Otherwise they all fail due to the space
instead of the lack of a feature.
Add an option to set the mtime of entries in a tarball so that one can
create a tarball with a consistent content hash (e.g. MD5) for a given
set of files regardless of their current timestamps on disk. This will
be useful for submission of tarballs to CDash, which tracks content
hashes to avoid duplication.
Inspired-by: Bill Hoffman <bill.hoffman@kitware.com>
The CXX_STANDARD property validates the values allowed for it,
and '90' is not a valid value for it. However, the validation is only
done for languages enabled for the target, and as C_undefined has no
CXX files, that language is not enabled for it.
This amends commit v3.1.0-rc3~23^2 (WCDH: Make the header -Wundef
safe for the C language., 2014-11-24).
004e1540 Features: Record for GNU 4.4.
2a5ca650 Features: Wrap failure-test in UNIX condition.
1ae2c6b2 Features: Blacklist cxx_constexpr test for GNU 4.5.
c66e3317 Features: Use a more-common feature in cycle-test.
c43a6dc5 Features: Update comment in test to match the code.
78259135 Features: Test presence of cxx_auto_type with genex.
7b9fc88b Features: Remove outdated comment.
00194ea7 Help: Add notes for topic 'SolarisStudio-compile-features'
bcb04783 Features: Record for SolarisStudio 12.4.
536c535c Features: Adjust cxx_variadic_templates unit test for SolarisStudio.
5d57970d Features: Use variable in cxx_inheriting_constructors test.
69182ce4 Features: Ensure that the cxx_auto_type test is correct.
Replace the TODO comment with calls to run_ctest_submit_FailDrop. For
now leave curl errors out of expected stderr. They can be ported over
from CTestTestFailedSubmit-* tests later.
Cover command invocation argument errors. Prepare infrastructure
to cover failed 'drop' cases, but do not cover them yet because
we need to collect error messages from each platform.
We require that the $<$<COMPILE_FEATURES:cxx_feature>:empty2> generates
a '1' when CXX11 is enabled. GNU 4.4 does not support cxx_nullptr, but
does support cxx_auto_type.
The purpose of that test is to cover the case where the genex
reports '1', and the feature is chosen to be present on all/most
supported compilers. GNU 4.4 does not support cxx_nullptr.
The change in commit 1f19ac4d (Features: Adjust cxx_variadic_templates
unit test for GNU < 4.7., 2015-01-11) pacified GNU 4.6, but leaves
SolarisStudio 12.4 complaining:
"cxx_variadic_templates.cpp", line 5: Error: Partial specialization for Interface<Is...> has identical arguments.
1 Error(s) detected.
Implement a preprocessor test for using the partial specialization
workaround needed by GNU 4.6.
SolarisStudio considers 'auto' to be a storage class specifier in
C++98 mode (as appropriate), and considers variables without a specified
type to be of type int. So, it treats
auto x = 3.14;
as
auto int x = 3.14;
which in C++98 mode is equivalent to
int x = 3.14;
and it does not fail to compile as expected.
Change the test to use a reference so that the type must be known.
770ba876 Help: Add notes for topic 'GNU-4.6-compile-features'
b15c008f Features: Record for GNU 4.6.
f13a2eb1 Features: Adjust the RunCMake test to use more-common features.
1f19ac4d Features: Adjust cxx_variadic_templates unit test for GNU < 4.7.
0798d1e5 Features: Extend the generalized_initializers test for GNU < 4.7.
938bd94e Features: Test __GXX_EXPERIMENTAL_CXX0X__ macro for GNU < 4.7 compatibility.
462c630b Features: Don't test __cplusplus value for CXX98 on GNU.
222ec86f Features: Test an old value of __STDC_VERSION__ for GNU < 4.7 compatibility.
6e909035 Features: Record C/CXX dialect flags for GNU 4.6.
bfc995cc Features: Remove wrong content from else() condition.
1532b921 Features: Make cxx_noexcept available from GNU 4.6.
1f4649e6 Features: Update the default_dialect test for old GNU-like compilers.
a60027a6 Features: Ensure appropriate return value from feature test macros.
The unit test for this fails with GNU 4.6:
Building CXX object CMakeFiles/test_cxx_variadic_templates.dir/cxx_variadic_templates.cpp.o
CompileFeatures/cxx_variadic_templates.cpp: In static member function ‘static int Interface<I, Is>::accumulate()’:
CompileFeatures/cxx_variadic_templates.cpp:18:31: sorry, unimplemented: cannot expand ‘Is ...’ into a fixed-length argument list
CMakeFiles/test_cxx_variadic_templates.dir/build.make:54: recipe for target 'CMakeFiles/test_cxx_variadic_templates.dir/cxx_variadic_templates.cpp.o' failed
The workaround is to use a specialization:
http://stackoverflow.com/questions/1989552http://stackoverflow.com/questions/11297376
Add support for a special URL template to map the fetch operation
to a project-specified .cmake script insead of using file(DOWNLOAD).
Extend the Module.ExternalData test to cover the behavior.
Extend the RunCMake.ExternalData test to cover error cases.
dcd72a74 Help: Add notes for topic 'Xcode-clang-compile-features'
3ad893b5 Features: Record for historical Xcode clang versions.
98965fb1 Features: Record dialect flags for AppleClang 4.0+.
The second hunk of commit 07d1f6fc (Features: Properly evaluate if the
compiler supports cxx_final, 2014-12-31) was a workaround for a bug in
the COMPILE_FEATURES generator expression that caused it never to return
0. Revert the workaround so we can fix the bug instead.
Revert commits:
2d738ce3 Help: Add notes for topic 'feature_record_msvc'
f73718c9 Features: Enable writing of MSVC compiler feature header.
64c30bdc Features: Record for MSVC C++ 2015 and MSVC C 2010-2015.
225c0ef8 Features: Record for MSVC 2010-2013.
This topic was merged to master prematurely, so remove it.
Commit v3.1.0-rc1~297^2~5 (cmTarget: Drop 'head' argument from
GetSourceFiles, 2014-07-10) exposed a dormant bug in source file computation,
causing the test case to regress. After that commit, the source file
computation and caching finds an existing container of source files. Prior to
that patch, the GetSourceFiles method was called with either a null pointer
for the head cmTarget, or it was called with the this pointer. The
processSources method is eventually called, which normalizes the difference
between the null pointer and the this pointer for the head target. However,
the cache key depends on the actual pre-normalized pointer. The change in
that commit caused the entry to be found in the cache where it was not before,
which resulted in incorrect behavior.
Prior to that commit, the test case also fails if the GetSourceFiles overload
taking a vector<cmSourceFile*> is changed to normalize the head target at
the beginning of the method:
cmTarget const* head = head_ ? head_ : this;
Such a construct was correctly used in other locations where similar caching
was in place, before being removed in commit v3.1.0-rc1~310^2~25 (cmTarget:
Remove 'head' argument from GetLinkInformation, 2014-06-12), but is not
neccessary anymore.
Commit v3.1.0-rc1~674^2~2 (cmTarget: Cache the cmSourceFiles in
GetSourceFiles., 2014-04-05) introduced the caching, but fails the test case
for an unrelated reason. That unrelated error was introduced in
commit v3.1.0-rc1~688^2~5 (cmTarget: Allow any generator expression in
SOURCES property., 2014-03-18) and fixed in
commit v3.1.0-rc1~561^2~1 (cmTarget: Fix listing of source files at
configure-time., 2014-04-13).
All commits which fail the test case in the testable way do so when such a
cached version of the source files is found and returned at generate
time. In the test case, the cached content is populated at configure-time
through the use of the deprecated LOCATION property with CMP0026 OLD. The
cached content is an empty container for the bar target in the test case,
because its source file 'foo.cpp.o' is not known until generate-time. That
means that no source files are available to compute the link language and
the reported error is issued.
The actual problem is that the SourceFilesMap should be cleared after
configure time by cmTarget::ClearLinkMaps. Clear it there now.
Brad King
Raffi Enficiaud
Stephen Kelly
Gerald Combs
Nils Gladitz
Matt McCormick
Bill Hoffman
Domen Vrankar
Gregor Jasny
Gilles Khouzam
Bill Newcomb
Clinton Stimpson
Stanislav Ionascu
Zach Mullen
Gaëtan Lehmann
Joseph Snyder
Zack Galbreath
Joshua A Clayton
Robert Maynard
Robert Goulet
Justin Borodinsky