Rather than injecting `CMakeLists.txt` files into each target's
`SOURCES`, teach the generators to add them during generation using
dedicated code. This avoids mutating the original targets, and avoids
polluting `$<TARGET_PROPERTY:foo,SOURCES>` with generator-specific
content.
This also avoids listing the `CMakeLists.txt` sources in the results of
`CMAKE_DEBUG_TARGET_PROPERTIES==SOURCES` so the `RunCMake.TargetSources`
test no longer needs a separate case for IDEs.
1e56634f FetchContent: Add tests
60e74d2f FetchContent: New module for populating content at configure time
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1306
CPACK_DEBIAN_PACKAGE_VERSION variable could in the past also
contain release and epoch version so regex test should expect
the entire versioning if both CPACK_DEBIAN_PACKAGE_RELEASE
and CPACK_DEBIAN_PACKAGE_EPOCH are not set.
Also since the checks were not performed in the past the regex
test of CPACK_DEBIAN_PACKAGE_VERSION variable content should
only report author warnings instead of errors in case of the
test fail.
Fixes: #17339
The archiver output in the case of universal binaries has changed
slightly. Update our expected output to match. While at it, drop
unnecessary leading and trailing `.*`.
Add a test for this case to verify the messages. This test will also be
valuable to cover this code path in which we've had several regressions
recently.
Prior to commit v3.4.0-rc1~264^2~1 (GetPrerequisites: Add error checks
for execute_process() calls, 2015-07-29), `get_prerequisites` would
simply warn on a missing binary and not update the result list at all.
That commit accidentally made the case an error. This was fixed by
commit v3.8.0-rc1~110^2 (GetPrerequisites: Do not fail on files we
cannot find, 2017-01-10), but the fix also cleared the result list.
Clearing the list is incorrect because it is supposed to be able to
accumulate results over multiple calls.
Remove the list clearing behavior to restore the original behavior on a
missing binary.
Fixes: #17306
The change in commit 3b95ab5693 (Performance: Improve efficiency of
source file lookup in cmMakefile, 2017-08-17) broke some legacy behavior
of source file properties in which the order sources are first resolved
with extensions affects how setting properties without extensions works.
It has been reverted for now, but the discovery was made after merging
because the broken case was not covered by our test suite.
Add a test case representing the legacy behavior.
Issue: #15208
In post-3.9 development we've taught CMake to understand C++ language
standards for MSVC. The RunCMake.try_compile test needs to know whether
to expect support or not. Previously we depended on the host CMake
version to know this, but CMake 3.9 and below do not. Add special logic
to the test to account for this.
Adds support for epoch version to CPackRPM and
CPackDeb packagers.
Also adds better version checking to CPackDeb
that complies with Debian rules.
Fixes: #17057
Different CPack generators could produce checksum files with the same
name which were overwritten by each other since only package name
without extensions was used for checksum file name generation. This
patch adds package extension to checksum files to prevent collisions.
Fixes: #16840
5dc215e0 FPHSA: fix typo in comment
a8f70b32 FPHSA: replace left-over VERSION by _FOUND_VERSION.
549366cb FPHSA: Add (failing) unit test when checking EXACT on full version number
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1259
This commit changes the internal -E__run_iwyu to be -E__run_co_compile. This
is used for co-compile commands. These are tools that want to mirror the
compiler. For each compiler invocation the tool will be invoked first. This
started as a way to implement include what you use (iwyu), but has expanded
to include cpplint, cppcheck and others. Likely there will be more in the
future as well. This commit implements each one in its own function and
provides a way to add additional ones in the future with less work.
The KWSys `SystemTools::SplitProgramFromArgs` implementation goes into
an infinite loop when the value is just " " (a space). Since the
"program path with unquoted spaces plus command-line arguments"
operation it is trying to provide is poorly defined (string parsing
should not depend on filesystem content), just stop using it.
Instead consider the main two use cases the old approach tried to handle:
* The value is the name or absolute path of a program with no quoting
or escaping, but also no command-line arguments. In this case we
can use the value as given with no parsing, and assume no arguments.
* The value is a command-line string containing the program name/path
plus arguments. In this case we now assume that the command line
is properly quoted or escaped.
Fixes: #17262
9ed24280 VS: only add custom command line if it is not empty
34c4108b add HasOnlyEmptyCommandLines() method to cmCustomCommandGenerator
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1050
This commit changes the times reported by labels and subprojects to be
weighted by the PROCESSORS property. It is reported with `sec*proc`
instead of just `sec`.
This commit splits out the reporting of labels and labels used for sub
projects. If a label is a sub project label it will not be included in
the label summary. To implement this the commit creates
PrintLabelOrSubprojectSummary which is able to do the work of both
PrintLabelSummary and PrintSubprojectSummary avoiding code duplication.
91935d9e FPHSA: lower-case for cmake_parse_arguments + move config-mode related code
aec27c15 FPHSA: use more localized _FOUND_VERSION instead of VERSION.
0019bfeb FPHSA: Add (failing) unit test in case a cache variable VERSION is present
81844334 FPHSA: add (failing) unit test for found version being equal to 0.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Rolf Eike Beer <eike@sf-mail.de>
Merge-request: !1224
The known keywords for each function are obtained by scraping the
documentation for lines matching a particular regular expression. In
commit 8842a027 (ExternalProject: Improve documentation, 2017-07-09),
the docs were overhauled and the COMMAND docs subsequently matched the
regular expression when they shouldn't have. This made COMMAND appear as
a true keyword, which thwarted the special handling logic elsewhere for
the intended use of COMMAND arguments.
This commit contains a workaround for issue #17229 to force a dependency
of the patch step on the update step to ensure a predictable step order.
Fixes: #17198
