`include-what-you-use` diagnostics, in practice, are specific to
the environment's compiler and standard library. Update includes
to satisfy IWYU for our CI job under Debian 13. Some patterns:
* Types named in virtual `override` signatures no longer require
includes since the overridden signature already names them.
* A function argument's type needs to be included even if its constructor
is called only by implicit conversion. For example, constructing a
`std::function` from a lambda now requires `<functional>`.
* Some prior mysterious `<type_traits>` inclusions are no longer required.
Since commit 5420639a8d (cmExecuteProcessCommand: Replace cmsysProcess
with cmUVProcessChain, 2023-06-01, v3.28.0-rc1~138^2~8) we have not
actually terminated child processes on an `execute_process` timeout.
Similarly for other migrations from cmsysProcess to cmUVProcessChain.
Teach cmUVProcessChain clients that implement timeouts to actually
terminate remaining child processes when the timeout is reached.
Fixes: #27378
40b093649b cmake: Avoid calling GetCMakeListFile with empty directory
f6b3ba5f55 CMP0198: Maintain CMAKE_PARENT_LIST_FILE only when configuring projects
3593aa59ef cmake: Replace working mode with role
accfa7fa81 cmake: Infer command failure action from role
e290d4f2a3 cmake: Infer command set from role
bfaaec6179 cmake --workflow: Convert implementation to internal role
ea5d04975e cmake --build: Convert implementation to internal role
0b83750e14 cmake: Clarify name of role of internal instances
...
Acked-by: Kitware Robot <kwrobot@kitware.com>
Tested-by: buildbot <buildbot@kitware.com>
Merge-request: !11395
Since commit 5420639a8d (cmExecuteProcessCommand: Replace cmsysProcess
with cmUVProcessChain, 2023-06-01, v3.28.0-rc1~138^2~8) we've
occasionally observed immediate timeouts on processes that take longer
than the timeout to start, even though we only start the timer after the
child processes start. The problem is that:
* `uv_loop_init` initializes `uv_loop_t`'s cached "now" time.
* Starting processes takes time but does not update the "now" time.
* `uv_timer_start` computes expiry relative the cached "now" time,
so short timers may be expired as soon as they are started.
* `uv_run` invokes expired timer callbacks before polling for I/O
or process completion, so we "timeout" immediately.
Fix this by updating the cached "now" time via `uv_update_time` just
before starting timers. This is needed only for timers that start
before the `uv_run` event loop. Update our `uv_timer_ptr` wrapper
to make all callers consider the choice when calling `start()`.
The explicit initialization with empty paths from commit 57bdc1a2f7
(cmState: Compute and store directory components., 2015-05-04,
v3.3.0-rc1~61^2~4) has not been needed since commit 6afd35b98a (cmState:
remove unused code, 2016-06-16, v3.7.0-rc1~90^2).
Give the actual exit code of the command from the snippet as intended;
currently, this always gives 0 for sub-commands like compile and link,
as well as `ctest`.
For now, the result in build snippets will be `null`.
Create a single place to store target data to prevent duplication.
This moves `targetType` and `targetLabels` out of the snippet files
and into a target map in the `cmakeContent` file referenced by each
snippet.
Fixes: #27244
Since upstream curl commit `1a12663d06` (CURLOPT: bump `CURLPROXY_*`
enums to `long`, drop casts, 2025-07-28), the `CURLPROXY_*` constants
are integer literals instead of `enum curl_proxytype`. It turns out
that `curl_easy_setopt` has always expected a `long` anyway, and that
`curl_proxytype` is not documented for public use.
Fixes: #27178
The plain "Exit code" string had a newline that we do not have in the
other result strings. This typo was inherited from KWSys Process,
which has been fixed by its upstream.
Fixes: #27119
In commit e52eada2c2 (cmCTestConfigureCommand: Refactor command line
construction, 2025-01-28, v4.0.0-rc1~60^2~2) we switched from the
`BuildDirectory` setting, which was always an absolute path, to
`CTEST_BINARY_DIRECTORY`. Convert it to an absolute path too.
Fixes: #27026
CTest currently removes non-filename characters from CTEST_SITE and
CTEST_BUILDNAME in an inconsistent way, which leads to unconnected
information on CDash. Non-filename characters actually don't cause
any issue in CDash at all, nor are they invalid XML. The only place
where removing them may be needed is when an actual filename is
constructed.
Remove the filtering from the SafeBuildIdField function and place
it where a filename is constructed.
In commit f2596dfa0e (macOS: Work around bug in system curl 8.{3,4,5}
LibreSSL backend, 2024-07-16, v3.30.1~2^2) we tried to prefer
`secure-transport` on problematic versions of curl. However, the
`curl_global_sslset` setting must be applied before every
`curl_global_init` call, not just the first one. Otherwise a
second (or subsequent) download won't apply the work-around.
5805461074 cmSystemTools: Simplify call to FindProgram for our own executable
db0e2574cb cmSystemTools: Restore FindProgram look-up of on-disk case on Windows
5d700abda4 Source: Simplify FindProgram calls
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !10620
When `--schedule-random` is used in automated CI jobs, failures may
occur due to test order. We now log the seed. Provide a way for
developers to re-run the same order by specifying the seed.
Fixes: #26760
Co-authored-by: Brad King <brad.king@kitware.com>
This improves the output of JSON-related error messages. It adds the filename to
the output and excludes the column number.
This is particularly useful when there are multiple JSON files being read that
could be responsible for an error, ie CMakePresets.json and
CMakeUserPresets.json, or multiple instrumentation queries.
Issue: #26717
