Remove test_minor_version_does_not_reconfigure_wipe() because when run
during dev cycle that test reconfigure with .99 -> .100 which is
considered a major version change now. It is covered by a more efficient
internal test now anyway.
While at it, remove no-op `with Path(self.builddir):` statement, the
intention was clearly to set workdir.
Fixes: #9260
This adds a new category of tests that does not need to run on all
platforms during CI. For now only run them on Linux runners because they
are not the bottleneck.
Currently, if the test fails to produce XML (or valid XML) then the test
fails with a backtrace. It's actually pretty easy to get into this
situation, a total failure of the test will result in no XML being
written (this can happen, for example, if rpaths to gtest are not
correctly set up). If we can't read the test, go ahead and complete
using `TestRunExitCode.complete()`, which will fail for the bad exit
code.
When setup creates a Visual Studio environment, a message is logged
which contains a path to the build directory. Typically, this path is
converted to a relative path prior to printing. If the path cannot be
converted to a relative path (e.g., because buildpath is on a different
drive from the cwd), print out the full path instead of failing with an
unhandled exception.
This will always be wrong, because when a directory component is
provided we need to match an exact filename on a manual search path, for
example find_program with dirs: or the current meson.build subdir.
If we ever get this far, shutil.which will do the same "is there a
dirname, if so just check whether the filename exists relative to
cwd"... except that the documented meson lookup path is that we check
relative to meson.build subdir, not relative to the cwd, and the cwd
could be anything, but is probably the root sourcedir.
Since internally, meson does not actually os.chdir into the sourcedir,
it could be absolutely anything at all, though.
...
The actual returned name for shutil.which(name) given a literal pathname
with a directory component is "return name" without adding the absolute
path, which means that this is double-broken. Not only does it find
things we didn't expect, the resulting ExternalProgram object doesn't
have the correct path to the program, so it will report "found" and then
fail to actually run when the current directory is changed, for example
by ninja -C.
Fixes#9262
This removes the warning when using default_options without fallback
kwarg completely because a subproject does not know if the main project
has an implicit fallback or not, so it could set default_options even if
not fallback is available at all.
Fixes: #9278
When a static library B to a static library A with generated sources, B
directly references the object file corresponding to the generated source in
A. For that reference in B object_filename_from_source() is used. But A did
not specify the object file name, ending up with cl.exe's default.
Fixes: #9235
Since we changed to using a json file to avoid over long command lines
we created a situation where the generated files may not be ready when
the depscan happens. To avoid that, we need to add all of the generated
sources as order deps.
Fixes: #9258
This command is useful to e.g. update a cloned subproject which does not
have its packagefiles merged, or which has updated packagefiles. This
does not strictly require internet, so if we can satisfy the ref, simply
log a warning and continue.
This enables a workflow where for network-isolated builds, the
subproject is cloned and moved into place and later on the build
orchestrator runs `meson subprojects update` without a network interface
to initialize the subproject.
Since it cannot resolve `import typing as T` in order to figure out that
T.* is doing annotation-worthy stuff.
Since T.cast('Foo') is not actually using Foo except in an annotation
context (due to being a string) it requires extra work to resolve, and
the only thing that would currently work is actually using
'typing.cast'. However, we have decided to not use it except as T...
Since this import is only imported during mypy it's not so bad to noqa
it.
It is imported from a subpackage in __init__ alongside a big list of
other things which are all exported. And elsewhere, this import is
re-imported by other code. It's pretty clearly an oversight that it
didn't get added to __all__
Alias commands did not work with the vs backend, due to trying to access
target.command[0] with an empty command. Fix this by just not emitting a
CustomBuild node for alias targets - the project references are enough to
trigger the necessary actions.
Fixes: #9247
It's broken in various circumstances, no one seems to actually use it,
CI doesn't test it, no one is committed to maintaining it, etc. etc.
etc.
Also, projects doing trivially reasonable things, such as generating
"foo/util.py" and "bar/util.py", create clashing output names. This will
never, ever, ever, ever work with layout=flat.
Closes#996Closes#1521Closes#1908Closes#7133Closes#7135Closes#7480Closes#8378
Until we invoke interpreter.Interpreter(b, ...) the coredata options
still have their default values and thus cannot be used sensibly.
Currently the warning never shows (other than, unsurprising in
retrospect, during --internal regenerate).
In commit 3c4c7d0429 the qresource
variable stopped being overwritten with a mesonlib.File, which is
reasonable. However, one call site for it which relied on being a built
file did not get renamed when needed.
Make the build target use the built file.
Older meson versions would not honor the `<lang>_args` and `<lang>_link_args` in the built-in
options section, add a note about this to the relevant section as it can cause quite some surprises
when using a crossfile with an older meson version.
The clang compiler now reimplements and re-checks the c_std and cpp_std
options in order to use them for objc as well, but it didn't
consistently support the same options. First it completely excluded all
the gnu ones, and then it added a handful of them but not for C++.
Be fully consistent -- or at least as consistent as we can be, given a
minimally working fix. (The C/C++ compiler mixin actually gates
different stds depending on detected clang version, we do not do that
here.)
Fixes regression in c54dd63547
Fixes incomplete fix from #8766 (which didn't fix objcpp at all)
Fixes#9237
This requires a bit of extra code because the version might change, but
otherwise it fits in the existing AllPlatformTests.test_summary testcase
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Dependencies are currently printed as
[<mesonbuild.mlog.AnsiDecorator object at 0x7faa85aeac70>, ' ', <mesonbuild.mlog.AnsiDecorator object at 0x7faa85aeab50>]
This was introduced in commit adb1b2f3f6, due to
an incorrect type annotation on the AnsiText constructor. Fix both the
annotation and the usage.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Allow using the links method to test that the C++ driver (e.g. g++) can be used to
link C objects. One usecase is that the C compiler's libsanitizer might not be
compatible with the one included by the C++ driver.
This is theoretically backwards-incompatible, but it should be treated as a
bugfix in my opinion. There is no way in Meson to compile a .c file with the
C++ driver as part of a build target, therefore there would be no reason to
do something like meson.get_compiler(meson.get_compiler('cpp').links(files('main.c')).
Fixes: #7703
In some cases, link tests would like to use objects provided by a compiler
for a different language, for example linking a C object file with a C++
compiler. This kind of scenario is what link_language is for, but it is
impossible to test that it works with a linker test.
This patch implements the required support in the Compiler class. The
source code compiler is passed to the Compiler.links method as an
argument.
Xavier Claessens
Dylan Baker
Jesse Natalie
Eli Schwartz
Ryan Kuester
Andres Freund
Rohit Goswami
miebka
ePirat
Mat Cross
Paolo Bonzini