If we update e.g. glib.wrap from wrap-git with directory=glib to
wrap-file with directory=glib-2.70 we could still have the glib
directory that is not referenced by any .wrap file any more. We should
still ignore that directory otherwise it overrides the glib.wrap we
parsed.
When _subproject_impl() is called the wrap file could not have been
downloaded form wrapdb yet, it is done when fetching the subproject.
Delay getting the variable name to when we actually need it, at that
point we are sure the wrap file has been downloaded.
Download wrap file from wrapdb automatically when it is not found
locally but we have it in wrapdb.json.
This makes for example `dependency('glib-2.0')` work out of the box
simply by running `meson wrap update-db`, even if the project does not
provide any wraps.
First, check if the env program exists. If it does, it is faster than
doing it via a python script `basically-env.py` that maybe imports all
of mesonbuild.* as a side effect of project structure.
We do not, however, use env for setting up PATH additions, since env can
override an environment variable but not extend it. So in that case we
still need to wrap the command via python.
By default, all run_targets (at least) are wrapped and now wrap via the
`env` program as they export e.g.
MESONINTROSPECT='/usr/bin/meson introspect'
https://github.com/mesonbuild/meson/pull/9287 changed the `optimization=0`
to pass `-O0` to the compiler. This change is reasonable by itself
but unfortunately, it breaks `buildtype=plain`, which promises
that “no extra build flags are used”.
`buildtype=plain` is important for distros like NixOS,
which manage compiler flags for optimization and hardening
themselves.
Let’s introduce a new optimization level that does nothing
and set it as the default for `buildtype=plain`.
- this fixes#10792
- Qt6EntryPoint(d) replaces qtmain(d) from Qt 4 & 5
- uses the same mechanism: `main: true`
- uses a special mixin class for pkg-config and config-tool dependency calculation
This catches some optimization problems, mostly in the use of `all()`
and `any()`. Basically writing `any([x == 5 for x in f])` vs `any(x == 5
for x in f)` reduces the performance because the entire concrete list
must first be created, then iterated over, while in the second f is
iterated and checked element by element.
This tests stable maintenance branches, to ensure that build directories
created with older versions of Meson continue to work without wiping,
after updating to a bugfix release.
Downgrade to LLVM-14 from LLVM-15 which is somewhat broken when using
static linking at present:
https://github.com/Homebrew/discussions/discussions/3666#discussioncomment-3681821
We can't use LLVM's lld instead of ld on macOS because we don't detect
it as an Apple linker and pass --as-needed etc to it. Even when that
is fixed and we set -lto_library etc correctly, the linker just hangs.
LLVM@14 is keg-only, so we need to add CPPFLAGS / LDFLAGS to the keg
subdir inside /usr/local
The LLVM@15 test is shared-only now and moved to the qt4 macOS job.
Those classes are used by wrapper scripts and we should not have to
import the rest of mesonlib, build.py, and all their dependencies for
that.
This renames mesonlib/ directory to utils/ and add a mesonlib.py module
that imports everything from utils/ to not have to change `import
mesonlib` everywhere. It allows to import utils.core without importing
the rest of mesonlib.
We need to support cases where the library might be called "foo.so" and
therefore we check for exact matches too. But this also allows
`cc.find_library('libfoo')` to find libfoo.so, which is strange and
won't work in many cases. Emit a warning when this happens.
Fixes#10838
A subproject could have a sub-subproject as a git submodule, or part of
the subproject's release tarball, and still have a wrap file for it
(e.g. needed for [provide] section). In that case we need to use the
source tree for the sub-subproject inplace instead of downloading a new
copy into the main project.
This is the case with GLib 2.74, it has a subproject "gvdb" as git
submodule, and part of release tarball, it ships gvdb.wrap file as well.
It is common, at least in GNOME projects, to install tests. Files goes
into various locations, including:
- /usr/lib/x86_64-linux-gnu/installed-tests
- /usr/share/installed-tests
- /usr/libexec/installed-tests
It is safe to assume that everything that goes into a "installed-tests"
subdir should be tagged as "tests" by default.
itstool detects a language code from the mo file’s basename,
so when 26c1869a14
changed the file name to be prefixed with project name,
values like “my-project-xx” ended up in the `xml:lang` attribute
of the generated page files, instead of the expected
IETF BCP 47 language tag.
Let’s fix it by passing a locale code to itstool explicitly.
Type annotation, documentation string, and implementation were doing 3
different things. Change implementation to match type annotation which
makes the most sense because it match what get_target_sources() does.
All callers only use keys from the returned dictionary any way, but
that's going to change in next commits.
This allows early exit of the project tests once a certain number of
failures are detected. For example `meson test --maxfail=1` will abort
as soon as a single test fails.
Currently running tests are marked as failed via INTERRUPT.
Resolves#9352
This was initially added in eb76ba7031 to avoid picking
it up from azure/gha images.
Looks like it is no longer set for some time now though:
https://github.com/actions/runner-images/pull/540
so should no longer be needed.
The empty string makes some meson tests fail which expect a path.
A recent msys2-runtime change now lets empty env vars through to child processes
whereas before they would get filtered out, which is why this worked before.
This does force a number of uses of `# pylint: disable` comments, but it
also finds a couple of useless global uses and one place (in the
previous commit) that an easy refactor removes the use of global. Global
is a code smell, so forcing adding a comment to disable helps force
developers to really consider if what they're doing is a good idea.
The `global` statement is only needed to assign to global variables, not
read or mutate them. So calling `global.mutate()` is fine, but not
`var = foo`, which would otherwise shadow `var`.
This removes the need for the use of the global statement. I've also
updated the test that overrides this to use mock.patch instead of hand
monkey patching.
We *mostly* just need to do the same thing. Plug in one utility method
to make sanity_check_impl find the right compile args, and plug in
DEVNULL to the test run. It's that simple.
This solves a few inconsistencies. The main one is that fortran never
logged the sanity checks to the Meson debug log, making it hard to
debug.
There's also some interesting quirks we built up in the dedicated
fortran handling. For example:
- in commit 5b109c9ad2 we added cwd to
building the fortran executable, with a wordy comment about how the
compiler has defects. But the clike base has always done that on
general principle anyway, so we would never have had that bug in the
first place.
- in commit d6be7822a0 we added special
deletion of an old "bad existing exe file" just for fortran. Looking
at the PR discussion for this odd requirement, it turns out that the
real problem is mixing WSL and native Windows without deleting the
build directory. This is apparently fortran specific simply because
"contemporary Windows 10 Fortran users" switch between the two?
The actual problem is that this never used .exe as the output name, so
Windows thinks you want to run something other than the thing you
asked to run, because it's not even a Window executable. But... the
common clike handling could have fixed that without needing special
cases.
Xavier Claessens
Jussi Pakkanen
Tristan Partin
Eli Schwartz
Jan Tojnar
Christoph Reiter
Hagen Möbius
Dylan Baker
Andres Freund
Nirbheek Chauhan
Caleb Cornett
Paolo Borelli