| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
- Install pcre2, as it's a new GLib dependency.
- Do not use CFLAGS to inject -Werror
- Use idiomatic Meson subcommands
|
|
|
|
| |
We need to unblock the CI pipeline, in order to fix them.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Fedora 34 is EOL, and by the time we release GNOME 43, Fedora 35 will be
close to EOL too.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
MSYS2 recently updated Python to 3.9.10 which triggered a build
issued present in typed_ast, which was fixed in typed_ast 1.5.0:
https://github.com/python/typed_ast/issues/169
The currently used mypy version has an upper limit on typed_ast, so
this fixed version wasn't pulled in.
To fix this, update mypy, fix one new warning it complains about
(namedtuple not being named after the variable), and install the
markdown stubs explicitely, since mypy no longer bundles them.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
The pcre subproject in GLib was renamed, so we need to fix the
incantation that promotes the subproject in our own CI.
|
|
|
|
|
| |
GLib has dropped the internal copy libpcre, so we need to promote
libpcre's subproject to the top level.
|
|
|
|
|
|
|
| |
GLib dropped the internal libpcre copy, which means we end up with a
subproject inside a subproject.
Since libpcre is frozen, we can rely on using the installed copy.
|
|
|
|
| |
Don't install them as part of the CI job. It's pointless repetition.
|
| |
|
|
|
|
| |
Not every Linux distro ships with Docker, and Podman is nicer.
|
| |
|
|
|
|
| |
Fedora 29 was EOL'ed in 2019.
|
|
|
|
| |
This resulted in duplicate module errors for me.
|
|
|
|
|
| |
ccache is currently broken somehow. With 32bit Windows getting less and less
usage/testing it's a good idea to move to 64 bit anyway.
|
|
|
|
|
| |
The latest release (0.800) finds a handful of new problems that need
to be addressed first.
|
|
|
|
| |
3.5 is EOL now and we'd like to use newer typing features
|
|
|
|
|
|
| |
Silence some errors, run mypy in CI
Adding annotations to functions/classes will make mypy check them.
|
|
|
|
|
|
|
| |
This tests that after installing the g-i tools a simple project using
them can be build with meson and autotools.
Fixes #341
|
|
|
|
| |
powershell instead of cmd, win2016 instead of 2012r2
|
|
|
|
|
|
| |
This reverts commit d01a387475e2665275a6cd6b0059bf91121b8300.
The nightly run time does not have Meson 0.54
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When gobject-introspection-1.0 pkg-config is not found on the system,
Meson can fallback to configure g-i as subproject and needs a dependency
object to replace the pc file.
The dependency file needs to ensure that typelibs are created before
compiling any other gir and provide the girdir for files within build
directory. It also need to provide glib dependencies required to compile
girs.
Bump Meson version to 0.54.0 to use meson.override_dependency().
|
|
|
|
|
|
|
| |
In our case we have to do various special things in case glib isn't available
through pkg-config and not all tests are run.
Add a CI job that tests this case so we don't regress.
|
|
|
|
| |
See !186
|
|
|
|
|
| |
The different dir separator seems to confuse meson now
(likely a new change in mingw Python path handling..)
|
|
|
|
|
|
|
| |
Similar to !180 this should prevent devs from not running all tests by
accident.
This also adds some checks for the required doctool dependencies, mako and markdown.
|
|
|
|
|
|
|
|
|
| |
We require cairo only to run all tests and thus default it to false.
This usually results in developers not using it when working on changes and
tests depending on cairo then failing in CI.
This changes it to a feature option that defaults to auto, so that devs that
have cairo headers installed will automatically use it.
|
|
|
|
|
| |
It's what we use for testing in CI (so not even sure older versions work) and
glib master now also depends on this 0.49.
|
| |
|
| |
|
| |
|
|
|
|
| |
To unbreak continuous
|
| |
|
| |
|
|
|
|
|
| |
The cairo dependency fallback code is copied from pygobject. I assume that
will work for msvc users (cairo is disabled by default anyway)
|
|
|
|
|
| |
The headerid extension no longer exists in 3.x. Use the toc extension instead.
The toc_depth option was only added in 3.x so still try headerid first.
|
|
|
|
|
| |
Add a simple test using the docwriter so we trigger the docwriter related
Python imports.
|
|
|
|
|
| |
Getting rid of autotools wont be easy, so make sure we run all tests on Windows,
not just the ones ported to meson.
|
| |
|
|
|
|
|
|
| |
glib did the same thing this cycle, see
https://gitlab.gnome.org/GNOME/glib/merge_requests/196
for details
|
|
|
|
|
| |
GLib, a wrapped dependency, requires 0.48.0, so our CI environments should
have that installed.
|
|
|
|
|
|
|
|
| |
doctool is an auto option and was disabled by the recent addition of
a new dependency on the markdown Python module. Since we didn't have that
in our testing docker image the tests were just skipped.
This makes sure we error out if any dependencies are missing again.
|
|
|
|
|
| |
This prints the error count to stderr, and more importantly 0 when
everything is OK so we know it did something.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
g-i includes an old version of pep8 and pyflakes and uses that
during "make check".
It (1) doesn't catch all cases newer versions of pycodestyle/pyflakes catch
and (2) doesn't test all Python files (3) doesn't work with meson.
Instead of updating just remove them and depend on flake8 instead.
To run the checks simply run flake8 in the root dir.
This also makes it possible to run those checks when using meson and
not autotools.
To not get test suite failures on flake8 updates move the checks from
"make check" to an extra "make check.quality" target.
|