We were checking for builtins explicitly like this because the ordinary
checks don't work for builtins at all. We do exactly the same check as
Autoconf and it doesn't work with Autoconf either (Autoconf is broken!)
So now we check for it in two ways: if there's no #include in prefix, we
check if `__builtin_symbol` exists (has_function allows checking for
functions without providing includes). If there's a #include, we check
if `symbol` exists.
The old method was causing problems with some buggy toolchains such as
MSYS2 which define some builtins in the C library but don't expose them
via headers which meant that `__builtin_symbol` would be found even
though `symbol` is not available.
Doing this allows people to always get the correct answer as long as
they specify the includes that are required to find a function while
also not forcing people to always specify includes to find a function
which is cumbersome.
Closes#1083
On Windows, we can build with both 32-bit and 64-bit compilers, but the
Python is either 32-bit or 64-bit. Check the architecture of the found
Python libraries and don't use them if they don't match our
build_machine.
Also skip the tests if the Python 3 dependency is not found.
Unlike Linux and OS X, when a library is loaded, all the symbols aren't
loaded into a single namespace. You must fetch the symbol by iterating over
all loaded modules.
So, we shouldn't use /FORCE:UNRESOLVED since that is not what modules do
on Windows. Instead, we now do exactly what GModule does on Windows.
Also use `void` for functions that take no arguments.
This is already how it should've been, but:
a) The test for this was wrong since Dependency is a base class for all
dependencies and isinstance on an InternalDependency will also be true
b) Internal dependencies can't ever be used here anyway because compiler
checks are always run at configure time and internal dependencies are
only built after that.
Just checking that the version retrieved from the pkg-config file
matches is not enough. It's nearly tautological since it just checks
that we aren't returning garbage in dep.version(). Instead, check in the
test executable that the pkg-config version retrieved matches the
ZLIB_VERSION exported by zlib.
Also don't require zlib 1.2.8 since RHEL (EPEL) 7 still has 1.2.7 and we
don't really need 1.2.8 anyway.
Sometimes we want to restrict the acceptable versions to a list of
versions, or a smallest-version + largest-version, or both. For
instance, GStreamer's opencv plugin is only compatible with
3.1.0 >= opencv >= 2.3.0
We originally skipped on Windows for the same reason, and now we need to
do that for EPEL 7 and F23 which ship pkg-config 0.28 which doesn't have
--validate.
Closes#1129
We want compiler check arguments (-O0, -fpermissive, etc) to override
all other arguments, and we want extra_args passed in by the build file
to always override everything.
To do this properly, we must split include arguments out, append them
first, append all other arguments as usual, and then append the rest.
As part of this, we also add the compiler check flags to the
cc.compiles() and cc.links() helper functions since they also most
likely need them.
Also includes a unit test for all this.
Knowing whether a test failed to run as its prerequisites were not
available, or whether those prerequisites were available and produced
unexpected/incorrect results, is a useful differentiation.
Add support for skipped tests by testing for exit code 77, used through
autotools/piglit/etc to denote a test which detected this and decided to
skip.
The library from which we extract should have multiple sources
from which we choose a strict subset otherwise we're testing almost the
same case extracting all the objects.
This exposes a bug in how we were doing extract checking for unity
builds.
This defaults to not exporting resources as that is generally what
you want but that does make this a breaking change. Along with that
if you export your resources you would want to install the header.
MesonDep1 doesn't use symbols from MesonDep2, but uses the MesonDep2
structure definition (so it only needs the header). This means only
generate_gir needs a dependency on MesonDep2 and shared_library doesn't.
This was broken earlier.
MesonSample uses symbols from MesonDep1 and MesonDep2, so both the
library and the gir get a dependency on MesonDep1, and on MesonDep2
(transitively). The transitive dependency was broken earlier.
Move CCompiler.compile to Compiler.compile so that ValaCompiler can use
it. Also rewrite ValaCompiler.sanity_check to use it since it does
a simple compile check.
At the same time, it enhances ExternalLibrary to support arguments for
languages other than C-like.
Includes a test for this that links against zlib through Vala.
Closes#983
Instead of using a whitelist, use a blacklist. Also print a more useful
error if the regex fails to match.
Use an underscore in the gir test to trigger this.
Fixes#436
Since Vala require 'glib-2.0' and 'gobject-2.0' dependencies, it's
better to fail at 'valac' step with meaningful error.
Add missing 'gobject-2.0' dependency on the mixed source test case.
Instead of blindly searching in PATH, use
Qt5Dependency.compilers_detect() (same for qt4) to get moc/uic/rcc.
This is much more robust, and it improves the chances that the correct
ones will be found. We still manually verify for now because the
fallback in dependencies.py for searching is stll to look in PATH for
backwards-compat, and because people probably have setups like that.
Also sync the qt4 module with the qt5 module w.r.t. resource compilation
and make the compiled qrc.cpp file unique in terms of the framework
version used (4 vs 5). This is needed for the test to work properly,
which now covers both Qt4 and 5.
Elliott Sales de Andrade
Jussi Pakkanen
Nirbheek Chauhan
Patrick Griffis
Daniel Stone
Igor Gnatenko
Guillaume Poirier-Morency
Thibault Saunier