On macOS, we set the install_name for built libraries to
@rpath/libfoo.dylib, and when linking to the library, we set the RPATH
to its path in the build directory. This allows all built binaries to
be run as-is from the build directory (uninstalled).
However, on install, we have to strip all the RPATHs because they
point to the build directory, and we change the install_name of all
built libraries to the absolute path to the library. This causes the
install name in binaries to be out of date.
We now change that install name to point to the absolute path to each
built library after installation.
Fixes https://github.com/mesonbuild/meson/issues/3038
Fixes https://github.com/mesonbuild/meson/issues/3077
With this, the default workflow on macOS matches what everyone seems
to do, including Autotools and CMake. The next step is providing a way
for build files to override the install_name that is used after
installation for use with, f.ex., private libraries when combined with
the install_rpath: kwarg on targets.
The install name is used by consumers of the library to find the
library at runtime. If it's @rpath/libfoo.dylib, all consumers must
manually add the library path to RPATH, which is not what people
expect.
Almost everyone sets the library install name as the full path to the
library, and this is done at install time with install_name_tool.
This is a special type of option to be passed to most 'required' keyword
arguments. It adds a 3rd state to the traditional boolean value to cause
those methods to always return not-found even if the dependency could be
found.
Since integrators doesn't want enabled features to be a surprise there
is a global option "auto_features" to enable or disable all
automatic features.
When a test fails due to a signal (e.g., SIGSEGV) it can be somewhat
mysterious why the test failed. Also, even when a test fails due to a
non-zero exit status it would help if the exit status was reported. This
augments the result string to include the non-zero exit status or
signal number and name.
Resolves#3642
Added method concatenate_string_literals to CCompiler. Will concatenate
string literals.
Added keyword argument 'concatenate_string_literals' to Compiler.get_define.
If used will apply concatenate_string_literals to its return value.
Because vala is not listed in clike_langs, is_source(fname) is returning False
for Vala source files. Therefore, extract_all_objects() is completely empty
for Vala programs.
Fixes#791
Input files can be in any file encoding, not just utf-8 or isolatin1. Meson
should not make assumptions here and allow for the user to specify the
encoding to use.
e.g. 'meson x86_64-w64-mingw32 --cross-file x86_64-w64-mingw32' currently
fails with an IsADirectoryError exception.
Cross files must be files, so when searching, only accept a candidate path
which is an existing file, not just an existing path.
Otherwise we can end up searching for the same library tens of times,
because pkg-config does not de-duplicate -lfoo args before returning
them.
We use -Wl,--start-group/end-group, so we do not need to worry about
ordering issues in static libraries.
On Windows, if we are going to link with a shared module, we need the
implib.
Use case: The Xorg server builds some X protocol extensions as modules. The
implibs for these modules need to be shipped as part of the SDK, to enable
building of 3rd party extensions which reference symbols in (and hence on
Windows, need to be linked with) these modules.
Refine #3277
According to what I read on the internet, on OSX, both MH_BUNDLE (module)
and MH_DYLIB (shared library) can be dynamically loaded using dlopen(), but
it is not possible to link against MH_BUNDLE as if they were shared
libraries.
Metion this as an issue in the documentation.
Emitting a warning, and then going on to fail during the build with
mysterious errors in symbolextractor isn't very helpful, so make attempting
this an error on OSX.
Add a test for that.
See also:
https://docstore.mik.ua/orelly/unix3/mac/ch05_03.htmhttps://stackoverflow.com/questions/2339679/what-are-the-differences-between-so-and-dylib-on-osx
debugoptimized builds building against Qt would ultimately link against
both the debug and non-debug msvcrt, ntdll, etc libraries which causes
crashes in weird places and is very much not recommended by Microsoft.
This changes the selected Qt library(ies) correctly to not uses the
debug variants for debugoptimized builds.
https://github.com/mesonbuild/meson/pull/3680
Without this, building a module in a Flatpak app manifest that is a
newer version of a module already present in the Flatpak runtime will
fail. (The Flatpak file system is a bunch of hard links to readonly
files, which can be replaced but not written to.)
This instead creates a temporary file in the same directory as the
destination (to avoid cross-device renaming errors) and atomically
renames the temporary file to the destination, replacing it instead of
rewriting it as shutil.copyfile() would do.
If only 1 dir is provided, the 2nd defaults to '.' and if none is
provided they default to '.' and '..'. It should be builddir first,
followed by sourcedir, but validate_core_dirs() will still swap them if
builddir contains a meson.build file.
This simplifies a lot of code, and centralize "key=value" parsing in a
single place.
Unknown command line options becomes an hard error instead of
merely printing warning message. It has been warning it would become an
hard error for a while now. This has exceptions though, any
unknown option starting with "<lang>_" or "b_" are ignored because they
depend on which languages gets added and which compiler gets selected.
Also any option for unknown subproject are ignored because they depend
on which subproject actually gets built.
Also write more command line parsing tests. "19 bad command line
options" is removed because bad cmd line option became hard error and
it's covered with new tests in "30 command line".
All options are now the projectoptions list, regardless of how they got
defined in the command line.
This also delays setting builtin option values until the main project()
default options are parsed to simplify the code. This is possible
because we already delayed setting the backend after parsing main
project() in a previous commit.
This ensure all option groups are printed the same way. Also ensure that
we cannot miss some builtin options by taking the list of all builtin
options and excluding only directories/testing options.
Christoph Behle
Nirbheek Chauhan
Xavier Claessens
FFY00
Kurtis Rader
Jussi Pakkanen
Matthew Waters
Niklas Claesson
Paolo Bonzini
Sander Sweers
Jon Turney
Robert Ancell
Thibault Saunier
Philip Chimento