mirror of
https://github.com/ruby/ruby.git
synced 2026-01-28 04:54:23 +00:00
9ef51b0b89
The rb_define_method function takes a pointer to ANYARGS-ed functions, which in fact varies 18 different prototypes. We still need to preserve ANYARGS for storages but why not check the consistencies if possible. Q&As: Q: Where did the magic number "18" came from in the description above? A: Count the case branch of vm_method.c:call_cfunc_invoker_func(). Note also that the 18 branches has lasted for at least 25 years. See also 200e0ee2fd3c1c006c528874a88f684447215524. Q: What is this __weakref__ thing? A: That is a kind of function overloading mechanism that GCC provides. In this case for instance rb_define_method0 is an alias of rb_define_method, with a strong type. Q: What is this __transparent_union__ thing? A: That is another kind of function overloading mechanism that GCC provides. In this case the attributed function pointer is either VALUE(*)(int,VALUE*,VALUE) or VALUE(*)(int,const VALUE*,VALUE). This is better than void* or ANYARGS because we can reject all other possibilities than the two. Q: What does this rb_define_method macro mean? A: It selects appropriate alias of the rb_define_method function, depending on the arity. Q: Why the prototype change of rb_f_notimplement? A: Function pointer to rb_f_notimplement is special cased in vm_method.c:rb_add_method_cfunc(). That should be handled by the __builtin_choose_expr chain inside of rb_define_method macro expansion. In order to do so, comparison like (func == rb_f_notimplement) is inappropriate for __builtin_choose_expr's expression (which must be a compile-time integer constant but the address of rb_f_notimplement is not fixed until the linker). So instead we are using __builtin_types_compatible_p, and in doing so we need to distinguish rb_f_notimplement from others, by type.
# -*- rdoc -*-
= How to build ruby using Visual C++
== Requirement
1. Windows 7 or later.
2. Visual C++ 12.0 (2013) or later.
[Note] if you want to build x64 version, use native compiler for
x64.
3. Please set environment variable +INCLUDE+, +LIB+, +PATH+
to run required commands properly from the command line.
[Note] building ruby requires following commands.
* nmake
* cl
* ml
* lib
* dumpbin
4. If you want to build from GIT source, following commands are required.
* bison
* patch
* sed
* ruby 2.0 or later
5. Enable Command Extension of your command line. It's the default behavior
of +cmd.exe+. If you want to enable it explicitly, run +cmd.exe+ with
<tt>/E:ON</tt> option.
== How to compile and install
1. Execute <tt>win32\configure.bat</tt> on your build directory.
You can specify the target platform as an argument.
For example, run `<tt>configure --target=i686-mswin32</tt>'
You can also specify the install directory.
For example, run `<tt>configure --prefix=<install_directory></tt>'
Default of the install directory is <tt>/usr</tt> .
The default _PLATFORM_ is `+i386-mswin32_+_MSRTVERSION_' on 32-bit
platforms, or `+x64-mswin64_+_MSRTVERSION_' on x64 platforms.
_MSRTVERSION_ is the 2- or 3-digits version of the Microsoft
Runtime Library.
2. Change _RUBY_INSTALL_NAME_ and _RUBY_SO_NAME_ in +Makefile+
if you want to change the name of the executable files.
And add _RUBYW_INSTALL_NAME_ to change the name of the
executable without console window if also you want.
3. Run `<tt>nmake up</tt>' if you are building from GIT source.
4. Run `<tt>nmake</tt>'
5. Run `<tt>nmake check</tt>'
6. Run `<tt>nmake install</tt>'
== Icons
Any icon files(*.ico) in the build directory, directories specified with
_icondirs_ make variable and +win32+ directory under the ruby
source directory will be included in DLL or executable files, according
to their base names.
$(RUBY_INSTALL_NAME).ico or ruby.ico --> $(RUBY_INSTALL_NAME).exe
$(RUBYW_INSTALL_NAME).ico or rubyw.ico --> $(RUBYW_INSTALL_NAME).exe
the others --> $(RUBY_SO_NAME).dll
Although no icons are distributed with the ruby source, you can use
anything you like. You will be able to find many images by search engines.
For example, followings are made from {Ruby logo
kit}[https://cache.ruby-lang.org/pub/misc/logo/ruby-logo-kit.zip]:
* Small {favicon}[https://www.ruby-lang.org/favicon.ico] in the official
site
* http://ruby.morphball.net/vit-ruby-ico_en.html or
{icon itself}[http://ruby.morphball.net/icon/vit-ruby.ico]
== Build examples
* Build on the ruby source directory.
ex.)
ruby source directory: C:\ruby
build directory: C:\ruby
install directory: C:\usr\local
C:
cd \ruby
win32\configure --prefix=/usr/local
nmake
nmake check
nmake install
* Build on the relative directory from the ruby source directory.
ex.)
ruby source directory: C:\ruby
build directory: C:\ruby\mswin32
install directory: C:\usr\local
C:
cd \ruby
mkdir mswin32
cd mswin32
..\win32\configure --prefix=/usr/local
nmake
nmake check
nmake install
* Build on the different drive.
ex.)
ruby source directory: C:\src\ruby
build directory: D:\build\ruby
install directory: C:\usr\local
D:
cd D:\build\ruby
C:\src\ruby\win32\configure --prefix=/usr/local
nmake
nmake check
nmake install DESTDIR=C:
* Build x64 version (requires native x64 VC++ compiler)
ex.)
ruby source directory: C:\ruby
build directory: C:\ruby
install directory: C:\usr\local
C:
cd \ruby
win32\configure --prefix=/usr/local --target=x64-mswin64
nmake
nmake check
nmake install
== Bugs
You can *NOT* use a path name that contains any white space characters as
the ruby source directory, this restriction comes from the behavior of
<code>!INCLUDE</code> directives of +NMAKE+.
You can build ruby in any directory including the source directory,
except +win32+ directory in the source directory.
This is restriction originating in the path search method of +NMAKE+.