[ANN] CplusRuby - Gluing C and Ruby

[Michael Neumann]

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

----------------------------------------------
CplusRuby - Glue C and Ruby together (OO-like)
----------------------------------------------

COPYRIGHT

Copyright (c) 2007 by Michael Neumann ([email protected]).
All rights reserved.

LICENSE

Ruby License.

ABOUT

With CplusRuby you can define custom C structures from within Ruby and
wrap them easily in a Ruby class. You can as well define C functions
that can access this structure easily. CplusRuby generates setter/getter
methods for every property and wrapper methods for the C functions.

The purpose is speed! The C functions can access the C-structure,
which is much faster than accessing instance variables. Also, the C
functions can call each other directly. This is *much* faster than
invoking a method in Ruby. As wrappers are generated, the Ruby-side
can access all C-structures and functions as well.

I started to write CplusRuby to implement a high-performance pulsed
neural network simulator. My existing C++ implementation suffered
from - well - C++ . This enables me to write the core algorithms
in C or C++ and do all the other non performance-critical tasks in
Ruby.

EXAMPLE

Take a look at the following example. You should also take a look
at the generated C source file (inspire.c). Note that properties
are actually members of a C-struct, not instance variables, and as
such, their access from C is very fast. As calling a method is quite
slow in Ruby, method defined in C (method_c) can be called directly
from C, which again is very fast!

# example.rb
require 'cplusruby'

class NeuralEntity < CplusRuby
property :id
end

class Neuron < NeuralEntity
property otential, :float
property :last_spike_time, :float
property re_synapses, :value

method_c :stimulate, %(float at, float weight), %{
// this is C code
selfc->potential += at*weight;
}

def initialize
self.pre_synapses = []
end
end

# generate C file, compile it and load the .so
CplusRuby.evaluate("inspire.cc", "-O3", "-lstdc++")

if __FILE__ == $0
n = Neuron.new
n.id = "n1"
n.potential = 1.0
n.stimulate(1.0, 2.0)
p n.potential # => 3.0
end

END

 

[Daniel Berger]

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

<snip>

First patch!

C:\Documents and Settings\djberge\My Documents\My Downloads\Ruby>diff -
u cplusruby.orig cplusruby.rb
--- cplusruby.orig Fri Sep 21 15:50:38 2007
+++ cplusruby.rb Fri Sep 21 15:49:04 2007
@@ -11,6 +11,10 @@
#
# Implements a simple ordered Hash
#
+require 'rbconfig'
+require 'win32/process' if RUBY_PLATFORM.match('mswin')
+include Config
+
class OHash < Hash
def []=(k, v)
@order ||= []
@@ -146,20 +150,28 @@
dir = File.dirname(file)
mod, ext = base.split(".")

+ make = RUBY_PLATFORM.match('mswin') ? 'nmake' : 'make'
+
File.open(file, 'w+') {|f| f << self.generate_code(mod) }
Dir.chdir(dir) do
- system("make clean")
+ system("#{make} clean")
+
pid = fork do
require 'mkmf'
$CFLAGS = cflags
$LIBS << (" " + libs)
create_makefile(mod)
- exec "make"
+ exec "#{make}"
end
_, status = Process.waitpid2(pid)
- raise if status.exitstatus != 0
+
+ if RUBY_PLATFORM.match('mswin')
+ raise if status != 0
+ else
+ raise if status.exitstatus != 0
+ end
end
- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end

def self.generate_code_for_class

Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).

Regards,

Dan

 

[Konrad Meyer]

--nextPart1419151.1mao5Msloi
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Quoth Daniel Berger:

=20

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

=20
<snip>
=20
First patch!
=20
C:\Documents and Settings\djberge\My Documents\My Downloads\Ruby>diff -
u cplusruby.orig cplusruby.rb
--- cplusruby.orig Fri Sep 21 15:50:38 2007
+++ cplusruby.rb Fri Sep 21 15:49:04 2007
@@ -11,6 +11,10 @@
#
# Implements a simple ordered Hash
#
+require 'rbconfig'
+require 'win32/process' if RUBY_PLATFORM.match('mswin')
+include Config
+
class OHash < Hash
def []=3D(k, v)
@order ||=3D []
@@ -146,20 +150,28 @@
dir =3D File.dirname(file)
mod, ext =3D base.split(".")
=20
+ make =3D RUBY_PLATFORM.match('mswin') ? 'nmake' : 'make'
+
File.open(file, 'w+') {|f| f << self.generate_code(mod) }
Dir.chdir(dir) do
- system("make clean")
+ system("#{make} clean")
+
pid =3D fork do
require 'mkmf'
$CFLAGS =3D cflags
$LIBS << (" " + libs)
create_makefile(mod)
- exec "make"
+ exec "#{make}"
end
_, status =3D Process.waitpid2(pid)
- raise if status.exitstatus !=3D 0
+
+ if RUBY_PLATFORM.match('mswin')
+ raise if status !=3D 0
+ else
+ raise if status.exitstatus !=3D 0
+ end
end
- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end
=20
def self.generate_code_for_class
=20
Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).
=20
Regards,
=20
Dan

Does the C->ruby mapping map between char * and Strings?

Thanks,
=2D-=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

--nextPart1419151.1mao5Msloi
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQBG9D+0CHB0oCiR2cwRAm5fAJ4vJodjmvfVBltXxH2lUYm9EZpeSgCfQkW0
hBrZuB6OLZOfTkQINW3cfMY=
=1i3A
-----END PGP SIGNATURE-----

--nextPart1419151.1mao5Msloi--

 

[Konrad Meyer]

--nextPart66148048.BTmV49FagV
Content-Type: text/plain;
charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Quoth Konrad Meyer:

Quoth Daniel Berger:

=20

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-r=

uby
=20
<snip>
=20
First patch!
=20
C:\Documents and Settings\djberge\My Documents\My Downloads\Ruby>diff -
u cplusruby.orig cplusruby.rb
--- cplusruby.orig Fri Sep 21 15:50:38 2007
+++ cplusruby.rb Fri Sep 21 15:49:04 2007
@@ -11,6 +11,10 @@
#
# Implements a simple ordered Hash
#
+require 'rbconfig'
+require 'win32/process' if RUBY_PLATFORM.match('mswin')
+include Config
+
class OHash < Hash
def []=3D(k, v)
@order ||=3D []
@@ -146,20 +150,28 @@
dir =3D File.dirname(file)
mod, ext =3D base.split(".")
=20
+ make =3D RUBY_PLATFORM.match('mswin') ? 'nmake' : 'make'
+
File.open(file, 'w+') {|f| f << self.generate_code(mod) }
Dir.chdir(dir) do
- system("make clean")
+ system("#{make} clean")
+
pid =3D fork do
require 'mkmf'
$CFLAGS =3D cflags
$LIBS << (" " + libs)
create_makefile(mod)
- exec "make"
+ exec "#{make}"
end
_, status =3D Process.waitpid2(pid)
- raise if status.exitstatus !=3D 0
+
+ if RUBY_PLATFORM.match('mswin')
+ raise if status !=3D 0
+ else
+ raise if status.exitstatus !=3D 0
+ end
end
- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end
=20
def self.generate_code_for_class
=20
Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).
=20
Regards,
=20
Dan

=20
Does the C->ruby mapping map between char * and Strings?
=20
Thanks,
--=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

Okay, so I have added the following to DEFAULT_OPTIONS:

:int =3D> {
:default =3D> '0',
:ctype =3D> 'int %s',
:ruby2c =3D> '(int)NUM2INT(%s)',
:c2ruby =3D> 'INT2NUM(%s)'},

:char_p =3D> {
:default =3D> '""',
:ctype =3D> 'char *%s',
:ruby2c =3D> 'StringValuePtr(%s)',
:c2ruby =3D> 'rb_str_new2(%s)'}=20

My code is like this:

#---------------------start

require 'cplusruby'

class BankAccount < CplusRuby
property :balance, :int
property :name, :char_p

method_c :withdraw, %{int dollars}, %{
if (dollars <=3D selfc->balance){
selfc->balance -=3D dollars;
return true;
}
}

def initialize
end
end

CplusRuby.evaluate("foo.c", "-Os", "")

n =3D BankAccount.new
puts n.balance
puts n.name

#---------------------end

When I run it with ruby I get a whole bunch of errors stemming from
the fact that it generates C code that uses a type called
"BankAccount" without defining it anywhere (struct BankAccount is
defined though).

Ideas, suggestions, questions?

Thanks,
=2D-=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

--nextPart66148048.BTmV49FagV
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQBG9Fb6CHB0oCiR2cwRAjEHAKDIVdHFl0cWZnSd25iAtz7WDOhMjgCgmDo9
tgGbQ8qP7q5RtCQ7R6VjMqk=
=PjjL
-----END PGP SIGNATURE-----

--nextPart66148048.BTmV49FagV--

 

[Konrad Meyer]

--nextPart2750306.cXgpGYuoH4
Content-Type: text/plain;
charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Quoth Konrad Meyer:

Quoth Daniel Berger:

=20

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-r=

uby
=20
<snip>
=20
First patch!
=20
C:\Documents and Settings\djberge\My Documents\My Downloads\Ruby>diff -
u cplusruby.orig cplusruby.rb
--- cplusruby.orig Fri Sep 21 15:50:38 2007
+++ cplusruby.rb Fri Sep 21 15:49:04 2007
@@ -11,6 +11,10 @@
#
# Implements a simple ordered Hash
#
+require 'rbconfig'
+require 'win32/process' if RUBY_PLATFORM.match('mswin')
+include Config
+
class OHash < Hash
def []=3D(k, v)
@order ||=3D []
@@ -146,20 +150,28 @@
dir =3D File.dirname(file)
mod, ext =3D base.split(".")
=20
+ make =3D RUBY_PLATFORM.match('mswin') ? 'nmake' : 'make'
+
File.open(file, 'w+') {|f| f << self.generate_code(mod) }
Dir.chdir(dir) do
- system("make clean")
+ system("#{make} clean")
+
pid =3D fork do
require 'mkmf'
$CFLAGS =3D cflags
$LIBS << (" " + libs)
create_makefile(mod)
- exec "make"
+ exec "#{make}"
end
_, status =3D Process.waitpid2(pid)
- raise if status.exitstatus !=3D 0
+
+ if RUBY_PLATFORM.match('mswin')
+ raise if status !=3D 0
+ else
+ raise if status.exitstatus !=3D 0
+ end
end
- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end
=20
def self.generate_code_for_class
=20
Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).
=20
Regards,
=20
Dan

=20
Does the C->ruby mapping map between char * and Strings?
=20
Thanks,
--=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

Also, is there a way to define C functions from ruby that return a C type?

=2D-=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

--nextPart2750306.cXgpGYuoH4
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQBG9GMECHB0oCiR2cwRAp9GAJ9kJazwIyQ7vo6cIFXoffLdbeskqACfeqIX
WZeyve3R/k3nvJ5C7vrW2vU=
=UFRT
-----END PGP SIGNATURE-----

--nextPart2750306.cXgpGYuoH4--

 

[Nobuyoshi Nakada]

Hi,

At Sat, 22 Sep 2007 06:54:13 +0900,
Daniel Berger wrote in [ruby-talk:270261]:

+ if RUBY_PLATFORM.match('mswin')
+ raise if status != 0
+ else
+ raise if status.exitstatus != 0
+ end

I'm curious why win32/process doesn't return Process::Status.

 

[Daniel Berger]

Hi,

At Sat, 22 Sep 2007 06:54:13 +0900,
Daniel Berger wrote in [ruby-talk:270261]:

+ if RUBY_PLATFORM.match('mswin')
+ raise if status != 0
+ else
+ raise if status.exitstatus != 0
+ end

I'm curious why win32/process doesn't return Process::Status.

We had to reimplement waitpid and waitpid2 because of our custom fork
implementation. But, we weren't sure how to set $? manually to a
Process::Status object in pure Ruby. Do you have any suggestions?

Thanks,

Dan

 

[Joel VanderWerf]

Hi all,

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

It's been done, but I'm glad to see it done again and probably with new
ideas

For comparison:

require 'cgen/cshadow' # from cgen on RAA

class NeuralEntity
include CShadow
attr_accessor :id
end

class Neuron < NeuralEntity
shadow_attr_accessor otential => "double potential"
shadow_attr_accessor :last_spike_time => "double last_spike_time"
shadow_attr_accessor re_synapses => Object

define_c_method :stimulate do
arguments :at, :weight
body %{
shadow->potential += NUM2DBL(at)*NUM2DBL(weight);
}
end

def initialize
self.pre_synapses = []
end
end

NeuralEntity.commit # generate and load the shared lib

if __FILE__ == $0
n = Neuron.new
n.id = "n1"
n.potential = 1.0
n.stimulate(1.0, 2.0)
p n.potential # => 3.0
end

Cgen is middle aged, as ruby libraries go (written mostly in 2001), and
it probably isn't up to community standards in a number of ways. Still,
I've used it intensively in numerical and hybrid simulation since then.
It's the foundation for redshift, which is a DSL for simulation of
hybrid automata (networked state machines with continuous as well as
discrete dynamics).

Cgen/cshadow handles a number of issues for you:

- creating accessors (r, w, rw)
- type checking and conversion in the write accessor
- mark and free functions
- attr initialization
- persistence via marshal and yaml (or an attr may be specified
non-persistent)
- extensible attribute classes for defining new attr types, other than
the basic numeric, string, and Ruby object types
- inheritance (a subclass of a shadow class inherits the parent's shadow
attrs in its own C struct; struct may be safely cast to parent type)
- tools for managing how code is distributed among files (very important
with a large project, or else recompilation time gets painful).
- tools for managing declarations and preprocessor includes
- DSL for rb_scan_args()

This is pretty well documented, though it is in the old rd format,
rather than rdoc.

 

[Michael Neumann]

It's been done, but I'm glad to see it done again and probably with new
ideas

Wow, cool! I knew about cgen, but then didn't knew that it's that easy
to use.

In the example you give I only see one minor difference. In CplusRuby,
method :stimulate can be called directly from the C-side, without
NUM2DBL conversions of Ruby-types. That's very important for my
usage scenario.
But I think you can do the same with cgen.

Regards,

Michael

 

[Joel VanderWerf]

In the example you give I only see one minor difference. In CplusRuby,
method :stimulate can be called directly from the C-side, without
NUM2DBL conversions of Ruby-types. That's very important for my
usage scenario.
But I think you can do the same with cgen.

Actually, I don't think cgen has that feature.... one of the things I'll
have to er, borrow, from cplusruby. Currently the only way to do this is
to define a separate C function that the method calls.

Reading the cplusruby .c output, it looks like there is an outer
function (Neuron_call__stimulate) that accepts ruby value arguments, and
then converts them and calls an inner function (Neuron__stimulate),
which expects C-typed arguments. You could do this in cgen by using
define_c_method to define the outer function and define_c_function to
define the inner function.

I guess I was used to performing the conversion manually, so I never
added that feature as a convenience. I can see that it would be more of
a necessity than a convenience if the function needed to be called from
C as well as ruby.

The selfc->stimulate function pointer is another difference, and
necessary for the same reason, I guess. In a cshadow struct, this struct
member doesn't exist, but there is reference ("VALUE self;") back to the
original object, on which you can call methods (with the additional
friction of type conversions, if the caller is native C code).

The fact that you have pointers to functions that accept C-typed args
makes access from C easy and efficient, but it also uses a lot of space
in the struct with a pointer to each function that implements a method.
What about having a unique struct for each class that keeps these
pointers, and then the struct only needs a pointer to this "singleton"?

 

[Joel VanderWerf]

- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end

def self.generate_code_for_class

Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).

Why not just

+ require "#{dir}/#{mod}"

which should work if there are no stray .rb files in the dir?

 

[Joel VanderWerf]

One comment on inheritance in cplusruby:

def self.inherited(klass)
# inherit properties and c_methods
self.properties.each {|k,v| klass.properties[k] = v}
CplusRuby.registered_classes << klass
end

Re-opening a parent class after a child class has been defined and
adding more properties will not propagate them to the child. And you
can't simply add such properties to the end of the parent and child
struct definitions, since that will make them incompatible (in the sense
of casting).

 

[Eric Hodel]

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-
ruby

[...]
system("#{make} clean")

Why not use RubyInline?

You'll get more than twice the utility and drop a bunch of code.

 

[Michael Neumann]

One comment on inheritance in cplusruby:

def self.inherited(klass)
# inherit properties and c_methods
self.properties.each {|k,v| klass.properties[k] = v}
CplusRuby.registered_classes << klass
end

Re-opening a parent class after a child class has been defined and
adding more properties will not propagate them to the child. And you
can't simply add such properties to the end of the parent and child
struct definitions, since that will make them incompatible (in the
sense of casting).

Ah thanks. Well, I should definitively refactor cplusruby a bit .

A quick C question about casting:

struct A {
int a;
}

struct B {
int a;
int b;
}

Is it always safe (even with -O3) to cast a pointer of B to A?
Or do I have to do:

struct B {
A x;
int b;
}

which is very ugly IMHO. Are there pramgas like "__packed" for this?

Actually, I don't think cgen has that feature.... one of the things I'll
have to er, borrow, from cplusruby. Currently the only way to do this is
to define a separate C function that the method calls.

Reading the cplusruby .c output, it looks like there is an outer
function (Neuron_call__stimulate) that accepts ruby value arguments, and
then converts them and calls an inner function (Neuron__stimulate),
which expects C-typed arguments. You could do this in cgen by using
define_c_method to define the outer function and define_c_function to
define the inner function.

I guess I was used to performing the conversion manually, so I never
added that feature as a convenience. I can see that it would be more of
a necessity than a convenience if the function needed to be called from
C as well as ruby.

The selfc->stimulate function pointer is another difference, and
necessary for the same reason, I guess. In a cshadow struct, this struct
member doesn't exist, but there is reference ("VALUE self;") back to the
original object, on which you can call methods (with the additional
friction of type conversions, if the caller is native C code).

The "VALUE self" in the C structure is a feature I will borrow. It's a
great idea!

The fact that you have pointers to functions that accept C-typed args
makes access from C easy and efficient, but it also uses a lot of space
in the struct with a pointer to each function that implements a method.
What about having a unique struct for each class that keeps these
pointers, and then the struct only needs a pointer to this "singleton"?

I plan to make methods by default "non-virtual" in the C++ sense. That
is, they will by default not be included as function pointers in the
struct. If you want to have virtual methods, you'd then need to add:

property :a_virtual_method, :virtual_method

I also plan to support templates, which is a feature I really need for
some data types like priority queues (I noticed that using templates
they are a lot faster than a generic version):

class ImplicitBinaryHeap
include CplusRuby::Template

property :size, :int
property :capacity, :int
property :elements, "%ELEMENT_TYPE%*"

method :initialize do
arguments [:initial_capacity, :int]
body %{
@capacity = initial_capacity;
@size = 0;
@elements = ALLOC_N(%ELEMENT_TYPE%, @capacity+1);
}
end

method ush do
arguments [:element, "%ELEMENT_TYPE%*"]
body %{
@size += 1;
if (@size > @capacity) $resize(self, 2*@capacity + 1);
@elements[@size] = *element;
$propagate_up(self, @size);
}
end

#...
end

MyPQ = ImplicitBinaryHeap.bindELEMENT_TYPE => 'int)

# now MyPQ is similar ImplicitBinaryHeap<int>

self is now a pointer to a C struct (thanks to the "VALUE self"
reference). "@" is replaced by "self->", and $ is replaced by
"classname__", so that you don't have to hardcode the prefix of methods
(only works for non-virtual methods of course).

The good thing is, that you can use the structure without associating it
to a Ruby object (so you get a poor mans or not so poor mans "C with
classes"):

MyPQ mypq;
MyPQ__initialize(&mypq, 31);
for (int i = 0; i<10; i++)
{
MyPQ__push(&mypq, &i);
}

It's not that nice as the equivalent in C++, but it doesn't need a C++
compiler at all.

Regards,

Michael

 

[Michael Neumann]

I am proud to announce my latest project called CplusRuby.
Below I pasted it's README for further explanation.
Alternatively read the following blog article:

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

[...]
system("#{make} clean")

Why not use RubyInline?

You'll get more than twice the utility and drop a bunch of code.

Drop less than 20 lines of code? It really takes longer to get into
RubyInline than to write those 20 lines

Regards,

Michael

 

[Paul Brannan]

:char_p => {
:default => '""',
:ctype => 'char* %s',
:ruby2c => 'StringValuePtr(%s)',
:c2ruby => 'rb_str_new2(%s)'}

I think you'll want to store either a copy of the string or a reference
to the string object (which would need to be marked). If you don't, the
string could be destroyed before the containing struct, which would
leave you with a dangling pointer.

Paul

 

[Konrad Meyer]

--nextPart2647755.kQulo5Tmft
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Quoth Paul Brannan:

:char_p =3D> {
:default =3D> '""',
:ctype =3D> 'char* %s',
:ruby2c =3D> 'StringValuePtr(%s)',
:c2ruby =3D> 'rb_str_new2(%s)'}
=20
I think you'll want to store either a copy of the string or a reference
to the string object (which would need to be marked). If you don't, the
string could be destroyed before the containing struct, which would
leave you with a dangling pointer.
=20
Paul

Yes, I believe that's the problem I ran into.

=2D-=20
Konrad Meyer <[email protected]> http://konrad.sobertillnoon.com/

--nextPart2647755.kQulo5Tmft
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQBG9vb6CHB0oCiR2cwRAs2sAJwN0CpmhmDI2HiaQqqT4UZ5Gi3w8gCgrpHz
2T/+aWIC/h2edh8CBjaiMr8=
=IId5
-----END PGP SIGNATURE-----

--nextPart2647755.kQulo5Tmft--

 

[Gaspard Bucher]

http://www.ntecs.de/blog/articles/2007/09/21/cplusruby-gluing-c-and-ruby

Thanks for the announce. I will keep an eye on your project when it
comes to speed with Rubyk. By the way, did anyone play with
RubyInline+Ruby2c like in the example below ?
require 'inline'
class MyTest

def factorial(n)
f = 1
n.downto(2) { |x| f *= x }
f
end

inlineRuby) do |builder|
builder.optimize :factorial
end
end

Gaspard

 

[Nobuyoshi Nakada]

Hi,

At Sun, 23 Sep 2007 06:50:05 +0900,
Joel VanderWerf wrote in [ruby-talk:270374]:

- require "#{dir}/#{mod}.so"
+ require "#{dir}/#{mod}." + CONFIG['DLEXT']
end

def self.generate_code_for_class

Mostly fixes MS Windows issues, but the DLEXT bit also fixes platforms
that don't generate .so files (OS X, HP-UX).

Why not just

+ require "#{dir}/#{mod}"

which should work if there are no stray .rb files in the dir?

Actually, CONFIG['DLEXT'] is unnecessary at all and that code
works fine with ".so" on every platforms, since the extension
".so" will be substituted with DLEXT internally.