Re: Advocacy: Ruby on/with .net

In article <[e-mail]>, Mauricio
Fern疣dez wrote:
> On Wed, Jul 30, 2003 at 11:30:02AM +0900, Tim Sutherland wrote:
>> In article <007d01c35570$0d1d54b0$a30aa1c8@bambook>, Rodrigo B. de Oliveira
>> wrote:
>> > I feel exactly the same way. When do we start coding? :-)
>> [...]
>> 
>> There is a mailing list
>> 	http://sourceforge.net/mailarchive/forum.php?forum_id=34312
>> for discussing all things to do with Ruby and .NET.
>> 
>> I have code right now which allows you to do things like
>> 	require 'dotnet'
>> 
>> 	DotNet.System.Console.WriteLine("Hello DotNet World!")
>> 
>> 	foo = DotNet.System.Collections.Hashtable.new
>> 	foo['one'] = 1
>> 	puts(foo['one']) # -> 1
>> 
>> It also allows C# to call Ruby code.
> 
> This is *very* similar in spirit to what I'm doing w/ Java (through
> JNI). If I got it right (I did a quick inspection of your code), you're
> linking to the C# runtime, so you have both parts running on the same
> process at a time (as opposed to communicating through sockets for
> instance).

Yes, that is how I am doing it. NB: I'm planning to make the first
'real' release of my rubydotnetproxy library by the upcoming Monday.

> BTW, you have achieved about the same (if not more) in < 2Klocs than me
> in 8Kloc (although some 4Kloc are auto-generated).
> 
> Did you solve the problem w/ threads (does it happen at all w/
> C# ?)
> 
> In my case, Java threads are spawned as "real" (native) threads, so if
> any of them tries to call Ruby code the thing will bomb (Ruby not being
> thread safe).

No, I did not solve the thread problem. In addition to the problem with
dotnet threads, if you are using Ruby threads then a call to a dotnet
object can block, preventing other Ruby threads from running.

> A few more questions:
>  * are you doing proper method dispatching on method overload?
>    This can be tricky if you perform automatic Ruby -> C#/Java type
>    conversions as there might be several matching methods

I only do a few automatic type conversions, e.g. strings. What problems
are you thinking of?

The dotnet method I am using is Type#InvokeMember which claims to choose
the best method.

>  * do you have metaclasses for static methods or such? Do you think it's
>  sensible to provide them? (I did that in rjni)

I'm not sure what you mean.

You can do
	klass = DotNet.SomeNamespace.SomeClass
	klass.SomeStaticMethod("Hello")

	inst = klass.new(1, 2, 3)
	inst.SomeStaticMethod("World")
	inst.SomeInstanceMethod(42)
	
> I'll take another look at your code when I have some time but as I know
> nothing about .net I don't know if I will understand it.
> 
> And last: is it just me, or C# is insanely easier to interface with than Java?
> There's several things in your code that take many more lines when using JNI...

I've never used JNI, so I can't say.

I think it would have been easier for me if I'd been willing to use
Microsoft's "Managed C++" language. This is a language based off C++
which can easily call dotnet code, C code and be called by C code.

Unfortunately Mono is not planning to implement this language so I wrote
half of my library in C# instead. It is easy to call C code from C#,
however it is not so easy to call C# code from C. My solution is to have
a small (~10 lines) library written in Managed C++ which causes the C#
library to call a C function in my library, passing in a bunch of
function pointers.

The C code can then call these functions to access dotnet features.

Ben Schroeder and John R. Pierce have written a similar library which
uses sockets. It implements an interesting feature (not yet available in
my library) which is to turn Ruby blocks into C# Delegates (these are
like function pointers.)

Ben offers the following example:

     dotNet.loadLibrary('System.Windows.Forms')

     button = dotNet.Button.new

     button.click.add do |sender, args|
             puts "Button clicked!"
     end

     button.performClick # -> prints <<Button clicked!>>