extending ruby: ruby_init() instacloses process

[Nilbert Nullingsworth]

Hey guys, I am extending ruby in a C++ program. I have all the headers
set up and stuff so it compiles without error, but I do get one
warning (MSVC Express 2008):

c:\visual studio 2008\projects\remoteexec\remoteexec\rubyheaders\ruby/
missing.h(154) : warning C4273: 'vsnprintf' : inconsistent dll linkage
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\stdio.h(350) :
see previous definition of 'vsnprintf'

Because it doesn't keep it from compiling, I don't think this is the
source of my problem. The problem is, when I call ruby_init() to start
the interpreter, the program instantly terminates. Run by the command
line, it looks like this:

C:\Visual Studio 2008\Projects\remoteExec\Release>project.exe

C:\Visual Studio 2008\Projects\remoteExec\Release>

As you can see, it does nothing. There are no errors, even though my
code is wrapped in a try/catch thing. Here is the C++ Code:

#define WIN32_LEAN_AND_MEAN

#include <windows.h>
#include <cstdio>
#include <cstdlib>
#include <iostream>
#include "ruby.h"

using namespace std;

int main() {
//copied from teh internetz
/*HWND hWnd = GetConsoleWindow();
ShowWindow(hWnd, SW_HIDE);*/

try {
ruby_init();
} catch (char* e) {
cout << e;
}

system("PAUSE");

return 0;
}

This code isn't supposed to do anything, it's just confining the error
to that one line. No error is outputted and the program does not pause
after executing the try/catch, like it should with system("PAUSE").
Does anyone know why it is just stopping the program when ruby_init()
is called? I am using the ruby 1.9 libs and headers and stuff.

 

[Nilbert Nullingsworth]

Also as you can see, I have commented out the code that would have
stopped the console window from displaying, as I thought it may have
been the source of the problem. However, as evidenced by the command
line and the fact that no process that belongs to the program is
running, the program is indeed ceasing to exist upon reaching the
ruby_init(); line.

 

[Nilbert Nullingsworth]

Actually, placing ruby_init() anywhere in the file closes the program
instantly. Replacing the try/catch block above with this one:

try {
printf("ARGGGGGGGGGGGGGGGGGG");

printf("asdf");

ruby_init();
} catch (char* e) {
printf("%s", e);
}

and running the program via command line does not print anything.
What the heck. Anyone have any idea why the bloody blazes this is
happening?

 

[Eric Christopherson]

Actually, placing ruby_init() anywhere in the file closes the program
instantly. Replacing the try/catch block above with this one:

try {
=A0 =A0 =A0 =A0printf("ARGGGGGGGGGGGGGGGGGG");

=A0 =A0 =A0 =A0printf("asdf");

=A0 =A0 =A0 =A0ruby_init();
} catch (char* e) {
=A0 =A0 =A0 =A0printf("%s", e);
}

and running the program via command line does not print anything.
What the heck. Anyone have any idea why the bloody blazes this is
happening?

No idea, but I got it to run smoothly when compiled with MinGW (for
what little it's worth).

 

[Nilbert Nullingsworth]

Well, I may just have to learn how to use MinGW if no one can figure
out why VC++ is messing it up.

 

[Eric Christopherson]

Well, I may just have to learn how to use MinGW if no one can figure
out why VC++ is messing it up.

Back to that vsnprintf warning, what happens if you remove #include <cstdio>?

 

[Nilbert Nullingsworth]

Back to that vsnprintf warning, what happens if you remove #include <cstdio>?

Nothing happens to the error.