Topic: AT&T C++ 2.0 Efficacious?
Author: timh@.mentorg.com (Tim Harvey @ APD x1381)
Date: 18 Jan 91 22:41:31 GMT Raw View
Eric Valette (Article 496) thought that AT&T C++ 2.0 might be efficacious.
I don't think so in this case.
I ran C++/cfront version 2.0 release 2c on his example, and as you can
see below, the C output looks fine.
Eric, did you take your question to where ever you got your cfront port?
It looks like you posted gnu.g++.help. Are you using gnu g++?
// C++ source code
class example {
private:
int a ;
public:
int getVal() { return a ; }
void setVal( int val ) { a = val ; }
};
example glob ;
void f() {
example loc;
int a = 1 ;
int b = 2 ;
int resu ;
loc.setVal(1) ;
resu = loc.getVal() ;
glob.setVal(1) ;
resu = glob.getVal() ;
}
Compiled on an Apollo 4500. Generated C code follows.
C++/cfront version 2.0 release 2c [apollo.mc68000.sys5.]
"t.c", line 13: warning: loc used but not set
char *__vec_new() ;
char __vec_delete() ;
typedef int (*__vptp)() ;
struct __mptr {
short d ;
short i ;
__vptp f ;
};
struct example {
int a__7example ;
};
extern struct __mptr *__ptbl_vec__t_c_[] ;
static int getVal__7exampleFv( __0this )
struct example *__0this ;
{
return __0this->a__7example ;
}
static char setVal__7exampleFi( __0this, __0val )
struct example *__0this ;
int __0val ;
{
__0this->a__7example = __0val ;
}
struct example glob = { 0 } ;
char f__Fv() {
struct example __1loc ;
int __1a ;
int __1b ;
int __1resu ;
__1a = 1 ;
__1b = 2 ;
setVal__7exampleFi( &__1loc, 1 ) ;
__1resu = getVal__7exampleFv( &__1loc ) ;
setVal__7exampleFi( &glob, 1 ) ;
__1resu = getVal__7exampleFv( &glob ) ;
}