[ROOT] Re: another limitation or bug in rootcint

From: Masaharu Goto (MXJ02154@nifty.ne.jp)
Date: Sun Mar 09 2003 - 10:15:55 MET


Hello HP,

Thank you for your message. I'll fix this in next release.

Masa Goto


----- Original Message ----- 
From: "HP Wei" <hp@rentec.com>
To: <roottalk@pcroot.cern.ch>
Cc: <MXJ02154@nifty.ne.jp>
Sent: Thursday, March 06, 2003 11:36 PM
Subject: another limitation or bug in rootcint


> // --------- test.H
> class A;
>  
> enum Op{
>   OP_Neg,
>   OP_NOT,
>   OP_Sub};
> 
> A fxx(Op,A,A);
> inline A operator-(A x,A y){return fxx(OP_Sub,x,y);}
> 
> A fx(Op,A);
> inline A operator-(A x){return fx(OP_Neg,x);}
> inline A operator!(A x){return fx(OP_Not,x);}
> 
> // ---------- testLinkDef.h
> #ifdef __CINT__
> 
> #pragma link off all globals;
> #pragma link off all classes;
> #pragma link off all functions;
> 
> #pragma link C++ function fxx;
> #pragma link C++ function operator-(A,A);
> #pragma link C++ function fx;
> #pragma link C++ function operator-(A);
> #pragma link C++ function operator!(A);
> #endif
> 
> // ---------------------- Problem ***
> After the command: rootcint testDict.C -c -p test.H testLinkDef.h
> (-p does not make any difference),
> in the testDict.C
> the function for the second operator-(A) is not generated.
> 
> Is this a bug or limitation of rootcint ?
> Or do I need to change the testLinkDef.h ?
> 
> My Cint version is 5.15.76 on Sun Solaris 8 with gcc.
> 
> Thanks,
> HP
> 
> 



This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:10 MET