RE:CINT problem

From: Masaharu Goto (MXJ02154@nifty.ne.jp)
Date: Tue Jan 25 2000 - 13:57:40 MET


Hello Nicolas,

Thank you for reporting this problem. I will fix this in cint5.14.29.
Meanwhile, the problem can be worked around by using typedef.

  typedef void(*FTrans_t)(const Image &,Shadowgram &);
  void f(FTrans_t FTrans) { }
  void g() {
    FTrans_t FTrans;
    f(Ftrans);
  }

Thank you
Masaharu Goto

>
>I got the following:
>
>Error: No symbol MwdCF(flux,lambda,sigma,dobs,mu,omega,mem,def,z,LTrans)
>in current scope  FILE:mwdoneit.cc LINE:143
>Possible candidates are...
>filename       line:size busy function type and name
>mwdoneit.cc      37:32   0 public: void MwdCF(double flux,const
>Shadowgram& lambda,const Shadowgram& sigma,const Shadowgram& dobs,double
>mu,double omega,Image& mem,const Image& def,double& z,void (*)(const
>Shadowgram&,Image&) LTrans);
>
>But this call is made inside the suceffully called routine:
>void MwdOneIt(bool doinit,int nconj,double flux,Image &mem, 
>             const Image &def,
>             const Shadowgram &sigma,const Shadowgram &dobs,
>             Shadowgram &dpred,Shadowgram &dzdlam,
>             Shadowgram &dellam,
>             Shadowgram &pdzdla,
>             double iomega,double tol,double &length,
>             Shadowgram &lambda,double &mu,double &z,
>             void (*FTrans)(const Image &,Shadowgram &),
>             void (*LTrans)(const Shadowgram &,Image &)){
>    static double grad,pgrad,scale;
>    double step,stepm,stepp,omega,gds0,gds1,t;
>
>So it seems to me that the types of the variables do correspond fully to
>the 'Possible candidate'
>g++ is happy with this code.
>
>-- 
>              Nicolas Produit



This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:17 MET