Re: [ROOT] continue statement in interpreted code??

From: Mike Kordosky (kordosky@mail.hep.utexas.edu)
Date: Thu Aug 16 2001 - 23:15:20 MEST


Hi Rene,

On Thu, 16 Aug 2001, Rene Brun wrote:

I don't care about seeing those histograms after the function is done. I
understand that they go out of scope.  This is sample code designed to
illustrate a problem that I do not understand.  My question has to do with
the continue statement.  If I leave it in, CINT complains:

Error: non class,struct,union object $gPad used with . or ->
FILE:TestBug.C LINE:24
*** Interpreter error recovered ***

Commenting out the continue statement I see that the histograms are filled
1000 times and the screen updates once every 10 fills.  Yes, at the end of
the function call everything disappears.

mike


> Mike,
> You have many scoping problems in your macro. You should create
> your objects in the heap if you want to still see them once
> the macro has finished.
> You will find below a simplification of your macro:
>
> Rene Brun
>
>
> void test_bug(){
>         TH1F *h1 = new TH1F("h1", "h1", 100, 1, 100);
>         TH1F *h2 = new TH1F("h2", "h2", 100, 100, 200);
>         TCanvas *can = new TCanvas("can", "can", 600, 800);
>         can->Divide(1,2);
>         can->cd(1);
>         TPad *pad1 = (TPad*)gPad;
>         h1->Draw();
>         can->cd(2);
>         TPad *pad2 = (TPad*)gPad;
>         h2->Draw();
>         Double_t result = 0;
>         Int_t i = 0;
>         while(i<1000){
>                 result = gRandom->Gaus(50, 5);
>                 if(result < 50){
>                         i++;
>                         continue;
>                 }
>                 h1->Fill(result);
>                 h2->Fill(result*2);
>                 if(i%10 == 0){
>                         pad1->Modified();
>                         pad2->Modified();
>                         can->Update();
>                 }
>                 i++;
>         }
> }
>
>
> On Thu, 16 Aug 2001, Mike Kordosky wrote:
>
> > Hi,
> >
> > I have the following macro.  The macro works as expected if I comment out
> > the continue statement.  When I leave continue in, CINT complains:
> >
> > root [1] test_bug();
> > Error: non class,struct,union object $gPad used with . or ->
> > FILE:TestBug.C LINE:24
> >
> > void test_bug(){
> >         TH1F h1("h1", "h1", 100, 1, 100);
> >         TH1F h2("h2", "h2", 100, 100, 200);
> >         TCanvas can("can", "can", 1000, 1000);
> >         can.Divide(1,2);
> >         can.cd(1);
> >         h1.Draw();
> >         can.cd(2);
> >         h2.Draw();
> >         Double_t result = 0;
> >         Int_t i = 0;
> > //      for(Int_t i = 0; i<1000; i++){
> >         while(i<1000){
> >                 result = gRandom->Gaus(50, 5);
> >                 if(result < 50){
> >                         i++;
> >                         continue;
> >                 }
> >                 h1.Fill(result);
> >                 h2.Fill(result*2);
> >                 if(i%10 == 0){
> >                         can.cd(1);
> >                         h1.Draw();
> >                         gPad->Update();
> >                         can.cd(2);
> >                         h2.Draw();
> >                         gPad->Update();
> >                 }
> >                 i++;
> >         }
> > }
> >
> >
> > Is this a bug, or am I missing something?
> >
> > Mike Kordosky
> >
> > --
> > Graduate Research Assistant  // High Energy Physics Lab
> > kordosky@hep.utexas.edu     // University of Texas at Austin
> > kordosky@fnal.gov	   //
> > ph: (512) 471-8426 (RLM Lab, Office)
> >     (512) 475-8673 (ENS Lab)
> >
> >
> >
> >
>

-- 
Graduate Research Assistant  // High Energy Physics Lab
kordosky@hep.utexas.edu     // University of Texas at Austin
kordosky@fnal.gov	   //
ph: (512) 471-8426 (RLM Lab, Office)
    (512) 475-8673 (ENS Lab)



This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:50:57 MET