Re: [ROOT] TLegend strangeness: font size

From: Rene Brun (Rene.Brun@cern.ch)
Date: Mon Oct 29 2001 - 18:09:51 MET


Hi Dugan,

I tested your macro with 3.01/06 and 3.02/02. Same result. I do not see a
problem. Are you sure that you use 3.01/06 ?

You will find below an example with a TMultiGraph

Rene Brun

void test()
{

 //TCanvas *c1 = new TCanvas("c1");
 //c1->DrawFrame(0,0.7,20,1.1);
 TMultiGraph *g = new TMultiGraph("g",
        "Efficiency vs.Threshold for CEM(1,x) CJT(2,7 or 10)");

  Double_t x[] = {0,10};
  Double_t y[] = {0.95,1};
  TGraph *gr1 = new TGraph(2,x,y);

  x[0]=0.0;
  x[1]=20.0;
  y[0]=0.1;
  y[1]=1.1;
  TGraph *gr2 = new TGraph(2,x,y);

  g->Add(gr1);
  g->Add(gr2);
  //g->Draw("AP"); //if DrawFrame used
  g->Draw("AP");

  leg = new TLegend(0.6,0.7,0.9,0.9);
  leg->AddEntry(g,"CJT(2,7)","l");
  leg->AddEntry(gr1,"CJT(2,7)","l");
  leg->SetHeader("Thresholds");
  leg->Draw();

}


oneil@fnal.gov wrote:
> 
> Hi Rene,
>  I have attached a working example (test.C). I have set the y-axis range
> to 0.9-1.1, if I set the lower bound to 0.7 instead the font-size
> dramatically increases.
> 
> Thank you for your suggestions! I will try multigraph.
> 
> Cheers,
> Dugan.
> 
> On Mon, 29 Oct 2001, Rene Brun wrote:
> 
> > Hi Dugan,
> >
> > I cannot reproduce your problem with TLegend.
> > Could you send a simple but working macro reproducing this problem ?
> >
> > Note that you can simplify your program in two ways.
> >
> >  - Instead of:
> >
> >   const Int_t m=2;
> >   Float_t x1[m],z[m];
> >
> >   x[0]=0.0;
> >   x[1]=20.0;
> >   y[0]=0.75;
> >   y[1]=1.1;
> >   TGraph *g = new TGraph(m,x,y);
> >   g->SetTitle("Efficiency vs.Threshold for CEM(1,x) CJT(2,7 or 10)");
> >  you can do (assuming a TCanvas already craeted)
> >   gPad->drawFrame(0,0.75,20,1.1);
> >
> > or/and
> > use a TMultiGraph object.
> >
> > Rene Brun
> >
> >
> >   oneil@fnal.gov wrote:
> > >
> > > Hi,
> > >
> > > I am trying to overlay a bunch of graphs, so I make a dumb one to start
> > > with y-axis range from 0.75 to 1.1
> > >
> > >   const Int_t m=2;
> > >   Float_t x1[m],z[m];
> > >
> > >   x[0]=0.0;
> > >   x[1]=20.0;
> > >   y[0]=0.75;
> > >   y[1]=1.1;
> > >   TGraph *g = new TGraph(m,x,y);
> > >   g->SetTitle("Efficiency vs.Threshold for CEM(1,x) CJT(2,7 or 10)");
> > >
> > > Then I create a bunch of other graphs with stuff in them
> > >
> > >   TGraph *gr1 = new TGraph(n,x,y1);
> > >   TGraph *gr2 = new TGraph(n,x,y2);
> > >   etc.
> > >
> > > Then I draw my dumb graphs and create a legend
> > >
> > >   g->Draw("AP");
> > >   gr1->Draw("P");
> > >   etc.
> > >
> > >   leg = new TLegend(0.6,0.6,0.9,0.9);
> > >   leg->AddEntry(g,"CJT(2,7)","l");
> > >   leg->AddEntry(gr1,"CJT(2,7)","l");
> > >   leg->SetHeader("Thresholds");
> > >
> > > Then draw it.
> > >
> > >   leg->Draw();
> > >
> > > The problem is that the font in the legend comes out really small. It
> > > seems intimately connected to the y-range of the axis. If I make the range
> > > 0-1 it works great. The problem seems to occur when I make the lower
> > > bound non-zero....I get linear shrinkage as I approach the upper bound.
> > > Any ideas? It is as if it is not using NDC anymore....
> > >
> > > I am using ROOT v3.01/06.
> > >
> > > Thanks,
> > > Dugan.
> > > ----------------------------------------------------------------------------
> > >   Dugan O'Neil                        E-mail : oneil@fnal.gov
> > >                                                Dugan.O'Neil@cern.ch
> > >   Dept. of Physics and Astronomy      web    : http://www-d0.fnal.gov/~oneil
> > >   Michigan State University
> > >
> > >   phone:(630)840-2829                 It's too bad that whole families
> > >   fax  :(630)840-8886                 have to be torn apart by something
> > >                                       as simple as wild dogs.
> > >                                                - Jack Handey
> > > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> > > Mail:                                 Couriers:
> > >   MS 352                                Kirk and Wilson Streets
> > >   Fermilab                              Mail Station 352
> > >   P.O.Box 500                           Fermilab
> > >   Batavia, IL 60510-0500                Batavia, IL 60510-0500
> > > ----------------------------------------------------------------------------
> >
> 
> ----------------------------------------------------------------------------
>   Dugan O'Neil                        E-mail : oneil@fnal.gov
>                                                Dugan.O'Neil@cern.ch
>   Dept. of Physics and Astronomy      web    : http://www-d0.fnal.gov/~oneil
>   Michigan State University
> 
>   phone:(630)840-2829                 It's too bad that whole families
>   fax  :(630)840-8886                 have to be torn apart by something
>                                       as simple as wild dogs.
>                                                - Jack Handey
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> Mail:                                 Couriers:
>   MS 352                                Kirk and Wilson Streets
>   Fermilab                              Mail Station 352
>   P.O.Box 500                           Fermilab
>   Batavia, IL 60510-0500                Batavia, IL 60510-0500
> ----------------------------------------------------------------------------
> 
>   --------------------------------------------------------------------------------
> 
>                 Name: test.C
>    test.C       Type: Plain Text (TEXT/PLAIN)
>             Encoding: BASE64



This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:51:05 MET