Hi David,
I fixed this problem last week in the development version in CVS.
Rene Brun
On Fri, 10 Jan 2003, David d'Enterria wrote:
> Hello Rooters,
>
> A sequence of CINT commands showing two(?) problems:
>
> // This works:
>
> const Int_t n = 5;
> Float_t y[n] = {5.,7.,10.,14.,16.};
> Float_t x[n] = {5.,7.,9.,13.,15.};
> TGraph *g = new TGraph(n,x,y);
> g->GetHistogram()
> (const class TH1F*)0x87365c0 // non-null address. OK, no need of
> g->Draw("AP") after 3.04 release
>
> // although:
>
> g->GetXaxis()
> (const class TAxis*)0x0
> g->GetHistogram()->GetXaxis()
> (const class TAxis*)0x8736600
>
> // Is this above a bug or a feature ?
>
> // Let's go on and try now with TGraphErrors
>
> Float_t ex[n] = {0.01,0.02,0.02,0.01,0.01};
> Float_t ey[n] = {0.,0.,0.,0.,0.};
> TGraphErrors *graph = new TGraphErrors(n,x,y,ex,ey);
> graph->GetXaxis()
> (const class TAxis*)0x0
> graph->GetHistogram()
> *** Break *** segmentation violation
>
> Relevant part of back trace from gdb:
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 1024 (LWP 994)]
> 0x408d2431 in TGraphErrors::ComputeRange () from /usr/root/lib/libGraf.so
> (gdb) back
> #0 0x408d2431 in TGraphErrors::ComputeRange () from
> /usr/root/lib/libGraf.so
> #1 0x408c7a4e in TGraph::GetHistogram () from /usr/root/lib/libGraf.so
> #2 0x4091f694 in G__TGraph_GetHistogram_0_3 () from
> /usr/root/lib/libGraf.so
> ....
>
> Am I doing something wrong here ?
>
> Thanks in advance,
>
> PS: Using ROOT 3.04/02 binaries on a redhat 7.1 machine
>
> --
> David D'ENTERRIA PostDoctoral Research Scientist
> Nevis Labs http://www.nevis.columbia.edu/~denterria
> Columbia University, NY mailto:denterria@nevis.columbia.edu
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> PHENIX Group (Bldg 510C) tel: +1 631 344 7053
> Brookhaven National Lab, Upton, NY 11973 fax: +1 631 344 3253
> http://www.phenix.bnl.gov/~enterria mailto:denterri@bnl.gov
> ---------------------------------------------------------------------
>
>
This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:08 MET