Re: [ROOT] ROOT::Math::poisson_pdf(0,0)=NaN while TMath::PoissonI(0,0)=1

Hello Corey, 

 I agree that poisson_pdf(0,0) should return 1, although in theory the Poisson is defined only for value of
the Poisson parameter greater than zero. 
I have modified now ROOT::Math::poisson_pdf in the ROOT trunk

 Thank you for your report

 Lorenzo

On Jul 28, 2010, at 1:55 PM, Corey Reed wrote:

> Hi,
> 
> Just in case anyone (else) encounters this, I thought I would post it to this list.
> 
> In Root 5.26.00/b,
> 
> root [9] ROOT::Math::poisson_pdf(0,0)
> (double)nan
> 
> root [11] TMath::PoissonI(0,0)
> (Double_t)1.00000000000000000e+00
> 
> I personally would have expected the poisson_pdf to give 1 (the chance to get 0 events in the limit that the
mean goes to 0).
> 
> I don't know if this behavior is intended or not.
> 
> - Corey

Re: [ROOT] Error in <TTree::SetBranchStatus>: unknown branch

Re: [ROOT] Error in <TTree::SetBranchStatus>: unknown branch

Hi,

you can use TTree::GetListOfBranches()->FindObject("branchname") to
determine whether the tree has a branch called branchname:

if (tree->GetListOfBranches()->FindObject("jet.px"))
  b_jet_px = tree->GetBranch("jet.px")
else
  b_jet_px = 0;
...
if (b_jet_px) tree->SetBranchStatus("jet.px"...

Cheers, Axel.

Harinder Singh Bawa wrote on 08/02/2010 01:13 PM:
> Dear Axel, Philips,
> 
> Thanks for the reply but I figured out that the branches are not in my
> input rootfiles because these input files are older and my code is
> according to newer one.
>  So now my question is
> I need to analyse the rootfiles(all some with more variables and some
> with less variables). What needs to be done so that I dont get erorr
> like variables not found.
> 
> I did MakeClass with newer one and so I am having more variables defined
> in my .h file. Now If I run to older file, it is complaiing that these
> variables are unknown branches. I want to run on both older and newer files.
> 
> Kindly suggest
> 
> 
> 
> 
> Harinder
> *From: *"Philippe Canal" <pcanal <at> fnal.gov>
> *To: *"Harinder Singh Bawa" <hbawa <at> csufresno.edu>
> *Cc: *"roottalk" <roottalk <at> cern.ch>
> *Sent: *Friday, July 30, 2010 5:22:18 PM
> *Subject: *Re: [ROOT] Error in <TTree::SetBranchStatus>: unknown branch
> 
> Hi,
> 
> It is plausible that you do not have any branch with exactly those
> names.   You may want to try:
> 
>   fChain->SetBranchStatus("L2_xe12*",1);
> 
> or double check the names (via mytree->Print()).
> 
> Cheers,
> Philippe.
> 
> 
> On 7/30/10 7:55 AM, Harinder Singh Bawa wrote:
> 
>     Dear Root Experts,
> 
>     1. I need to understand 2 things. First I opened a rootfile and made
>     .C and .h using MakeClass. I am expecting that MakeClass
>     incorporated all the branch addresses and variables. But when I run
>     , I am getting the following error.
> 
>     Error in <TTree::SetBranchStatus>: unknown branch  -> trig_EF_ph
>     Error in <TTree::SetBranchStatus>: unknown branch  -> L2_xe12
>     Error in <TTree::SetBranchStatus>: unknown branch  -> xxxxxxxxx(blah
>     blah)
>     Error in <TTree::SetBranchStatus>: unknown branch  -> xxxxxxxxx
>     Error in <TTree::SetBranchStatus>: unknown branch  -> xxxxxxxxx
>     Error in <TTree::SetBranchStatus>: unknown branch  -> xxxxxxxxx
>     Error in <TTree::SetBranchStatus>: unknown branch  -> xxxxxxxxx
> 
>     What needs to be done ? The code is running but I want to get rid of it
> 
>     2. I did something in my .C file as
>      fChain->SetBranchStatus("*",0);
> 
>       if ( stream_analyzed==TYPE_DATA) {
>         fChain->SetBranchStatus("RunNumber",1);
>         fChain->SetBranchStatus("lbn",1);
>       }
>       fChain->SetBranchStatus("L1_EM10",1);
>       fChain->SetBranchStatus("vxp_nTracks",1);
>       fChain->SetBranchStatus("MET_Goodness_N90Cells_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_PtEM_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_N90Cells_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_HECf_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_QualityFrac_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_EMFraction_Jet",1);
>       fChain->SetBranchStatus("MET_Goodness_Timing_Jet",1);
>       fChain->SetBranchStatus("el_n",1);
>       fChain->SetBranchStatus("el_cl_eta",1);
>       fChain->SetBranchStatus("el_cl_phi",1);
>       fChain->SetBranchStatus("el_cl_pt",1);
>       fChain->SetBranchStatus("el_author",1);
>       fChain->SetBranchStatus("el_loose",1);
>       fChain->SetBranchStatus("el_medium",1);
>       fChain->SetBranchStatus("el_tightIso",1);
>       fChain->SetBranchStatus("el_isEM",1);
>       fChain->SetBranchStatus("el_expectHitInBLayer",1);
>       fChain->SetBranchStatus("MET_LocHadTopo_et",1);
>       fChain->SetBranchStatus("MET_LocHadTopo_phi",1);
> 
>     In order to read only those variables which are of my interest. I
>     thought , it will read only the above variables and branches and I
>     should not get the errors as pointed out in 1, because it is not
>     reading now all the branches. But I think It is reading but taking
>     only those which are defined above. Am I right?
> 
> 
> 
>     If yes,
>     Then My question is same for both points. How to get rid of the errors.
> 
>     thanks
> 
> 
> 
> 
> 
> 
> 
>     Harinder Singh Bawa
>     California State University Fresno
>     Alternative email:-  bawa <at> cern.ch
> 
>     Office:-
>     CERN B591-R-024                         CSU Fresno:- Science II Room326
>     Phone:- +41-22-767-78187                Telephone:- +1-559-278-5802
>     CERN Mobile:- 0764878721
>     Res:- 0033-457055309
> 

[ROOT] Aclic : option for compilation in 32 bits

Hello,

i use a program in compiled mode which needs to be in 32 bits. This 
programs need a shared object of a program, that is created "at hand" by 
doing :
root
...
//create a TChain
...
return mychain_data->Process("myselector.C+");

==>it creates myselector_C.so

but when i compile the main project in 32 bits with -m32 (mandatory 
because of some packages only available in 32 bits)
and use in the compilation the so-called myselector_C.so, i occur the 
following problem at link :

bin/myselector_C.so: could not read symbols: File in wrong format

from google, it would mean an imcompatibility of architecture (typically 
32 bits versus 64 bits)

would you have an idea ?

thanks
(unfortunately here, i don't have a minimal program to reproduce the 
problem, but i could try if necessary)

Re: [ROOT] Aclic : option for compilation in 32 bits

Hi Marc,

can you use a 32bit version of ROOT to create myselector_C.so?

Cheers, Axel.

Marc Escalier wrote on 08/02/2010 05:58 PM:
> Hello,
> 
> i use a program in compiled mode which needs to be in 32 bits. This
> programs need a shared object of a program, that is created "at hand" by
> doing :
> root
> ...
> //create a TChain
> ...
> return mychain_data->Process("myselector.C+");
> 
> ==>it creates myselector_C.so
> 
> but when i compile the main project in 32 bits with -m32 (mandatory
> because of some packages only available in 32 bits)
> and use in the compilation the so-called myselector_C.so, i occur the
> following problem at link :
> 
> bin/myselector_C.so: could not read symbols: File in wrong format
> 
> from google, it would mean an imcompatibility of architecture (typically
> 32 bits versus 64 bits)
> 
> would you have an idea ?
> 
> thanks
> (unfortunately here, i don't have a minimal program to reproduce the
> problem, but i could try if necessary)
> 
> 

Re: [ROOT] Aclic : option for compilation in 32 bits

Hi Marc,

With a recent version of ROOT, the following should work:

      TString cmd = gSystem->GetMakeSharedLib();
      cmd.ReplaceAll("-m64","-m32");
      gSystem->SetMakeSharedLib();

Cheers,
Philippe.

PS.   Of course the simpliest and cleanest solution is to use a 32bits 
ROOT to do the compilation.

On 8/2/10 10:58 AM, Marc Escalier wrote:
> Hello,
>
> i use a program in compiled mode which needs to be in 32 bits. This 
> programs need a shared object of a program, that is created "at hand" 
> by doing :
> root
> ...
> //create a TChain
> ...
> return mychain_data->Process("myselector.C+");
>
> ==>it creates myselector_C.so
>
> but when i compile the main project in 32 bits with -m32 (mandatory 
> because of some packages only available in 32 bits)
> and use in the compilation the so-called myselector_C.so, i occur the 
> following problem at link :
>
> bin/myselector_C.so: could not read symbols: File in wrong format
>
> from google, it would mean an imcompatibility of architecture 
> (typically 32 bits versus 64 bits)
>
> would you have an idea ?
>
> thanks
> (unfortunately here, i don't have a minimal program to reproduce the 
> problem, but i could try if necessary)
>

Re: [ROOT] Aclic : option for compilation in 32 bits

thanks Philippe and Axel,

by using a root version 32 bits, as you say, it fixes the problem 
(there is other problems on the computer that has the version 32 bits 
of root, but i will try to fix it)

by the way, on the root 64 bits (which is root 5.26/00b) when i did, as 
suggested by Philippe :

TString cmd = gSystem->GetMakeSharedLib();
cmd.ReplaceAll("-m64","-m32");
gSystem->SetMakeSharedLib(cmd);

i got a problem at the compilation with libCint, even with the 
"SetMakeSharedLib function" :

==> 
/afs/in2p3.fr/system/amd64_sl5/usr/local/root/root_v5.26.00b-gcc3.4/root/lib/libCint.so: 
could not read symbols: File in wrong format

example :
root [0] .x CreateDictionnary.C
(cout) before ReplaceAll : cmd=cd $BuildDir ; g++ -c $Opt -pipe 
-m64 -Wall -W -Woverloaded-virtual -fPIC -pthread 
$IncludePath $SourceFiles ;  g++ $ObjectFiles -shared 
-Wl,-soname,$LibName.so -m64 -O2  $LinkedLibs -o 
$SharedLib
(cout)after ReplaceAll : cmd=cd $BuildDir ; g++ -c $Opt -pipe 
-m32 -Wall -W -Woverloaded-virtual -fPIC -pthread 
$IncludePath $SourceFiles ;  g++ $ObjectFiles -shared 
-Wl,-soname,$LibName.so -m32 -O2  $LinkedLibs -o 
$SharedLib
Info in <TUnixSystem::ACLiC>: creating shared library 
/afs/in2p3.fr/home/e/escalier/public/DiPhotons/4DMatrix/WithSelector/./ComputeCategories_2Steps_C.so
/afs/in2p3.fr/system/amd64_sl5/usr/local/root/root_v5.26.00b-gcc3.4/root/lib/libCint.so: 
could not read symbols: File in wrong format
collect2: ld returned 1 exit status
Error in <ACLiC>: Compilation failed!
Error in <TSelector::GetSelector>: file 
ComputeCategories_2Steps.C+ does not have a valid class 
deriving from TSelector

thanks

===============
On Mon, 2 Aug 2010, Philippe Canal wrote:

> Hi Marc,
>
> With a recent version of ROOT, the following should work:
>
>     TString cmd = gSystem->GetMakeSharedLib();
>     cmd.ReplaceAll("-m64","-m32");
>     gSystem->SetMakeSharedLib();
>
> Cheers,
> Philippe.
>
> PS.   Of course the simpliest and cleanest solution is to use a 32bits ROOT 
> to do the compilation.
>
> On 8/2/10 10:58 AM, Marc Escalier wrote:
>> Hello,
>> 
>> i use a program in compiled mode which needs to be in 32 bits. This 
>> programs need a shared object of a program, that is created "at hand" by 
>> doing :
>> root
>> ...
>> //create a TChain
>> ...
>> return mychain_data->Process("myselector.C+");
>> 
>> ==>it creates myselector_C.so
>> 
>> but when i compile the main project in 32 bits with -m32 (mandatory because 
>> of some packages only available in 32 bits)
>> and use in the compilation the so-called myselector_C.so, i occur the 
>> following problem at link :
>> 
>> bin/myselector_C.so: could not read symbols: File in wrong format
>> 
>> from google, it would mean an imcompatibility of architecture (typically 32 
>> bits versus 64 bits)
>> 
>> would you have an idea ?
>> 
>> thanks
>> (unfortunately here, i don't have a minimal program to reproduce the 
>> problem, but i could try if necessary)
>> 
>

Re: [ROOT] Aclic : option for compilation in 32 bits

 > i got a problem at the compilation with libCint, even with the 
"SetMakeSharedLib function" :

Humm .. indeed.   So you must use a 32bit build of ROOT.

Cheers,
Philippe.

On 8/2/10 2:30 PM, Marc Escalier wrote:
> thanks Philippe and Axel,
>
> by using a root version 32 bits, as you say, it fixes the problem 
> (there is other problems on the computer that has the version 32 bits 
> of root, but i will try to fix it)
>
> by the way, on the root 64 bits (which is root 5.26/00b) when i did, 
> as suggested by Philippe :
>
> TString cmd = gSystem->GetMakeSharedLib();
> cmd.ReplaceAll("-m64","-m32");
> gSystem->SetMakeSharedLib(cmd);
>
> i got a problem at the compilation with libCint, even with the 
> "SetMakeSharedLib function" :
>
> ==> 
> /afs/in2p3.fr/system/amd64_sl5/usr/local/root/root_v5.26.00b-gcc3.4/root/lib/libCint.so: 
> could not read symbols: File in wrong format
>
> example :
> root [0] .x CreateDictionnary.C
> (cout) before ReplaceAll : cmd=cd $BuildDir ; g++ -c $Opt -pipe -m64 
> -Wall -W -Woverloaded-virtual -fPIC -pthread $IncludePath $SourceFiles 
> ;  g++ $ObjectFiles -shared -Wl,-soname,$LibName.so -m64 -O2  
> $LinkedLibs -o $SharedLib
> (cout)after ReplaceAll : cmd=cd $BuildDir ; g++ -c $Opt -pipe -m32 
> -Wall -W -Woverloaded-virtual -fPIC -pthread $IncludePath $SourceFiles 
> ;  g++ $ObjectFiles -shared -Wl,-soname,$LibName.so -m32 -O2  
> $LinkedLibs -o $SharedLib
> Info in <TUnixSystem::ACLiC>: creating shared library 
>
/afs/in2p3.fr/home/e/escalier/public/DiPhotons/4DMatrix/WithSelector/./ComputeCategories_2Steps_C.so 
>
> /afs/in2p3.fr/system/amd64_sl5/usr/local/root/root_v5.26.00b-gcc3.4/root/lib/libCint.so: 
> could not read symbols: File in wrong format
> collect2: ld returned 1 exit status
> Error in <ACLiC>: Compilation failed!
> Error in <TSelector::GetSelector>: file ComputeCategories_2Steps.C+ 
> does not have a valid class deriving from TSelector
>
> thanks
>
> ===============
> On Mon, 2 Aug 2010, Philippe Canal wrote:
>
>> Hi Marc,
>>
>> With a recent version of ROOT, the following should work:
>>
>>     TString cmd = gSystem->GetMakeSharedLib();
>>     cmd.ReplaceAll("-m64","-m32");
>>     gSystem->SetMakeSharedLib();
>>
>> Cheers,
>> Philippe.
>>
>> PS.   Of course the simpliest and cleanest solution is to use a 
>> 32bits ROOT to do the compilation.
>>
>> On 8/2/10 10:58 AM, Marc Escalier wrote:
>>> Hello,
>>>
>>> i use a program in compiled mode which needs to be in 32 bits. This 
>>> programs need a shared object of a program, that is created "at 
>>> hand" by doing :
>>> root
>>> ...
>>> //create a TChain
>>> ...
>>> return mychain_data->Process("myselector.C+");
>>>
>>> ==>it creates myselector_C.so
>>>
>>> but when i compile the main project in 32 bits with -m32 (mandatory 
>>> because of some packages only available in 32 bits)
>>> and use in the compilation the so-called myselector_C.so, i occur 
>>> the following problem at link :
>>>
>>> bin/myselector_C.so: could not read symbols: File in wrong format
>>>
>>> from google, it would mean an imcompatibility of architecture 
>>> (typically 32 bits versus 64 bits)
>>>
>>> would you have an idea ?
>>>
>>> thanks
>>> (unfortunately here, i don't have a minimal program to reproduce the 
>>> problem, but i could try if necessary)
>>>
>>

[ROOT] no dictionary for class TPaletteAxis is available

Re: [ROOT] no dictionary for class TPaletteAxis is available

Hi,

it was discussed some time ago here :
(thanks to Henso who reads this hypernews to have found the link and 
explanation)

http://root.cern.ch/phpBB3//viewtopic.php?t=6947&view=next&sid=cd602857c8522df6ac8123ae9f29ebfd

to summarize, it would be due to the version of root that you use (below 
5.20 in cint)

==>if you create a dummy canvas before your loop, the warning should 
disappear
TCanvas *mydummycanvas=new TCanvas();

regards
=============
=============
Harinder Singh Bawa a écrit :
> Dear root experts,
>
> I am getting this strange error when I call the Object Quality mapfile 
> in my analysis. Although it doesnot effect or crash my analysis but I 
> want to get rid of it. Can anybody suggest what is this for??
> *********************************
> checkOQ: opening file ObjectQualityMaps_run152166.root
> Warning in <TClass::TClass>: no dictionary for class TPaletteAxis is 
> available
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> Error in <TBufferFile::ReadObject>: trying to read an emulated class 
> (TPaletteAxis) to store in a compiled pointer (TObject)
> ***********************
>
> Harinder