Re: Failure in TTree filling

Akira,

In your script, you create the Tree in a directory of a file open in readonly mode.
This is like creating a memory-resident tree.

Rene Brun

OKUMURA, Akira wrote:
> Hello ROOTers,
>
> I encountered a problem with TTree which is similar to roottalk
> messages posted in June/2006.
> I have read the messages previously posted, but could not solve the
> problem.
>
> root [0] .x PHD.C+
> Error in <TTree::Fill>: Failed filling branch:ch2.pedestal, nbytes=-1
> This error is symptomatic of a Tree created as a memory-resident Tree
> Instead of doing:
> TTree *T = new TTree(...)
> TFile *f = new TFile(...)
> you should do:
> TFile *f = new TFile(...)
> TTree *T = new TTree(...)
> ( continue ... )
>
> I wrote a macro (PHD.C) to analyze pulse height distribution of PMTs.
> It opens some .root files which contain TH1D branch of 10000 entries.
> All TH1D is curve data of an oscilloscope.
>
> I will attache my macro.
> What is wrong with it ?
>
> Sincerely,
>
> OKUMURA, Akira oxon_at_icrr.u-tokyo.ac.jp
> Institute for Cosmic Ray Research, University of Tokyo
> 5-1-5 Kashiwanoha Kashiwa Chiba 277-8582 Japan
> Phone/Fax : +81 4-7136-3153
> Skype : okumura.akira
>
> On 2006/06/30, at 21:28, Rene Brun wrote:
>
>> You can simplify your code as shown below.
>> In your case it looks like you did not create T1 in f1, but you saved T1
>> via T1->Write() in f1.
>>
>> Rene Brun
>>
>> TFile *f1 = new TFile(...)
>> TTree *T1 = new TTree(...)
>> T1->Branch...
>> ...
>> TFile *f2 = new TFile(...)
>> TTree *T2 = new TTree(...)
>> T2->Branch...
>> ...
>> fill branches (no need to cd to the file)
>> T1->Fill();
>> T2->Fill();...
>>
>> //save headers (no need to cd() if using Autosave)
>> T1->AutoSave();
>> T2->Autosave();
>> ...
>> delete f1;
>> delete f2;
>> ...
>>
>> Francois-Xavier Girod wrote:
>>> Dear rooters,
>>>
>>> I am using root version 5.11/06, but the problem also appears on
>>> earlier
>>> versions.
>>>
>>> What should I do if I want to write several trees in several files ?
>>>
>>> I did
>>> TFile *f1 = new TFile(...)
>>> TTree *T1 = new TTree(...)
>>> T1->Branch...
>>> ...
>>> TFile *f2 = new TFile(...)
>>> TTree *T2 = new TTree(...)
>>> T2->Branch...
>>> ...
>>> ...
>>> (process stuff and fill variables)
>>> ...
>>> f1->cd()
>>> T1->Fill()
>>> f2->cd()
>>> T2->Fill()
>>> ...
>>> f1->cd()
>>> T1->Write()
>>> f2->cd()
>>> f2->Write()
>>>
>>>
>>> f2 file is ok, with T2 tree containing its data. However, f1 file
>>> contains
>>> T1 tree, which has correct keys (for instance I can get the number of
>>> events or the list of leaves) but contains no data. More
>>> specifically if I
>>> try to draw something I get :
>>>
>>> Error in <TFile::ReadBuffer>: error reading all requested bytes from
>>> file
>>> outTreeResultDVCS.root, got 0 of 29892
>>>
>>> I should add that, this only happens for the analysis of a large
>>> number of
>>> events. When I process a few events, the output is correct.
>>>
>>> My guess is that, the way I proceed only the last declared tree is disk
>>> resident, whereas the previous ones are memory resident.
>>>
>>> Thank you for help.
>>>
>>> Regards,
>>> Francois-Xavier
>>>
>>> --
>>>
>>> Girod Francois-Xavier
>>>
>>> CEA Saclay
>>> DSM / DAPNIA / SPhN
>>> Orme des Merisiers
>>> Bat 703 piece 133 C
>>> 91191 Gif-sur-Yvette
>>> France
>>>
>>> tel : (+33) 1 69 08 74 29
>>> fax : (+33) 1 69 08 75 84
>>>
>>> On Wed, 28 Jun 2006, Rene Brun wrote:
>>>
>>>
>>>> When this message is printed with teh CVS head version, you also
>>>> get the
>>>> following print:
>>>>
>>>> This error is symptomatic of a Tree created as a memory-resident Tree
>>>> Instead of doing:
>>>> TTree *T = new TTree(...)
>>>> TFile *f = new TFile(...)
>>>> you should do:
>>>> TFile *f = new TFile(...)
>>>> TTree *T = new TTree(...)
>>>>
>>>>
>>>> Rene Brun
>>>>
>>>> Alberto Pulvirenti wrote:
>>>>
>>>>> Dear rooters,
>>>>>
>>>>> sometimes, while filling a TTree, I encountered this message:
>>>>>
>>>>> Error in <TTree::Fill>: Failed filling branch:matches.matches,
>>>>> nbytes=-1
>>>>>
>>>>> can someone explain me what does it mean? What mistake did I do while
>>>>> creating the code to fill that TTree?
>>>>> For completeness I must say that this TTree is filled with a branch
>>>>> made from a struct of primitive data types.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Alberto
>>>>>
>>>>> **************************************
>>>>> ----- Alberto Pulvirenti, Ph. D. -----
>>>>> Universita' di Catania,
>>>>> Dipartimento di Fisica e Astronomia
>>>>> Via Santa Sofia, 64
>>>>> I-95129, Catania, Italy
>>>>>
>>>>> Tel :+39-095-3785286
>>>>> **************************************
>>>>>
>>>>
>>
>>
>
Received on Thu Sep 28 2006 - 21:54:46 MEST