Re: Problem with merging trees via hadd

From: Michael Boehler <michael.boehler_at_cern.ch>
Date: Tue, 15 May 2012 19:57:38 +0200


Hi Philippe,

On 05/15/2012 07:48 PM, Philippe Canal wrote:
> Hi Michael,
>
> How did the merge fails?

The number of entries in the trees of the merged file (newFile.root) does not match with my control histograms in the directory.
> (The instances are 'cycles' or backup copies of the TTree meta-data
> and hadd will ignore all but the highest cycle version).

I saw, that the highest cycle version differes from the file in output_0001 to output_0002 and output_0003.

If the first file in hadd newFile.root file1.root file2.root file3.root has another cycle number compared to the other files, the trees are not merged.
(the tree in the merged file only contains the entries from file1.root ).

If the first file is output_0002/ZmumuNp5_pt20.root for example, which has the same cycle number as output_0003/ZmumuNp5_pt20.root, the newfile has exactly the same number of entries in the tree as output_0002/ZmumuNp5_pt20.root + output_0003/ZmumuNp5_pt20.root, but the entries in output_0001/ZmumuNp5_pt20.root are missing in the merged file.

Thanks for any help.

Cheers,
Michael

>
> Cheers,
> Philippe.
>
> On 5/15/12 12:43 PM, Michael Boehler wrote:
>> Dear RootTalkers,
>>
>> I have the following problem:
>>
>> I'm trying to merge (three) root files with a directory with
>> histograms and trees via hadd.
>>
>> hadd newFile.root output_0001/ZmumuNp5_pt20.root
>> output_0002/ZmumuNp5_pt20.root output_0003/ZmumuNp5_pt20.root
>>
>>
>> The content of the files is listed below. The files have different
>> instances of the trees e.g. tau;26 in output_0001 and tau;17 in
>> output_0002.
>> The histogram merging is working perfectly. But the trees can only be
>> merged if the number behind ; is the same.
>>
>> First I was working with ROOT530. But also checking the *Pro*,
>> version 5.32/03
>> <http://root.cern.ch/drupal/content/production-version-532> didn't
>> solve this problem.
>> Is this a known problem? Is there any quick solution to solve this
>> problem?
>>
>> Thanks for any help.
>>
>> Cheers,
>> Michael
>>
>>
>>
>>
>> [lxplus247] /tmp/boehlerm > root -l output_0001/ZmumuNp5_pt20.root
>> root [0]
>> Attaching file output_0001/ZmumuNp5_pt20.root as _file0...
>> root [1] .ls
>> TFile** output_0001/ZmumuNp5_pt20.root
>> TFile* output_0001/ZmumuNp5_pt20.root
>> KEY: TDirectoryFile histos;1 histos
>> KEY: TTree tau;26 physics
>> KEY: TTree tau;25 physics
>> KEY: TTree tmvatree;1 tmvatree
>> KEY: TTree tauinfo;1 tauinfo
>>
>>
>> [lxplus247] /tmp/boehlerm > root -l output_0002/ZmumuNp5_pt20.root
>> root [0]
>> Attaching file output_0002/ZmumuNp5_pt20.root as _file0...
>> root [1] .ls
>> TFile** output_0002/ZmumuNp5_pt20.root
>> TFile* output_0002/ZmumuNp5_pt20.root
>> KEY: TDirectoryFile histos;1 histos
>> KEY: TTree tau;17 physics
>> KEY: TTree tau;16 physics
>> KEY: TTree tmvatree;1 tmvatree
>> KEY: TTree tauinfo;1 tauinfo
>>
>>
>> [lxplus247] /tmp/boehlerm > root -l output_0003/ZmumuNp5_pt20.root
>> root [0]
>> Attaching file output_0003/ZmumuNp5_pt20.root as _file0...
>> root [1] .ls
>> TFile** output_0003/ZmumuNp5_pt20.root
>> TFile* output_0003/ZmumuNp5_pt20.root
>> KEY: TDirectoryFile histos;1 histos
>> KEY: TTree tau;17 physics
>> KEY: TTree tau;16 physics
>> KEY: TTree tmvatree;1 tmvatree
>> KEY: TTree tauinfo;1 tauinfo
>>
>>
>> the files are stored on the lxplus machine 247 in the tmp dir:
>> /tmp/boehlerm
>>
Received on Tue May 15 2012 - 19:57:43 CEST

This archive was generated by hypermail 2.2.0 : Tue May 15 2012 - 23:50:01 CEST