Hi, I'm currently working with data sets in a TTree, which have to be processed by an analysis/simulation/etc. The output of these programs has to be written to a TTree together with the input data. At the moment this means to copy the existing tree to a new file and adding new branches to the tree. This works, but it also means copying large amounts of data from one file to another. So my first question is: is it in some way possible to add new branches to an existing tree on a file without changing the old branches? Another solution (which I'm actually using now) is to write only the new information to the new file, in a tree which could have the same name as the old one. In a later analysis one has to open both files and go through both trees in parallel. I would suggest to add a class similar to TChain, but not working with file chains, but parallel. Then it would be possible to process on input file by several independant analysis programs with individual output files. Later on, one could define such a file collection, which should than act as one file (e.g. when having trees with the same name in more than one file, the objects in these trees could be accessed by only one tree pointer). What do you think about it? Best regards, Volker -- Volker Hejny Tel: 02461/616853 ** Institut f. Kernphysik Fax: 02461/613930 ** ---------------------------------------------------------------- ** ** --- Forschungszentrum Juelich GmbH, D-52425 Juelich **
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:35 MET