Nope, it is not the same.
On lxplus I have 5.27.02.
When I initialize athena I have:
for Rel. 22.214.171.124.1 - 5.26.00f_python2.6
for Rel. 17.0.4 - 5.28.00g
So I guess ROOT has this auto dictionary loader since at least 5.27.02, but then why it does not work with 5.28.00g in athena Rel.17.0.4?
Actually I don’t care which athena release to use to process data with my script on the GRID. The only restriction - so that 2-dimensional vector could be read, all other stuff seems working fine since long ago. I would use some athena release related to ROOT 5.27.02 with pleasure, I simply do not see any release of this kind.
humm … apperently, it is working for vector<vector > for these 3 releases.
So the problem is likely slightly different and a dictionary provided by Athena is likely to be preventing the autodictionary generation. I recommend that you either use the Loader.C+ trick and use
gInterpreter->GenerateDictionary("vector<vector<float> >","vector"); which is available since v5.28.
yeah, I tried to roll-back to my previous configuration when I had loader.C with content
#pragma link C++ class vector<vector<float> >+;
So if I do now
It works as well if I do not do
.L loader.C+ but instead of it -
But the reason why I wanted to have it all in the same .c file is that it is much easier to submit this job to the GRID via prun. Now I need to figure out how to write a submitting script with “.L loader.C+” or with "gInterpreter->GenerateDictionary(“vector<vector >”,“vector”);"
because I don’t see anything like that in twiki.cern.ch/twiki/bin/viewauth/Atlas/PandaRun
You can have the call to GenerateDictionary has the first operation done in your script.
yep, already tried it - it does not work unfortunately
Same errors shown:
Error: Can't call vector<vector<float> >::at(trkpt5_trackNumber) in current scope myanalysis.c:699:
Possible candidates are...
(in vector<vector<float> >)
Error: non class,struct,union object at(trkpt5_trackNumber) used with . or -> myanalysis.c:699:
*** Interpreter error recovered ***
I mean I did
and the very first line in myanalysis.c is
So. We’ve found so far two ways to succeed here, but both require 2 commands being executed in the same ROOT session. And this is the thing I’m stuck with right now: it doesn’t seem obvious how to make it with prun.
[quote]and the very first line in myanalysis.c is[/quote]It should no be the very first line of the file,
it should be the very first line of the function myanalysis … (i.e. being the first first line is not valid C++).
I see now! It works! Thank you very much!
Tomorrow I’ll try to send this to the GRID - I’ll let you know about results.
Thanks again, it all works fine now in the GRID using prun!