Root Event causes crash in develop

[Laurie Nevay]

Using root event output affects the read out geometry somehow causing a crash at its instantiation. Without recompiling, the code works fine with other forms of output. This is quite surprising as the output should not affect the geometry creation or sensitivity.

libc++abi.dylib: terminating with uncaught exception of type std::out_of_range: basic_string
BDSIM is about to crash or was interrupted! 
With signal: Abort trap: 6
Trying to write and close output file
Terminate run
BDSRunAction::EndOfRunAction> Run 0 end. Time is Mon Apr 11 11:49:41 2016

Run Duration >> 0 s
Ave, Imperator, morituri te salutant!
Abort trap: 6

[Laurie Nevay]

bdsim --file=sm.gmad --output=rootevent --batch --ngenerate=10 --outfile=t1

Will fail. Wheras

bdsim --file=sm.gmad --output=root --batch --ngenerate=10 --outfile=t1

will run ok.

[Jochem Snuverink]

Can't reproduce it, nor the test simpleMachineROOTEvent that tests almost exactly this (just less events): http://abp-cdash.web.cern.ch/abp-cdash/testSummary.php?project=11&name=simpleMachineROOTEvent&date=2016-04-11

My guess is that there might be a memory corruption somewhere.

[Laurie Nevay]

After restart also can' t reproduce. On hold for now.

[Jochem Snuverink]

We have seen similar crashes in the last months, and it was always due to some build products not being rebuilt after a pull or branch switch. This should be improved now. Closing the issue.