Limitation on the number of event for hepevt

[BrieucF]

Hi, what was the reason for this limitation to 1 M particles with the hepevt format:

DD4hep/DDG4/plugins/Geant4EventReaderHepEvt.cpp

Lines 186 to 189 in 665f7c4

	if( NHEP > 1e6 ){
	printout(ERROR,"EventReaderHepEvt::readParticles","Cannot read in more than million particles, but %d requested", NHEP );
	return EVENT_READER_EOF;
	}

[andresailer]

Probably this

DD4hep/DDG4/plugins/Geant4EventReaderHepEvt.cpp

Lines 183 to 184 in 665f7c4

	//check loop variable read from input file and chack that is reasonable
	// should fix coverity issue: "Using tainted variable NHEP as a loop boundary."

[andresailer]

Hi @BrieucF

Did this answer your question? Would you like to increase this limit?

[BrieucF]

Hi Andre, I am afraid I do not fully understand the meaning of the two comments you quoted... If that limit is just to remain within what is deemed reasonable, yes I would like to increase it. Some beam induced background simulation provide a lot of particles, even with some filtering, but the simulation can still be run in a decent amount of time. If there is a deeper reason for this limit then we will have to find workarounds.

[andresailer]

Coverity static Analysis complained that we read some value from a file and without checking used that as a loop boundary, so someone put a limit that seemed reasonable at the time. If you think a larger number is also reasonable we can increase the limit.

Regardless of the limit, time and memory for simulating one event should be kept in mind.

[BrieucF]

Indeed, time and memory should be kept in mind. But providing "sane" input file is somehow the user responsibility. I also do not see such limit for e.g. the HepMC3 reader. So yes, if a limit has to be kept for hepevt, it would be nice to increase it to e.g. 50M. Otherwise, I am also happy if we just remove this.

[andresailer]

PR to increase the limit is welcome!

HepMC3 does its own validation, or is at least outside our purview.

[BrieucF]

Here it is: #1359