Fix failure during merge when histogram limits are invalid

[mark-grimes]

This should stop the relval failures for HGCal. The problem was that validation plots were being booked with equal minimum and maximum limits which causes root to fail when adding them together during merging. This adds a check during the merging to ignore histograms where the min is not less than the max. This does not fix the underlying problem, it merely stops it failing the whole merge.

The underlying problem is for the following plots:

HcalHitsV/SimHitsValidationHcal/HcalHitEtaHE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitEtaHE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitEtaHE2-z
HcalHitsV/SimHitsValidationHcal/HcalHitTimeAEtaHE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitTimeAEtaHE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitTimeAEtaHE2-z
HcalHitsV/SimHitsValidationHcal/HcalHitE100HE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitE100HE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitE100HE2-z
HcalHitsV/SimHitsValidationHcal/HcalHitE250HE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitE250HE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitE250HE2-z
HcalHitsV/SimHitsValidationHcal/HcalHitE25HE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitE25HE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitE25HE2-z
HcalHitsV/SimHitsValidationHcal/HcalHitE50HE0-z
HcalHitsV/SimHitsValidationHcal/HcalHitE50HE2+z
HcalHitsV/SimHitsValidationHcal/HcalHitE50HE2-z

It appears that the limits are set dynamically in SimHitsValidationHcal::getLimits and this is failing for HGCal and returning equal numbers.

@vandreev11, @pfs, @boudoul, @davidlange6 (also @lgray since you were on some thread about it).

[cmsbuild]

A new Pull Request was created by @mark-grimes (Mark Grimes) for CMSSW_6_2_X_SLHC.

Fix failure during merge when histogram limits are invalid

It involves the following packages:

DataFormats/Histograms

@Dr15Jones, @danduggan, @rovere, @cmsbuild, @nclopezo, @deguio, @ojeda, @ktf can you please review it and eventually sign? Thanks.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
@fratnikov, @mark-grimes you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[vandreev11]

@mark-grimes Thanks a lot, Mark.

[mark-grimes]

merge

I think the DQMIO request that just went in also has a check at a different point, but no harm in belt and braces.