Porting changes for beaudett: was pullID 120

[deguio]

including in the pull request 120 the last changes to the files:
Validation/RecoEgamma/python/egammaValidation_cff.py
Validation/RecoEgamma/plugins/PhotonPostprocessing.cc

which ware lost during the porting to GIT

[deguio]

@beaudett
I have rejected the pull request 120 and merged your changes with the updates to the files in [1] in the new pull request 156.

Since I am new to git, please have a look and cross check that everything is correct.

thanks,
F.

[1]
Validation/RecoEgamma/python/egammaValidation_cff.py
Validation/RecoEgamma/plugins/PhotonPostprocessing.cc

[cmsbuild]

The following categories have been signed by @deguio: DQM

@cms-git-dqm

[beaudett]

Hello,

I am taking a look right now. I am new to git as well and I'll do my best.

Thanks
Florian

On 21/7/13 6:07 PM, deguio wrote:

@beaudett https://github.com/beaudett
I have rejected the pull request 120 and merged your changes with the
updates to the files in [1] in the new pull request 156.

Since I am new to git, please have a look and cross check that
everything is correct.

thanks,
F.

[1]
Validation/RecoEgamma/python/egammaValidation_cff.py
Validation/RecoEgamma/plugins/PhotonPostprocessing.cc

—
Reply to this email directly or view it on GitHub
#156 (comment).

[beaudett]

Hello,
[adding Lindsey]

it runs fine, and it does what it is supposed to do.
Looking more carefully, I have however some doubts about a change in
RecoParticleFlow/PFProducer/src/PFEGammaAlgo.cc
that you copied from he:
deguio@a4eaf31

Lindsey, from the pull request page, it seems that you had anticipated
such possible problem. Should be go ahead this way ?

Florian

On 21/7/13 6:07 PM, deguio wrote:

@beaudett https://github.com/beaudett
I have rejected the pull request 120 and merged your changes with the
updates to the files in [1] in the new pull request 156.

Since I am new to git, please have a look and cross check that
everything is correct.

thanks,
F.

[1]
Validation/RecoEgamma/python/egammaValidation_cff.py
Validation/RecoEgamma/plugins/PhotonPostprocessing.cc

—
Reply to this email directly or view it on GitHub
#156 (comment).

[ktf]

@deguio, have you pushed the stuff you reverted?

[cmsbuild]

The following categories have been rejected by @vadler: Analysis

@cms-git-analysis

[cmsbuild]

The following categories have been rejected by @vadler: Analysis

@cms-git-analysis

[vadler]

I rejected it for the moment, since the changes at DataFormat level (renaming of data members) are not propagated to PAT and beyond. I will assist @beaudett to get this done, based on https://hypernews.cern.ch/HyperNews/CMS/get/swReleases/3922/1/1/1.html.
The MCMatcher issue in David's log file should be uncorrelated to that.

[vadler]

Finally, things become clearer, thanx to Chris and Philippe. To summarise it here:

• The MCMatcher issue is a consequence of the changes in this topic.
• It seems to a ROOT bug.

Details can be found in https://hypernews.cern.ch/HyperNews/CMS/get/edmFramework/3152.html ff.
I will stay with my rejection for the moment, since there is nothing, AT can do about it. If this topic should go in now, the ROOT ioread directives have to be removed and backward incompatibility accepted for the time being, until the ROOT fix becomes available.

[Dr15Jones]

Here is the issue posted by Philippe Canal to the ROOT bug tracking system

https://sft.its.cern.ch/jira/browse/ROOT-5437

[beaudett]

Hello,

I don't know the policy about the backward incompatibility in this
release. Given that it is the third attempt to get this change in, and
each time it is submitted it requires ~120 packages to be recompiles, I
hope that the (temporary) backward incompatibility is acceptable at this
stage of the release building.
Slava & Thomas, please let us know.

Cheers,
Florian

On 20/8/13 11:58 PM, Volker Adler wrote:

Finally, things become clearer, thanx to Chris and Philippe. To
summarise it here:

• The MCMatcher issue /is/ a consequence of the changes in this topic.
• It seems to a ROOT bug. Details can be found in
  https://hypernews.cern.ch/HyperNews/CMS/get/edmFramework/3152.html
  ff. I will stay with my rejection for the moment, since there is
  nothing, AT can do about it. If this topic should go in /now/, the
  ROOT |ioread| directives have to be removed and backward
  incompatibility accepted for the time being, until the ROOT fix
  becomes available.

—
Reply to this email directly or view it on GitHub
#156 (comment).

[Dr15Jones]

The problem has been fixed in ROOT

@ktf Giulio, can we get this ROOT patch in?
https://sft.its.cern.ch/jira/browse/ROOT-5437

[ktf]

@davidlt is looking into this right now.

[ktf]

@davidlt ping?

[davidlt]

It's in CMSSW_7_0_0_pre3.

[ktf]

Ok, perfect.@vadler can you try this again in today IB and see if it works now?

[vadler]

It does :-)
However, the file
CommonTools/ParticleFlow/python/Isolation/pfIsolatedElectrons_cfi.py
has been forgotten to be updated to the new GsfElectron interface in the topic, which breaks PFBRECO and the corresponding PAT integration tests.
I leave this one rejected and have re-queued it including the necessary fix as #666 (yes, the number of the beast!!!)
So, in principle this one can be closed.

[nclopezo]

Superseded by #680