Hello Norman,

I don’t think we need to do a full new pass for a few events where the current version of the software throws an exception. Can’t we simply only re-process the files that failed before, once we have a solution on how to deal with those few events?

Unless of course these exceptions indicate that we are doing something something wrong and the fix changes the output in a real way.

Best,
Maurik


On Jan 19, 2018, at 9:12 PM, Graf, Norman A. <[log in to unmask]> wrote:

Hello Jeremy,

Your proposal sounds fine to me. We should merge the detector into the 4.0.1 release without any other code changes.

There are two outstanding issues (dphi and bordered band matrix) which throw uncaught Exceptions. But these are rare and have not yet been resolved (Matt, do you have any updates?). As inelegant as it may sound, I am proposing that we make another pass once we have patches for those two issues. I do not want to hold up production reconstruction of the 1.5mm data any longer. 

Norman


From: [log in to unmask] <[log in to unmask]> on behalf of McCormick, Jeremy I. <[log in to unmask]>
Sent: Friday, January 19, 2018 3:37 PM
To: Rafayel Paremuzyan
Cc: hps-software
Subject: Re: Pull request #285
 
Hi,

I moved this conversation to the software list…

If we really need to use the exact same version of recon from 4.0.1 with these updates, I propose the following:

- create a branch 4.0.2-pre from the 4.0.1 tag

- merge in the new detector update to this branch (can use the commit hash from the PR if the branch was deleted)

- merge in any other fixes/updates for 4.0.2 (are there any?)

- change hps-java version to 4.0.2

- make a 4.0.2 git tag from the branch

- build the java jars from the tag

Then we can run the 2015 pass from this jar which will contain the 4.0.1 recon with the new detector data.

I don’t think we really need a proper release for 4.0.2, though if needed it could be done from the command line if required (the default Jenkins build uses master for releases).

—Jeremy

On Jan 19, 2018, at 9:00 AM, Rafayel Paremuzyan <[log in to unmask]> wrote:

Right Norman, 

this is important to merge to the 4.0.1!!

Rafo


On 01/19/2018 11:54 AM, Graf, Norman A. wrote:
There have been changes to the master which will change the results. We should merge this detector into the branch that was used for 4.0.1 and tag that as 4.0.2.

Norman

From: McCormick, Jeremy I.
Sent: Thursday, January 18, 2018 2:32 PM
To: Graf, Norman A.
Cc: Maurik Holtrop; Graham, Mathew Thomas; Nathan Baltzell; Rafayel Paremuzyan
Subject: Re: Pull request #285
 
I can make a 4.0.2 release with this detector but I’m not sure that the physics results have not changed since 4.0.1 was tagged - can someone confirm/deny?

On Jan 18, 2018, at 2:20 PM, Graf, Norman A. <[log in to unmask]> wrote:

Hello Maurik,

Thanks for taking care of this. This will also need to be included in the production branch (4.0.2?). Could you please coordinate with Jeremy and Rafo?

Thanks,
Norman


From: Maurik Holtrop <[log in to unmask]>
Sent: Thursday, January 18, 2018 2:19 PM
To: Graf, Norman A.
Cc: Graham, Mathew Thomas; Nathan Baltzell; McCormick, Jeremy I.; Rafayel Paremuzyan
Subject: Re: Pull request #285
 
Hi Norman,

I approved the changes and then merged the code and deleted the branch.
Sorry for the delay. Thank you for pointing me to this.

Best,
Maurik


On Jan 18, 2018, at 2:20 PM, Graf, Norman A. <[log in to unmask]> wrote:

Hi Matt, Maurik,

What's holding up my pull request #285? This is on the critical path for running pass8 over the 1.5mm SVT data. I've tested it and Alessandra has signed off on it. Please let me know what you are waiting on.

Thanks,
Norman





Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1

Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1


Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1