Core Minutes 3/18/2014

• ScienceTools: (Jim) ScienceTools-09-33-00, requested by FSSC, exists. It includes all recent package tags except for work concerned with new Pass8 IRFs.

  there is a new Likelihood tag incorporating Johann's updates

  See Science Tools Development Notes for details.

• FSSC: (Joe)

  re:Mountain Lion builds, [See report from Tom S. below] we compile using the native xcode compiler (llvm5-gcc4.2) but I have not gotten the tools to sucessfully compile using any of the macports compilers.

  I've been working on some tip changes Jim requested to allow the reading and writing of bit array columns to fits files, and that's been coming along (I'm shooting to have that all ready to go by next week.)

  I've also been getting the ScienceTools release tag that Jim created ready here for prerelease testing.

• Reprocessing: (Tom G.) Status of redo of diffuse repsonse for Pass7 reprocessing is unchanged from last week: done through February; waiting for decision on transition date.

  (Warren) is working on renaming 1-second ft2s. They will be delivered as part of P203 and exported by future versions of L1.

  (Tom) Generation of merit and fits files for Pass8: various issues are still under discussion. One concerns tools — whether to use Gleam (as usual) or TMine.

• Core talk:

  • GR (Heather) is implementing ACD changes discussed last week: change default to not assume Overlay, and improve reporting of jo set-up problems.

  • WIRED (Heather) There continues to be a problem with a system library on fermilnx14. It's not missing (as previously thought) but the native version is newer than the one WIRED is looking for.

    She will update Dima's script, which currently specifies GR 18. There is no reason to be using this version which is not suitable either for Pass8 (too old) or L1. Dima should be using a version appropriate for current Pass8 development and perhaps should also be able to switch to something compatible with L1.

  • Mountain Lion (Tom S.) Tony never got a personal account but there is no need since all the Jenkins set-up is done.

    There are 41 different parts of the code giving compile errors. About a dozen can be fixed by judicious insertion of const qualifiers. Others appear to be conflict with the compiler versions used to build ROOT and python — but this is mysterious since he used the same compiler version to build everything! He is thinking of retreating to the compiler version used by FSSC for which it should all work, then pushing forward.

    (Heather) There are newer ROOT patch versions of 5.34 which we could try. Some of the updates were specifically to address Mac issues.

previous

minutes index

next

J. Bogart, Last Modified: