Skip to content
Snippets Groups Projects
Forked from atlas / athena
Source project has a limited visibility.
  • Zach Marshall's avatar
    db8f278f
    Moving to functions for tau truth kernel · db8f278f
    Zach Marshall authored
    This change has two related effects.  What was in TauTruthCommon is now
    wrapped up in a function.  All clients have been migrated to use the
    function form, and several clients were identified that don't actually
    need to be calling this function at all (they get it for free when using
    the common tools).
    
    The first good benefit is we avoid the `import *` syntax, and treat the
    module as a real module that has a function we can call rather than as a
    replacement for `include`.
    
    The second good benefit is that we can optionally pass this function an
    algSequence in which to schedule the truth tools.  At the moment, the
    truth code is the only client of this functionality, but it allows the
    scheduling of private instances of the kernels that could run (for
    example) after skimming choices have been made, rather than public
    kernels that are always a part of the main sequence.  This satisfies a
    request of the SUSY group, to ensure that the recommended workflow
    (private kernels) really is possible, even if it's not heavily used.
    
    Tested a TRUTH and EXO derivation to make sure that things are behaving
    after these changes, and everything seems to be behaving normally.  This
    change was not discussed at any length with the tau group, though, so it
    would be useful to have a comment from someone on that end before
    merging this request in.
    
    
    Former-commit-id: f3e5cae04c2ab2a533d4a25edcc0e3a91d14b0dd
    db8f278f
    History
    Moving to functions for tau truth kernel
    Zach Marshall authored
    This change has two related effects.  What was in TauTruthCommon is now
    wrapped up in a function.  All clients have been migrated to use the
    function form, and several clients were identified that don't actually
    need to be calling this function at all (they get it for free when using
    the common tools).
    
    The first good benefit is we avoid the `import *` syntax, and treat the
    module as a real module that has a function we can call rather than as a
    replacement for `include`.
    
    The second good benefit is that we can optionally pass this function an
    algSequence in which to schedule the truth tools.  At the moment, the
    truth code is the only client of this functionality, but it allows the
    scheduling of private instances of the kernels that could run (for
    example) after skimming choices have been made, rather than public
    kernels that are always a part of the main sequence.  This satisfies a
    request of the SUSY group, to ensure that the recommended workflow
    (private kernels) really is possible, even if it's not heavily used.
    
    Tested a TRUTH and EXO derivation to make sure that things are behaving
    after these changes, and everything seems to be behaving normally.  This
    change was not discussed at any length with the tau group, though, so it
    would be useful to have a comment from someone on that end before
    merging this request in.
    
    
    Former-commit-id: f3e5cae04c2ab2a533d4a25edcc0e3a91d14b0dd
Code owners
Assign users and groups as approvers for specific file changes. Learn more.