[ET Trac] [Einstein Toolkit] #900: GRHydro: do not use the Slopelimiter function

Einstein Toolkit trac-noreply at einsteintoolkit.org
Mon May 21 17:29:13 CDT 2012


#900: GRHydro: do not use the Slopelimiter function
------------------------------------+---------------------------------------
  Reporter:  bmundim                |       Owner:  bmundim             
      Type:  defect                 |      Status:  review              
  Priority:  minor                  |   Milestone:  ET_2012_05          
 Component:  EinsteinToolkit thorn  |     Version:                      
Resolution:                         |    Keywords:  GRHydro slopelimiter
------------------------------------+---------------------------------------

Comment (by rhaas):

 The problem with relying on ParamCheck is what happens when a new slope
 limiter is introduced in param. ccl (not very likely admittedly). Without
 an aborting "else" case, the new option would be silently ignored.
 ParamCheck will easily catch removed options but will not help you (the
 programmer) if you would like to add a new option. I see it analogous to
 always having a "default:" case in a switch statement even if all it does
 is abort (with maybe the possible exception if the same test was done in
 the same routine close by before. But even then, what happens if this
 piece of code is copied or lot's of other code added in between?).

-- 
Ticket URL: <https://trac.einsteintoolkit.org/ticket/900#comment:6>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit


More information about the Trac mailing list