[ET Trac] [Einstein Toolkit] #625: CarpetIOHDF5 sliced output does not output symmetry points unless it also outputs buffer points

[Einstein Toolkit]

#625: CarpetIOHDF5 sliced output does not output symmetry points unless it also
outputs buffer points
---------------------+------------------------------------------------------
  Reporter:  hinder  |       Owner:  eschnett    
      Type:  defect  |      Status:  new         
  Priority:  major   |   Milestone:              
 Component:  Carpet  |     Version:              
Resolution:          |    Keywords:  CarpetIOHDF5
---------------------+------------------------------------------------------

Comment (by rhaas):

 Tanja had similar issues to mine and after some poking around I think I
 now understand the reason for Tanja's and mine's problems. It turns out
 that the fix for that likely also fixes your problem, Ian. Attached please
 find a preliminary patch for CarpetIOHDF5. If you need these options to
 work right now, please go ahead and apply the patch otherwise I'll do some
 cleanup (eventually) and try to test some of the corner cases (and remove
 support for deprecated parameters from new code).

 Buffered and unbuffered output are now handled completely separately
 (which is ugly), with the old, buffered output removing symmetry and ghost
 points from the exterior if the are not requested, and the buffered output
 adding them to the active region if they are requested.

 As of late the VisIt reader actually uses the cctk_nghostzones information
 and passes it on the VisIt, so that we should not longer set the
 cctk_nghostzones attribute to a non-zero number when we remove the ghost
 points (and have to make sure that there are ghost points if we claim so).

-- 
Ticket URL: <https://trac.einsteintoolkit.org/ticket/625#comment:6>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit