Graveyard updates#855
Conversation
… the bounding box using vertex coords.
gonuke
left a comment
There was a problem hiding this comment.
Thanks @pshriwise - just a couple of maintainability questions
6336330 to
0a53e2a
Compare
|
Should we ever replace an existing graveyard? Should we just not add another to the set? |
I can see someone wanting to replace a spherical shell graveyard in case they're learning DAGMC and didn't realize how many triangles this might add to their model without having to re-tessellate the entire thing. Granted, some kind of utility program in the repo would be better than writing it yourself own, which is what they'd have to now. |
|
I'm ready to merge but want to give @makeclean a chance to either withdraw or amplify his concern? |
gonuke
left a comment
There was a problem hiding this comment.
Only one typo in a comment....
Co-authored-by: Paul Wilson <paul.wilson@wisc.edu>
|
Thanks @pshriwise |
Description
These changes update some behavior of the
DagMC::create/remove_graveyardfunction.create_graveyardcreate_graveyardcan now create a graveyard without acceleration data structures using the vertices of the model to create a bounding box as a basis for the graveyard volume.create_graveyardwill create an implicit complement if none exists so the graveyard volume can be inserted into the model topology correctly.create_graveyardwill only (re)construct acceleration data structures for the new graveyard volume and the implicit complement if these data structures were present when the method was initially called.remove_graveyardThere's a bug when calling
create_graveyardwithoverwrite == truewhen a graveyard is not present. Previously, this would fail when trying to remove a graveyard that isn't present. This adds a check at the beginning ofremove_graveyardto simply returnMB_SUCCESSif no graveyard is present.