[Users] dependencies & CST warnings

Roland Haas roland.haas at physics.gatech.edu
Wed Jul 17 17:01:50 CDT 2013


Hello Bernard,

> Now these are warnings, not errors; perhaps the code would compile fine
> and run OK anyway (I haven't tried). But surely trying to schedule a
> routine in a particular group or timebin means that the thorn responsible
> for such groups/timebins should be an actual dependency? Why are Boundary
> and SymBase not *required* in the interface.ccl of CartGrid3D?
To my knowledge these routines implement the old-style boundary routines
for octant, quadrant, bitant symmetry (ie if cartgrid3d::domain is not
"full"). The thorn would run fine without and in fact using those
options is deprecated (in favor of ReflectionSymmetry which does require
SymBase which does require Boundary). Having said that, I would not
*want* to run without Boundary on grounds of not knowing which thorns
silently assumes it is present, but it is not required for CartGrid3D
(eg the wavetoysimple example does not use it).

Yours,
Roland


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 260 bytes
Desc: OpenPGP digital signature
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20130718/4836ed59/attachment.bin 


More information about the Users mailing list