error when using `FlatMetric->True` with xTras

52 views
Skip to first unread message

Daniel Mahler

unread,
Jun 23, 2016, 12:49:53 AM6/23/16
to xAct Tensor Computer Algebra
Defining a metric with FlatMetric->True throws an error when xTras is loaded.
For example:

Needs["xAct`xTras`"]
DefManifold[M,4,IndexRange[a,n]]
DefMetric[-1,met[-a,-b],CD, FlatMetric->True]

produces the following:

** DefManifold: Defining manifold M. 
** DefVBundle: Defining vbundle TangentM. 
** DefTensor: Defining symmetric metric tensor met[-a,-b]. 
** DefTensor: Defining antisymmetric tensor epsilonmet[-a,-b,-c,-d]. 
** DefTensor: Defining tetrametric Tetramet[-a,-b,-c,-d]. 
** DefTensor: Defining tetrametric Tetramet\[Dagger][-a,-b,-c,-d]. 
** DefCovD: Defining covariant derivative CD[-a]. 
** DefTensor: Defining vanishing torsion tensor TorsionCD[a,-b,-c]. 
** DefTensor: Defining symmetric Christoffel tensor ChristoffelCD[a,-b,-c]. 
** DefTensor: Defining vanishing Riemann tensor RiemannCD[-a,-b,-c,-d]. 
** DefTensor: Defining vanishing Ricci tensor RicciCD[-a,-b]. 
** DefTensor: Defining vanishing Ricci scalar RicciScalarCD[]. 
** DefTensor: Defining vanishing Einstein tensor EinsteinCD[-a,-b]. 
** DefTensor: Defining vanishing Weyl tensor WeylCD[-a,-b,-c,-d]. 
** DefTensor: Defining vanishing TFRicci tensor TFRicciCD[-a,-b]. 
** DefTensor: Defining vanishing Kretschmann scalar KretschmannCD[]. 
** DefTensor: Defining vanishing symmetrized Riemann tensor SymRiemannCD[-a,-b,-c,-d]. 
** DefTensor: Defining vanishing symmetric Schouten tensor SchoutenCD[-a,-b]. 
** DefTensor: Defining symmetric cosmological Schouten tensor SchoutenCCCD[LI[_],-a,-b]. 
** DefTensor: Defining symmetric cosmological Einstein tensor EinsteinCCCD[LI[_],-a,-b]. 
MakeRule::error: LHS of rule evaluates to zero.
** DefTensor: Defining weight +2 density Detmet[]. Determinant.


This does not happen if only xTensor is loaded.

thanks
Daniel





Teake Nutma

unread,
Jul 27, 2016, 3:51:54 PM7/27/16
to xAct Tensor Computer Algebra
Hi Daniel,

Thanks for reporting this. This is indeed a bug -- I forgot to check for vanishing curvature when defining some additional curvature relations. I've fixed it on Github (see https://github.com/xAct-contrib/xTras/commit/48018f2d3e5d4fb9ffd13c8eed299b156a36d56a), but it might take a while to propagate to a point release due to the (extremely) slow release cycle of xTras these days.
Best,

Teake
Reply all
Reply to author
Forward
0 new messages