I was hoping to be able to manipulate vector objects in lieu of arrays of components. When you work on a component basis, it's gets hard to recognize cross and dot products when there chain of dots and crosses. It seems like they could be another symbol class that has a slightly different set of rules. I wonder if sagemath has something, although using sagemath seems to be way more complicated...
I am using the mechanics pkg (it's awesome!). I going through a text on Kane's method and trying to do every problem in the book with sympy. I run into difficulties with some of the proofs where you have to expand multiple dot product, cross products, and triple products. Currently I default to a tablet and just do it by hand. I can't imagine grinding on all the algebra by hand...you'd need to take my belt and shoelaces... On Monday, July 21, 2025 at 12:08:47 PM UTC-4 [email protected] wrote: > Yes, I think the stuff in mechanics is the best way to achieve what > you are looking for right now. It might be good to try to unify some > of this stuff at some point, since much of what is in mechanics isn't > really specific to mechanics or even physics. > > Aaron Meurer > > On Sat, Jul 19, 2025 at 10:44 AM Peter Stahlecker > <[email protected]> wrote: > > > > In sympy.physics.mechanics, I can create, say, three point P1, P2, P3 > > make > > u = P1.pos_from(P2) > > v = P2.pos_from(P3) > > > > and can get > > w = u.cross(v) > > > > but maybe I am missing a point. > > > > [email protected] schrieb am Samstag, 19. Juli 2025 um 18:10:36 UTC+2: > >> > >> I would also suggest using matrix expressions for this. The main issue > >> is that matrix expressions currently don't have a Cross operation. It > >> wouldn't be too hard to make one by subclassing MatrixExpr. I would > >> suggest opening an issue about this. > >> > >> The diff thing is another issue because currently there isn't a way to > >> represent functions in the matrix expressions > >> https://github.com/sympy/sympy/issues/5855 > >> https://github.com/sympy/sympy/issues/19265 > >> > >> Aaron Meurer > >> > >> On Sat, Jul 19, 2025 at 1:24 AM Jason Moore <[email protected]> wrote: > >> > > >> > I do not think there is a VectorSymbol type of object. If there was > you could imagine doing things like: > >> > > >> > R = VectorSymbol('R', 3) > >> > r = VectorSymbol('r', 3) > >> > R.diff(t).cross(r.diff(t)) > >> > > >> > We do have a MatrixSymbol object that may do what you desire. You can > create nx1 column matrices and then do such operations. > >> > > >> > Jason > >> > moorepants.info > >> > +01 530-601-9791 <(530)%20601-9791> > >> > > >> > > >> > On Thu, May 29, 2025 at 11:28 PM Frank Senkel <[email protected]> > wrote: > >> >> > >> >> Does sympy handle vector expressions vs instances of specific > vectors?..so AxBxC vs [a1,a2,a3] x [b1, b2, b3]...something like this. > >> >> > >> >> -- > >> >> You received this message because you are subscribed to the Google > Groups "sympy" group. > >> >> To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> >> To view this discussion visit > https://groups.google.com/d/msgid/sympy/b22199b5-810b-4371-8479-eabd71e30d0dn%40googlegroups.com > . > >> > > >> > -- > >> > You received this message because you are subscribed to the Google > Groups "sympy" group. > >> > To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> > To view this discussion visit > https://groups.google.com/d/msgid/sympy/CAP7f1AiS2juGowfs0%3DYQcyR4mgaDZsHjRoZtwXAWEToi2r8nvg%40mail.gmail.com > . > > > > -- > > You received this message because you are subscribed to the Google > Groups "sympy" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > > To view this discussion visit > https://groups.google.com/d/msgid/sympy/70440eaa-7bc7-4ab8-a5d6-06bc7abf53f9n%40googlegroups.com > . > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/sympy/043b2623-f30c-45e5-bf4b-540b32652bd6n%40googlegroups.com.
