Re: Unwanted Recursion
- To: mathgroup at smc.vnet.net
- Subject: [mg120138] Re: Unwanted Recursion
- From: Simon <simonjtyler at gmail.com>
- Date: Sat, 9 Jul 2011 07:33:42 -0400 (EDT)
- Reply-to: comp.soft-sys.math.mathematica at googlegroups.com
Yeah, but NonCommutativeMultiply[a] and NonCommutativeMultiply[] do not simplify like you want them to. I'm not sure how Plus and Times do it... they might use a slightly nonstandard evaluation order - maybe you could customize the evaluation of your p expressions.
The only (simple) automated solution I can think of at the moment is to use
$Post = (# /. {p[a_] :> a, p[] -> 1} &); (* order is important *)
When I implemented some fermionic algebra (arXiv:1102.3043), I actually included these types of rules in my NonCommutativeMultiply sorting algorithm. It's not perfect (like a lot of my code), but since I always needed to canonicalize my expressions and this could be considered part of the canonicalization procedure, it seemed like an acceptable compromise.