You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We create a "subring" of the Symbolic "Ring" that only includes value objects without variables (like SymbolicConstantsRing) and has eager evaluation of relational operators.
Compare:
sage: s = SR('x') == 1
sage: s
x == 1
sage: s in SR
True
This could then be used for example with Polyhedron(..., backend='field').
Approach 1 to implement this is to wrap elements of SR in some wrapper class.
Approach 2 is to implement it through refactoring:
We refactor the Symbolic Ring so that there is a convenient parent class for symbolic expressions with normal Python comparison semantics instead of creating relations.
The magic behavior of comparison operators for SR elements, creating relations, will be implemented in a wrapper class.
Ideally, this will be unified with the separate implementation of this functionality in sage.numerical.linear_functions.
mkoeppe
changed the title
Wrapper class for symbolic equations, inequalities
Refactor SR through a wrapper class for symbolic equations, inequalities
Nov 2, 2020
We create a "subring" of the Symbolic "Ring" that only includes value objects without variables (like SymbolicConstantsRing) and has eager evaluation of relational operators.
Compare:
with:
This could then be used for example with
Polyhedron(..., backend='field')
.Approach 1 to implement this is to wrap elements of
SR
in some wrapper class.Approach 2 is to implement it through refactoring:
We refactor the Symbolic Ring so that there is a convenient parent class for symbolic expressions with normal Python comparison semantics instead of creating relations.
The magic behavior of comparison operators for
SR
elements, creating relations, will be implemented in a wrapper class.Ideally, this will be unified with the separate implementation of this functionality in
sage.numerical.linear_functions
.Here we do Approach 1.
CC: @mjungmath @egourgoulhon @DaveWitteMorris @yuan-zhou
Component: symbolics
Issue created by migration from https://trac.sagemath.org/ticket/30234
The text was updated successfully, but these errors were encountered: