_sympy_ methods for Set_object_binary subclasses

[mkoeppe]

Formal unions, intersections, ...

Depends on #32013

CC: @tscrim @kcrisman @egourgoulhon

Component: interfaces

Author: Matthias Koeppe

Branch/Commit: 689986b

Reviewer: Travis Scrimshaw, Michael Jung

Issue created by migration from https://trac.sagemath.org/ticket/32015

[mkoeppe]

Description changed:

--- 
+++ 
@@ -1 +1,2 @@
+Formal unions, intersections, ...
 

[mkoeppe]

Branch: u/mkoeppe/_sympy__methods_for_set_object_binary_subclasses

[mkoeppe]

Commit: 3dd50e4

[mkoeppe]

Author: Matthias Koeppe

[mkoeppe]

Last 10 new commits:

6e5cac6	sage.interfaces.sympy_wrapper, Sets.ParentMethods._sympy_: New
3cac256	sage.interfaces.sympy_wrapper: Add doctests
eef604e	SageSet: Finish docstrings; handle symbolic _contains
2baae58	Sets.ParentMethods._sympy_: Call sympy_init
153b3e5	Merge #31938
c06c965	sage.interfaces.sympy_wrapper.SageSet: Add another doctest
f535127	Merge #31938
32cdd5c	Merge #31877
789dc05	Merge #31931
3dd50e4	sage.sets.set.Set_object: Add `_sympy_` methods to subclasses

[mkoeppe]

Changed dependencies from #32013, #31931, #31938 to #32013

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

200b1ef	Merge tag '9.4.beta4' into t/32013/initialize_a_set_from_a_convexset_base_instance
7f35aee	Merge #32013

[sagetrac-git]

Changed commit from 3dd50e4 to 7f35aee

[tscrim]

comment:6

I am not sure about caching the output of _sympy_? It creates a clone of the set and ties it to the Sage object. Can you give some reasoning for this?

[mkoeppe]

comment:7

If one makes one computation with the sympy version of a set, it's likely that another computation will be made -- so saving the result seemed a good idea.

[tscrim]

Reviewer: Travis Scrimshaw

[tscrim]

comment:8

I would expect someone to store said Sympy version if they were going to use it again. Granted, I expect this to not be a problem either way. My instinct is to err on the side of not caching, but I don't have a strong opinion on this. So you either change it or not; once decided, you can set a positive review.

[mkoeppe]

comment:9

Thanks! I'm keeping it as is.

[vbraun]

comment:10

[docpdf] 
[docpdf] Overfull \hbox (22.51231p
[docpdf] ! Missing { inserted.
[docpdf] <to be read again> 
[docpdf]                    _
[docpdf] l.1488 ...n{}in class providing the operators \(__
[docpdf]                                                   add__\) and \(__sub__\).
[docpdf] ? 
[docpdf] [809
[docpdf] ! Emergency stop.
[docpdf] <to be read again> 
[docpdf]                    _
[docpdf] l.1488 ...n{}in class providing the operators \(__
[docpdf]                                                   add__\) and \(__sub__\).
[docpdf] !  ==> Fatal error occurred, no output PDF file produced!