Do not convert unsigned elements to InfinityRing

[dkrenn]

sage: InfinityRing(x)
A negative finite number
sage: InfinityRing("junk")
A positive finite number

Component: coercion

Author: Jeroen Demeyer

Branch/Commit: u/jdemeyer/do_not_convert_unsigned_elements_to_infinityring @ 93a039f

Reviewer: Ralf Stephan

Issue created by migration from https://trac.sagemath.org/ticket/17825

[jdemeyer]

Author: Jeroen Demeyer

[jdemeyer]

Branch: u/jdemeyer/do_not_convert_unsigned_elements_to_infinityring

[jdemeyer]

Commit: e16686d

[jdemeyer]

New commits:

e16686d

Do not convert unsigned elements to InfinityRing

[rwst]

comment:4

Does not apply.

[jdemeyer]

comment:5

Do you plan to review this ticket if I fix the conflict?

[rwst]

comment:6

Yes. I have asked others this question in the past. Perhaps some people use the code but don't want to review it...

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

93a039f

Do not convert unsigned elements to InfinityRing

[sagetrac-git]

Changed commit from e16686d to 93a039f

[rwst]

Reviewer: Ralf Stephan

[rwst]

comment:9

The sardonis patchbot has one relevant doctest fail. It appears that to determine RIF(-1,1) < oo it has to be coerced to the infinity ring but it has no sign. OTOH, RIF(-1,1) < oo is perfectly easily solvable so should not give an error.

[jdemeyer]

comment:10

Replying to @rwst:

OTOH, RIF(-1,1) < oo is perfectly easily solvable so should not give an error.

That doesn't really matter. The coercion model always coerces. And it is the coercion which fails in this case. That is not easy to fix.