#29836 closed defect (fixed)

normaliz backend isn't ready for generators

Reported by: gh-kliem Owned by:
Priority: critical Milestone: sage-9.2
Component: geometry Keywords: polytopes, dilation
Cc: jipilab, gh-LaisRast Merged in:
Authors: Jonathan Kliem Reviewers: Matthias Koeppe
Report Upstream: N/A Work issues:
Branch: 21f9d90 (Commits, GitHub, GitLab) Commit: 21f9d90bba887bbd2d734c41273d0f6a2f2596a5
Dependencies: Stopgaps:

Status badges

Description

sage: P = polytopes.simplex(backend='normaliz')
sage: K.<sqrt2> = QuadraticField(2)
sage: P.dilation(sqrt2)
Traceback (most recent call last):
...
AttributeError: 'generator' object has no attribute 'list'

The reason for this is simple. With optimization in #29200, it seems that generators are passed down to the normaliz backend and this isn't ready for this yet (when converting the data to the normaliz field).

Change History (18)

comment:1 Changed 10 months ago by gh-kliem

I just marked this as critical, because this is regression.

comment:2 Changed 10 months ago by gh-kliem

  • Branch set to public/29836
  • Commit set to cb13ddb433735cd25e92a870385b9f7faa2875f1
  • Status changed from new to needs_review

New commits:

cb13ddballow generators for Vrep/Hrep for backend normaliz

comment:3 Changed 10 months ago by chapoton

missing the # optional tag

comment:4 follow-up: Changed 10 months ago by chapoton

and you could use

isinstance(thing, (A, B, C))

comment:5 Changed 10 months ago by git

  • Commit changed from cb13ddb433735cd25e92a870385b9f7faa2875f1 to 002e5436d235e9a6cd5dbde1b04356e33ac31afe

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

002e543added optional tag; small improvement

comment:6 in reply to: ↑ 4 Changed 10 months ago by gh-kliem

Replying to chapoton:

and you could use

isinstance(thing, (A, B, C))

Much better. Thanks.

comment:7 Changed 10 months ago by mkoeppe

Probably Polyhedron_base should get some _test_... methods to make sure that methods are tested with all backends instead of relying on coverage by copy-paste doctests.

comment:8 Changed 10 months ago by gh-kliem

That would probably a good thing to do. I'm not familiar with TestSuite? at all, but that should probably make things better.

comment:9 Changed 10 months ago by mkoeppe

Take a look at sage.numerical.backends.GenericBackend. Some time ago I did a similar thing for the MIP backends there. There's also LoggingBackend, which provides a semiautomatic way to make _test... methods from existing doctests.

comment:10 Changed 10 months ago by gh-kliem

Thank you for the reference. I will do this, but better in a separate ticket.

comment:11 Changed 10 months ago by gh-kliem

I opened #29842 for this.

comment:12 Changed 10 months ago by mkoeppe

  • Reviewers set to Matthias Koeppe
  • Status changed from needs_review to positive_review

comment:13 Changed 10 months ago by gh-kliem

Thank you.

comment:14 Changed 10 months ago by fbissey

  • Status changed from positive_review to needs_work

In

+            sage: P = polytopes.simplex(backend='normaliz')  # optional - pynormaliz
+            sage: K.<sqrt2> = QuadraticField(2)              # optional - pynormaliz
+            sage: P.dilation(sqrt2)

the last line needs to be optional as well. If you don't have pynormalize you get a nice NameError: name P is not defined message in your doctests.

comment:15 Changed 10 months ago by git

  • Commit changed from 002e5436d235e9a6cd5dbde1b04356e33ac31afe to 21f9d90bba887bbd2d734c41273d0f6a2f2596a5

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

21f9d90optional tag again

comment:16 Changed 10 months ago by gh-kliem

  • Status changed from needs_work to needs_review

Thanks for catching that.

comment:17 Changed 10 months ago by mkoeppe

  • Status changed from needs_review to positive_review

comment:18 Changed 10 months ago by vbraun

  • Branch changed from public/29836 to 21f9d90bba887bbd2d734c41273d0f6a2f2596a5
  • Resolution set to fixed
  • Status changed from positive_review to closed
Note: See TracTickets for help on using tickets.