Listing an infinite vector space just hangs

[rbeezer]

Asking for a list of elements in an infinite vector space does not raise an error, but requires a Ctrl-C to get back the command-line.

sage: V=QQ^2
sage: V
Vector space of dimension 2 over Rational Field
sage: V.list()
<hang>

Of course, finite examples work properly:

sage: V=GF(3)^2
sage: V.list()
[(0, 0), (1, 0), (2, 0), (0, 1), (1, 1), (2, 1), (0, 2), (1, 2), (2, 2)]

Maybe this should be addressed in the module code, I have not looked into the cause much. Perhaps a fix could be as simple as having a non-zero dimension with an infinite base ring throwing an error?

---

Apply both patches, attachment: trac_10470-listing-infinite-sets.patch and attachment: trac_10470-referee-doc.patch.

Component: linear algebra

Author: Rob Beezer

Reviewer: John Palmieri

Merged: sage-4.7.alpha4

Issue created by migration from https://trac.sagemath.org/ticket/10470

[rbeezer]

comment:1

It seems ZZ and QQ and (maybe others) have never-ending iterators. So an attempt to iterate over modules like ZZ^2 or QQ^2 just keep making linear combinations like crazy, but never coming to any finite conclusion. Testing a fix now.

[rbeezer]

Attachment: trac_10470-listing-infinite-sets.patch.gz

[rbeezer]

Author: Rob Beezer

[rbeezer]

comment:4

This was discussed further on sage-devel:

http://groups.google.com/group/sage-devel/browse_thread/thread/3bb0f163b1fb1020

Patch just checks to see if a parent knows that is infinite, and if so, bails out with an explanation. Otherwise behavior should be the same. Passes the full test suite.

[jhpalmieri]

Attachment: trac_10470-referee-doc.patch.gz

apply on top of other patch

[jhpalmieri]

Reviewer: John Palmieri

[jhpalmieri]

comment:5

I'm happy with the main patch: it gets a positive review. I've added a little documentation about using X.list() vs. list(X). If it's not appropriate, let me know and we can just go with the main patch.

[rbeezer]

comment:6

Additional documentation looks good.

[jdemeyer]

Merged: sage-4.7.alpha4