get doctest coverage of monoids/monoid.py to 100%

[williamstein]

Base ticket - 90% doctest coverage - #12024

Apply:

• attachment: trac_12025.patch
• attachment: trac_12025_review-TS.patch

CC: @orlitzky @sagetrac-tmonteil

Component: doctest coverage

Keywords: Cernay2012

Author: William Stein

Reviewer: Michael Orlitzky, Thierry Monteil, Karl-Dieter Crisman

Merged: sage-5.0.beta5

Issue created by migration from https://trac.sagemath.org/ticket/12025

[williamstein]

comment:1

See the #12024 metaticket.

[williamstein]

Attachment: trac_12025.patch.gz

[williamstein]

Author: wstein

[orlitzky]

comment:4

sage: is_Monoid(ZZ)   # not technical math meaning of monoid 
False

Does this work for any set-like objects? Wouldn't it depend on the operation?

I think this gives the wrong impression (if I read it correctly) that the test fails due to a technicality rather than because the test simply doesn't make sense.

There are binary operations that make ZZ a monoid, right?

[williamstein]

comment:5

If anything, this test gives the right impression, showing that is_Monoid is only a typecheck, and not testing whether the object has some sort of monoid structure. There are hundreds of "is_"'s in Sage, and they are all typechecks. That's why you get a warning when you use them from the sage: prompt. This was discussed a lot on sage-devel years ago and sorted out.

[orlitzky]

comment:6

Exactly, I only have a problem with the comment, not the test itself. The "technical math meaning" of monoid has no bearing whatsoever on the result: it's a typecheck.

My suggestion would be no comment at all, or something like "ZZ does not inherit from Monoid_class."

[williamstein]

comment:7

Replying to @orlitzky:

Exactly, I only have a problem with the comment, not the test itself. The "technical math meaning" of monoid has no bearing whatsoever on the result: it's a typecheck.

My suggestion would be no comment at all, or something like "ZZ does not inherit from Monoid_class."

The comment is "# not technical math meaning of monoid". Perhaps a clearer comment would be:

   "The technical math meaning of monoid has no bearing whatsoever on the result: it's a typecheck which is not satisfied by ZZ since it does not inherit from Monoid_class."

-- William

[orlitzky]

comment:8

Replying to @williamstein:

The comment is "# not technical math meaning of monoid". Perhaps a clearer comment would be:

   "The technical math meaning of monoid has no bearing whatsoever on the result: it's a typecheck which is not satisfied by ZZ since it does not inherit from Monoid_class."

-- William

That's even better. In hindsight, I see that the original comment was intended read "this does not test whether or not ZZ is mathematically a monoid." I missed it the first time around, though, so better to make it explicit for the next guy.

[orlitzky]

Description changed:

--- 
+++ 
@@ -1 +1,4 @@
+I tried to update the patch, but it looks like it's based off a different monoid.py than I've got (4.8.alpha2).
 
+Do I need to apply another patch first?
+