#11895: Make p-adic numbers unhashable
-------------------------------------+-------------------------------------
Reporter: mmasdeu | Owner: roed
Type: defect | Status: needs_work
Priority: critical | Milestone: sage-6.4
Component: padics | Resolution:
Keywords: p-adic, hash | Merged in:
Authors: | Reviewers:
Report Upstream: N/A | Work issues: one pickle from the
Branch: | pickle jar does not unpickle
u/saraedum/ticket/11895 | correctly
Dependencies: #11670, #15897, | Commit:
#15898, #15956, #16122, #16124, | 42afa4d47da18d38095cd54313731f856ae5f083
#16129, #16250, #16251, #16316, | Stopgaps: todo
#16317, #16318, #16321, #16339, |
#16341, #16342 |
-------------------------------------+-------------------------------------
Comment (by saraedum):
Replying to [comment:45 vbraun]:
> But IMHO thats more of an issue with having two separate and competing
ways of specifying precision, one in the parent and one in the element
Is it really about competing ways? I guess this happens because it is
possible to specify precision in the element at all.
> Replying to [comment:44 saraedum]:
> > I agree that disabling hashing has its drawbacks but I do not see an
alternative.
> I don't have a good answer right now either, but it once again shows the
importance of this issue.
> We need to fix our failure to adhere to Python's convention about
comparison and hashes.
The problem is that sometimes you want `a==b` to mean, `a` is essentially
the same as `b`, say in an algorithm that doese `while(x!=0)`, and
sometimes to you want it to mean `a` is indistinguishable from `b`. The
latter is maybe what you want in dictionary lookups. But not always, for
example if you have a dict which maps powers of `x`, the variable in a
polynomial ring, to whatever, then you probably do not care about the
precision of the `1` coefficient.
So what I'm trying to say is that precision is tricky. Sure, many
algorithms might just throw an exception if we disable hashing. But would
they really work correctly otherwise? Whenever we make a dictionary lookup
with elements with precision, it seems that we need to decide on a case by
case basis which version of `==` would produce the right result. Throwing
an exception and requiring people to explicitly specify what they want to
happen seems to be a better solution IMO.
--
Ticket URL: <http://trac.sagemath.org/ticket/11895#comment:47>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/d/optout.
