#11316: Weighted degree term orders added
--------------------------------+-------------------------------------------
Reporter: klee | Owner: AlexGhitza
Type: enhancement | Status: needs_review
Priority: minor | Milestone: sage-4.7.1
Component: basic arithmetic | Keywords:
Work_issues: | Upstream: N/A
Reviewer: | Author: Kwankyu Lee
Merged: | Dependencies:
--------------------------------+-------------------------------------------
Comment(by SimonKing):
Concerning unpickling of old pickles, I think the following holds in this
case:
The pickling of a `TermOrder` is actually quite simple, standard Python.
It is an object with a `__dict__`, and a new un-initialised instance can
be created with `__new__`.
'''Example'''
First, we obtain an order:
{{{
sage: P.<x,y>=QQ[]
sage: T = P.term_order()
}}}
Now, if I am not mistaken, `S=loads(dumps(T))` essentially boils down to
the following.
First, we create a new instance of the class:
{{{
sage: S = T.__new__(T.__class__)
}}}
And then, we provide the new instance with a copy of the old
instance's `__dict__`:
{{{
sage: S.__dict__ = copy(T.__dict__)
}}}
As a result, we have a copy of T:
{{{
sage: S
Degree reverse lexicographic term order
sage: S == T
True
sage: S is T
False
}}}
If I understand correctly, the problem here is that your patch
introduces a new attribute `_weights`, that has a default value for
trivial degree weights. So, if `__dict__` was taken from an old pickle,
then it lacks the key `'_weights'`. Hence, requesting `S._weights` would
fail.
Two solutions were suggested:
1. Introduce a `__getattr__` method, that returns the default weights and
puts them into `S.__dict__['_weights']`. That ensures that
`S.__getattr__('_weights')` is called at most once: `__getattr__` is only
called if an attribute can not be found in the `__dict__` and not as a
class attribute.
2. Introduce the default `_weights` as a class attribute. If a term order
has non-default weights then they are put into `__dict__` (this is what
happens if you do `T._weights = [1,2,3]`). But attributes found in
`__dict__` have precedence over class attributes. Hence, the class
attribute will only be considered if `_weights` is not in `__dict__`.
It seems to me that in both cases unpickling of old pickles should just
work. One should test, though, whether one of the solutions has a speed
penalty. I reckon that introducing `__getattr__` may slow things down. How
often are attributes of a term order called when you do polynomial
arithmetic?
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11316#comment:15>
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 post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
