ordering, I'm just suggesting that since there is an inconsistent
definition of for Sage sets, and since they get used frequently as
vertices, that we define atotalordering. You yourself admitted that
you can't think of a natural mathematical meaning for for arbitrary
sets...
If you
On 2010-Aug-04 14:51:21 -0700, Nils Bruin nbr...@sfu.ca wrote:
There are other areas of mathematics where gets used for proper
inclusion. A group theorist is going to be very surprised if for two
groups H,G, the expression H G is valid but does not mean H is a
subgroup of G.
I expect this is
On Fri, 6 Aug 2010 07:06:33 +1000, Peter Jeremy peterjer...@acm.org wrote:
On 2010-Aug-04 14:51:21 -0700, Nils Bruin nbr...@sfu.ca wrote:
There are other areas of mathematics where gets used for proper
inclusion. A group theorist is going to be very surprised if for two
groups H,G, the
I'm not sure what the point of a comparison function is if we don't
implement a total ordering. The main place cmp methods get used is in
sorting. If you have a large list of objects, and, for example, you
want to know whether X is in the list, you might find this out by
looping over the whole
On Aug 4, 4:23 am, Robert Miller rlmills...@gmail.com wrote:
I'm not sure what the point of a comparison function is if we don't
implement a total ordering. The main place cmp methods get used is in
sorting.
But it is a fact that Python has already abandoned total ordering
semantics for .
If
Nils,
So you want Sage Sets to implement a b to mean a is a subset of
b? I'll admit that that is reasonable, and it is a fact that it
follows Python convention. But I think that the Python convention is
bizarre, especially given how they implement sorting lists. I would
also rather the sort
Also, consider the fact that many Sage functions use return
sorted(output) to guarantee a consistent ordering of the output. What
you're advocating means that this wouldn't work in many cases...
On Wed, Aug 4, 2010 at 4:16 PM, Robert Miller r...@rlmiller.org wrote:
Nils,
So you want Sage Sets
Wow, I'm 100% certain I have nothing to add to this quandary, and
couldn't possibly know the correct answer/choice, only supply the link
to the relevant context Robert refers to:
http://docs.python.org/release/3.0.1/whatsnew/3.0.html#ordering-comparisons
Down the rabbit hole indeed.
D
On Aug
On Aug 4, 1:16 pm, Robert Miller r...@rlmiller.org wrote:
So you want Sage Sets to implement a b to mean a is a subset of
b? I'll admit that that is reasonable, and it is a fact that it
follows Python convention.
My initial reaction for this is affirmative: If python makes a choice,
then sage
On Wed, Aug 4, 2010 at 5:51 PM, Nils Bruin nbr...@sfu.ca wrote:
On Aug 4, 1:16 pm, Robert Miller r...@rlmiller.org wrote:
So you want Sage Sets to implement a b to mean a is a subset of
b? I'll admit that that is reasonable, and it is a fact that it
follows Python convention.
My initial
An alternative approach would be to realise that vertices need to be
stored in a sequence that has quick membership testing, for instance a
dictionary {'v1': 0, 'v2':1} etc. Depending on the type of access,
perhaps you would also need to keep the thing as a sequence
['v1','v2']. In magma,
I don't know how easy it is to make such a data structure efficient
and what the memory penalty is for having both hashing and ordering.
http://docs.python.org/py3k/whatsnew/3.1.html#pep-372-ordered-diction...
The reference implementation in the PEP(372) states O(n) for
deletion of keys.
Another except from the PEP: A version written in C could use a
linked list. The code would be more complex than the other two
approaches but it would conserve space and would keep the same big-oh
performance as regular dictionaries. It is the fastest and most space
efficient.
The trade-off
On Aug 4, 3:14 pm, Robert Miller r...@rlmiller.org wrote:
If Python jumped off a cliff...
Then Sage might as well :-). Or be reimplemented in CL.
So then you would rather have Sage sets give an error rather than sort
in a list? I can't imagine why this is a good thing. You seem to be
ignoring
On Aug 4, 3:44 pm, mda_ donmorri...@gmail.com wrote:
http://docs.python.org/py3k/whatsnew/3.1.html#pep-372-ordered-diction...
Thanks! That makes me want to use Python 3.1.
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
Nils,
First of all, I hope my distaste for the muggy weather hasn't made me
unbearable. 0:-)
When I was referring to things being much more difficult, I was
thinking of implementing a cmp() function that took arbitrary
hashables and defined a total ordering on all of them, consistent
regardless
Hopefully this all agrees with you, and if not, I guess I can start
learning Lisp...
My apologies for the cross-posting (I am not yet approved for sage-
flame)
http://www.buayacorp.com/wp-content/uploads/2007/10/john-mccarthy-poster1.jpg
(Above: A portrait, and tribute! ;-), to the
mda_ wrote:
Hopefully this all agrees with you, and if not, I guess I can start
learning Lisp...
My apologies for the cross-posting (I am not yet approved for sage-
flame)
http://www.buayacorp.com/wp-content/uploads/2007/10/john-mccarthy-poster1.jpg
The fact that you're doing it
The fact that you're doing it at all implies the imperative which, as
we all know, is not the way to program (this week).
http://en.wikipedia.org/wiki/Touché
my friend(?)
=)
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email
The problem of not providing a total ordering seems to come up
again and again in different situations. There are two cases now
already where it is clear that even in Python does not signify a
total ordering anymore:
- python complex numbers ( gives an error)
- python sets ( denotes proper
I think there lies the problem :
sage: mcl_v = mcl.vertices()
sage: mclc_v = mcl.complement().vertices()
sage: mcl_v == mclc.v
False
So they are not equal. Do they contain the same elements ?
sage: Set(mcl_v) == Set(mclc_v)
True
So it seems... But then, what does THAT mean ?
sage:
21 matches
Mail list logo
