[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Raymond Hettinger added the comment:
Thanks Terry.
And yes, your reading of the set.update() docs is correct, Update the set,
adding elements from all others means that it updates the set by adding the
elements from the other sets.
FWIW, getting into details about which value wins goes against the core
concept of the data structure. Sets (in a mathematical sense) are about
treating elements of an equivalence class as being identical. We intentionally
treat {1, 1.1} as being of length one and equal to {1.1, 1} eventhough the
integer 1 and the float 1.0 are actually distinguishable in ways not tested by
equality.
This isn't just a concept with sets. The dict.update() method works similarly
as does other container operations that depend on a notion of equivalence that
is independent of other distinctive traits (object identity, attached values,
type, etc).
--
resolution: - invalid
status: open - closed
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20902
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Terry J. Reedy added the comment:
Defining equality to ignore the .value attribute (and the id), says that they
*do not matter*. Python believes you and the interpreter does what it does. If
we had made a 'first or second operand wins' claim, we would not have been able
to optimize intersection by starting with the smaller set. I may mis-remember
but I believe we changed after someone noticed that something like {1]
set(range(1)) versus set(range(1)) {1} had very different run times.
I agree with Raymond that the docs for intersection and union should be left as
is.
I could read the doc for A.intersection_update(B) as saying that it keeps all
items in A that also appear in B, with appear defined on the basis of ==, so
that the result is a subset of the actual items in A. If that is the intent, it
could be made clearer.
The doc A.update(B) (union) come close to saying that all items in A remain and
items in B not in A get added, so that the result all the items in A. Again, if
that is the intent, it could be made explicit, and people who want to determine
which operand wins could use the update functions to do so.
--
nosy: +terry.reedy
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20902
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Giacomo Alzetta added the comment: I asked this because, for example, in Haskell it *is* a well-defined behaviour (see its documentation at: http://hackage.haskell.org/package/containers-0.5.4.0/docs/Data-Set.html): the left operand is preferred by all operations. In fact, working with Haskell, I also have used such behaviour in the past. For example when writing a simple type-checker it's quite convenient to use such behaviour when handling environments where you want inner blocks to hide outer variables. Not defining such behaviour means that you must re-implement the wheel in order to achieve the correct behaviour. In any case, this information should be made explicit in the docs, whether we want to make the behaviour defined or not. -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Changes by Raymond Hettinger raymond.hettin...@gmail.com: -- assignee: docs@python - rhettinger ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Raymond Hettinger added the comment: As far as I can tell currently there is no rule about this. Intersection prefers the second operand, while union prefers the first. The implementation uses the same logic as found in Lib/sets.py where the intersection operator loops over the SMALLER of the two input sets and does membership testing over the LARGER set. In any case, this information should be made explicit in the docs Not really. Historically, we've resisted the urge to over-specify or to declare something as undefined. In general, we make affirmative guarantees when they are useful and when we're prepared to make sure the guarantee will always hold (for example, it took a long while before sorts were guaranteed to be stable). Another consideration is that for most users, additional notes of behavior xxx is undefined is confusing, disconcerting, distracting, and rarely helpful. Language lawyers tend to request this sort of wording out of some sense of completeness, but it doesn't actually make the docs better for users. -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
New submission from Giacomo Alzetta:
Currently the documentation for set (at:
http://docs.python.org/2/library/stdtypes.html#set) does not mention which
operand is preferred when performing the usual binary operations.
For example the following sample code doesn't have a defined result according
to the documentation:
class MyObj:
def __init__(self, key, value):
self.key = key
self.value = value
def __key(self):
return self.key
def __hash__(self):
return hash(self.__key())
def __eq__(self, other):
return type(self) is type(other) and self.__key() == other.__key()
a = {MyObj(1, 'a')}
b = {MyObj(1, 'b')}
print((a b).pop().value) # 'a' or 'b'?
As far as I can tell currently there is no rule about this. Intersection
prefers the second operand, while union prefers the first. These are the only
operations where this is important since they include common elements.
I believe that this kind of information ought to be included explicitly near
such operations. At the very least, if the behaviour should really be somehow
undefined, it should be stated there so that people don't rely on such
behaviour.
The same should be stated for |= and =, since the first will not modify common
elements while the latter will.
PS: I can't find any resource about the formatting of issues (e.g. if and which
syntax is used for code blocks. Sorry for that.
--
assignee: docs@python
components: Documentation
messages: 213305
nosy: Giacomo.Alzetta, docs@python
priority: normal
severity: normal
status: open
title: Which operand is preferred by set operations? Missing information in the
documentation
type: enhancement
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20902
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
R. David Murray added the comment: If you take the intersection or union of a set, it shouldn't matter which set the element came from, by the axioms that apply to sets. So it if does, that's an application bug, IMO. -- nosy: +r.david.murray ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Changes by Éric Araujo mer...@netwok.org: -- nosy: +stutzbach ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
R. David Murray added the comment: Or, to put it another way, which set it comes from is not documented because it is an implementation detail and can not be depended on. I'm sure someone will correct me if I'm wrong :) -- nosy: +rhettinger ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Steven D'Aprano added the comment: If this is undefined, and I think it should be, the docs should explicitly say so. How is this? The union and intersection operations select elements which appear in both operands, e.g. set([a, b, c]) set([c, d, e]) returns set([c]). The selection is performed by equality, not object identity, and which specific object is returned (the c object from the first set or the second set) is an implementation detail and cannot be relied on. -- nosy: +stevenjd ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
R. David Murray added the comment: Sounds good to me. -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Changes by Balakrishnan B balakrishnan.er...@gmail.com: -- nosy: +Balakrishnan.B ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
[issue20902] Which operand is preferred by set operations? Missing information in the documentation
Changes by Benjamin Peterson bp+pyb...@benjamin-peterson.org: -- keywords: +easy ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue20902 ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
