Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.soft-sys.math.mathematica > #15515
| From | Itai Seggev <itais@wolfram.com> |
|---|---|
| Newsgroups | comp.soft-sys.math.mathematica |
| Subject | Re: list mutability (very basic question) |
| Date | 2013-09-23 07:39 +0000 |
| Message-ID | <l1or8a$je6$1@smc.vnet.net> (permalink) |
| References | <20130921083830.751226A65@smc.vnet.net> |
| Organization | Time-Warner Telecom |
> I just realized that I am unclear on a very basic point: in what ways are
> lists mutable?
>
> Example:
> {0}[[1]]=1; (* a. this is an error *)
> x={0}
> x[[1]]=1 (* b. this is not an error *)
> ReplacePart[x, 1 -> 2] (* c. this is differently not an error *)
>
> http://reference.wolfram.com/mathematica/tutorial/ManipulatingListsByTheirI
>ndices.html It is important to understand that ReplacePart always creates a
> new list. It does not modify a list that has already been assigned to a
> symbol the way does.
There two levels here; what I'll call user-mutability and internal-mutability.
On a user level, Set modifies the (definition of) the variables it acts upon,
so after it has evaluated the variable will have a new value, and it returns
the new value of the variable. ReplacePart does not; it will first make a
copy, modify it, and then return that new expression. I think this is
relatively clear
The there's internal mutability, whether and when expressions can be shared and
mofied in place. This depends very sensitivily on your state. For example, if
a = {1,0}
b = a
b[[1]]= 2
Then a and b can point the same list after the 2nd input is evaluated, but not
after the third. So in this case it would likely generate a copy and b would
be updated to point to the new location. For a user perspective how, it is
still "b that was modified".
> I think this means that after b. (i.e., Set[Part[x,1],1]) that x still
> references the same location in memory. If so, why is a. an error? (I
> realize that someone will think that saying that {0} is not a L-value will
> be a helpful answer, and perhaps it should be, but I'm looking for a
> different angle on the answer.)
I'm confused by this comment. Do you think that 0=1 should not be an error? If
you think it should be an error (as it is in every language I'm familiar with),
then why should {0}[[1]]=1, which would essentially be the same thing, not be
an error?
--
Itai
Itai Seggev
Software Technology
Back to comp.soft-sys.math.mathematica | Previous | Next | Find similar | Unroll thread
Re: list mutability (very basic question) Itai Seggev <itais@wolfram.com> - 2013-09-23 07:39 +0000
csiph-web