Ahoj, Petre, hezke vanoce (Hi, Peter, Merry Christmas)
can't resist to show you a few examples of Rebol code (just for fun, don't
be afraid).
Ex1:
f: func [/local a][
do func [] [
a: ""
]
insert a "1"
]
Results:
>> f
== ""
>> f
== ""
interesting, isn't it?
Ex2:
a: [""]
b: copy a
change/only a "x"
a
b
>> a: [""]
== [""]
>> b: copy a
== [""]
>> change/only a "x"
== []
>> a
== ["x"]
>> b
== [""]
You see, that by changing A we didn't change it's copy B.
Ex3:
a: [""]
b: copy a
insert first a "x"
a
b
Results:
>> a: [""]
== [""]
>> b: copy a
== [""]
>> insert first a "x"
== ""
>> a
== ["x"]
>> b
== ["x"]
Now, even B looks changed. How did I manage to do it?
Ladislav
[EMAIL PROTECTED] wrote:
> Let's consider now another version of the Mystery Function that has
> been discussed in this thread.
>
> >> container: ["" "x"]
> == ["" "x"]
>
> This word just holds a block of data. The first element of the block
> is a reference to an empty string; the second element is a reference
> to a string of length 1.
>
Exuse me folks - I am far from being expert in language theories or
programming,
so I will represent "simple" view here.
I know we all try to find best way to describe the behavior, but I wonder in
following:
1) use language features to describe it's components:
- there is no 'reference nor 'reference? function, so?
- type? first container returns string!, so it is string and from that point
of
view should not be regarded anything else, dot.
- value? first container returns true, so the empty string itself has a
value
- >> source value?
value?: native [
"Returns TRUE if the word has been set."
value
]
hmm, "... if the word has been set." What word are we talking about here?
The
description for 'value? shoul be changed then.
So, as for Rebol itself, it fails to describe what "" and [] really are.
2) I've thought REBOL is about simplicity, not confusion. If there is a
problem
because of "" [] only, remove the problem!
What about one exception in REBOL - always create copy of "" [] by default
...
Wouldn't it help to remove all unintuitiveness we are talking about here?
Look
also at following:
->> container: ["" "x" ""]
== ["" "x" ""]
->> source same?
same?: native [
"Returns TRUE if the values are identical."
value1
value2
]
->> same? first container third container
== false
or
->> a: ""
== ""
->> b: ""
== ""
->> insert a "Ahoy"
== ""
->> print a
Ahoy
->> print b
->> print head b
So, from another point of view ""s reference to some empty strings, but not
the
same ones! If so, how's that 'a, used as a local in a function, initialised
to ""
in the body of function, refers to the same string?
example: func [/local a][a: "" insert a "1" print a]
In above example, 'a and 'b refer to empty strings, the different ones. So,
if 'a
is local in our function, each function call should refer to DIFFERENT ""
string.
Or indefinite extent in action here? I think not.
>From the Amiga land, I remember one motto: KISS - keep is simple, stupid.
And
simple means - the thing works the way most people expect it to work. Can we
afford such exceptions in the language? What about introduction of new
datatypes?
any-string, empty-string ... Could it help?
3) I just wonder, why are REBOL Tech. folks so silent? Why not enter the
discussion, introducing proper solution? I would expect it from language
creators.
As for me, I can live with a: copy "" or a: make string! "", but care should
be
taken of REBOL newcomers ...
Regards,
-pekr-
>
> -jn-
