«

Willem wrote:

> ) Why should pop return a value?
>
> For the same reason as push() takes two arguments.
>

And what reason is that?


Rui Maciel

Chris M. Thomasson wrote:

> That does create a "gap" between `top()' and `pop()'; no problem for
> single-threaded access...

And what leads you to believe that writing pop() so that it returns a value
will make the code immune to concurrency issues, or at least more vulnerable
than the version which doesn't return a value?


Rui Maciel

Rui Maciel <> writes:
> Willem wrote:
>> ) Why should pop return a value?
>>
>> For the same reason as push() takes two arguments.
>
> And what reason is that?

Convenience.

--
Keith Thompson (The_Other_Keith) kst- <http://www.ghoti.net/~kst>
Nokia
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"

"Rui Maciel" <> wrote in message
news:4dfa7ac7$0$23523$...
> Chris M. Thomasson wrote:
>
>> That does create a "gap" between `top()' and `pop()'; no problem for
>> single-threaded access...
>
> And what leads you to believe that writing pop() so that it returns a
> value
> will make the code immune to concurrency issues, or at least more
> vulnerable
> than the version which doesn't return a value?

NO IMMUNITY!


I simply prefer the conditional `try_pop()' function because it can be so
easily used by the inherent pattern contained within the almighty
"eventcount" algorithm:

http://groups.google.com/group/comp....f4655c4e0b7c95

;^)

Rui Maciel wrote:
) Willem wrote:
)
)> ) Why should pop return a value?
)>
)> For the same reason as push() takes two arguments.
)>
)
) And what reason is that?

I don't know. Ask the person who proposed the API upthread.

If you insist that pop() shouldn't return a value, then push()
should not take a value either, just a stack. Otherwise it
is plainly inconsistent.


SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT

Willem <> writes:
> Rui Maciel wrote:
> ) Willem wrote:
> )
> )> ) Why should pop return a value?
> )>
> )> For the same reason as push() takes two arguments.
> )>
> )
> ) And what reason is that?
>
> I don't know. Ask the person who proposed the API upthread.
>
> If you insist that pop() shouldn't return a value, then push()
> should not take a value either, just a stack. Otherwise it
> is plainly inconsistent.

pop() not returning a value makes some sense if you think that
shrinking the stack and retrieving the top element should be
separate operations.

push() without an argument would have to push an indeterminate or
default value onto the stack.

--
Keith Thompson (The_Other_Keith) kst- <http://www.ghoti.net/~kst>
Nokia
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"

Keith Thompson wrote:
) pop() not returning a value makes some sense if you think that
) shrinking the stack and retrieving the top element should be
) separate operations.

push() not taking a value makes some sense if you think that
expanding the stack and setting the top element should be
separate operations.

) push() without an argument would have to push an indeterminate or
) default value onto the stack.

pop() without a return value would have to discard the value it
takes from the stack.


SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT

In article <>,
Keith Thompson <kst-> wrote:

> Willem <> writes:
> > Rui Maciel wrote:
> > ) Willem wrote:
> > )
> > )> ) Why should pop return a value?
> > )>
> > )> For the same reason as push() takes two arguments.
> > )>
> > )
> > ) And what reason is that?
> >
> > I don't know. Ask the person who proposed the API upthread.
> >
> > If you insist that pop() shouldn't return a value, then push()
> > should not take a value either, just a stack. Otherwise it
> > is plainly inconsistent.
>
> pop() not returning a value makes some sense if you think that
> shrinking the stack and retrieving the top element should be
> separate operations.
>
> push() without an argument would have to push an indeterminate or
> default value onto the stack.

Maybe the names are just poorly chosen. I always think of pop()
returning the top value and shrinking the stack. I would use drop() to
just shrink the stack. A push() without arguments could just duplicate
the top of the stack, but then you'd be better off calling that dup().

Willem <> writes:
> Keith Thompson wrote:
> ) pop() not returning a value makes some sense if you think that
> ) shrinking the stack and retrieving the top element should be
> ) separate operations.
>
> push() not taking a value makes some sense if you think that
> expanding the stack and setting the top element should be
> separate operations.

*And* if you don't mind the new top of the stack having a meaningless
value. The two cases are not equivalent.

> ) default value onto the stack.
>
> pop() without a return value would have to discard the value it
> takes from the stack.

Of course. Sometimes that's what you want to do.

I'm not arguing that pop() *shouldn't* have a return value, just
that it's not entirely insane for it not to.

--
Keith Thompson (The_Other_Keith) kst- <http://www.ghoti.net/~kst>
Nokia
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"

Willem <> writes:

> Keith Thompson wrote:
> ) pop() not returning a value makes some sense if you think that
> ) shrinking the stack and retrieving the top element should be
> ) separate operations.
>
> push() not taking a value makes some sense if you think that
> expanding the stack and setting the top element should be
> separate operations.

This would be an extraordinarily idiosyncratic usage.

> ) push() without an argument would have to push an indeterminate or
> ) default value onto the stack.
>
> pop() without a return value would have to discard the value it
> takes from the stack.

pop() without a return value wouldn't actually have to take a value from
the stack; it just moves the pointer.

> SaSW, Willem