Re: Question on the behavior of length()

[arnold]

And because this is purely an implementation issue, the details
are purposely not documented, since implementation mechanisms
are always subject to change.

Christian, I will answer your original mail, probably tomorrow, I
am short of time today.

Arnold

"Andrew J. Schorr" <aschorr@telemetry-investments.com> wrote:

> Hi,
>
> Internally, associative arrays are stored in different ways depending on the
> nature of the indices. There are currently 3 array implementations: string
> arrays, integer arrays, and arays with (mostly) consecutive non-negative 
> integer
> indices.
>
> Regards,
> Andy
>
> On Mon, Dec 11, 2023 at 02:06:38PM +0100, Wolfgang Laun wrote:
> > That's interesting; I must have missed this in the documentation. - What
> > causes the switch from fast to slow? Would an index such as ToS = -42
> > result in a speed reduction? What happens if a string as index is
> > introduced late in the game? Will all accesses to integer keys be slowed
> > down?
> > 
> > Also, you wrote "makes *the script* run about 3 times slower". Is it true
> > that these accesses to a single array affect the entire script?
> > 
> > -W
> > 
> > On Mon, 11 Dec 2023 at 13:55, Ed Morton <mortoneccc@comcast.net> wrote:
> > 
> > > Using a string `"tos"` instead of a number `0` for the top of stack index
> > > makes the script run about 3 times slower. You can always do something 
> > > like
> > > `Tos=0` and `x[Tos]` if you like.
> > >
> > >      Ed.
> > >
> > > On 12/11/2023 6:45 AM, Wolfgang Laun wrote:
> > >
> > > Call it x["tos"] and x["doc"] to avoid me being confused and having to
> > > remember what is what.
> > > -W
> > >
> > > On Mon, 11 Dec 2023 at 13:38, Ed Morton <mortoneccc@comcast.net> wrote:
> > >
> > >> Just a suggestion regarding calling `length()` for push()/pop() stack
> > >> operations:
> > >>
> > >> On 12/10/2023 8:18 AM, Christian Schmidt wrote:
> > >> > Hi all,
> > >> >
> > >> > First of all I'd like to state that I do not consider the following
> > >> > necessarily a bug; however I'd like to discuss the current
> > >> > implementation, and have not found a better place to do so.
> > >> >
> > >> > My issue is that I can't use
> > >> >
> > >> > x[length(x)] = y
> > >> >
> > >> > e.g. to emulate to push to a stack, without explicitly converting x
> > >> > into an array first, e.g. by "delete x".
> > >>
> > >> No need to call `length(x)` for every push()/pop(), you can implement a
> > >> stack by storing it's top index in `x[0]` (or some other unused,
> > >> probably negative, index instead if you prefer):
> > >>
> > >>       function push(x,y) { x[++x[0]] = y }
> > >>       function pop(x) { delete x[x[0]--] }
> > >>
> > >> As well as not having the problem you described that also allows you to
> > >> use other unused indices to store other attributes about the stack, e.g.
> > >> a description like `x[-1] = "this is a stack of connection requests"`,
> > >> which you couldn't do if you were relying on `length(x)` to tell you the
> > >> index of the top of the stack.
> > >>
> > >> Regards,
> > >>
> > >>      Ed.
> > >>
> > >
> > >
> > > --
> > > Wolfgang Laun
> > >
> > >
> > >
> > 
> > -- 
> > Wolfgang Laun
>
> -- 
> Andrew Schorr                      e-mail: aschorr@telemetry-investments.com
> Telemetry Investments, L.L.C.      phone:  917-305-1748
> 152 W 36th St, #402                fax:    212-425-5550
> New York, NY 10018-8765