Re: [If/then and sequences...]
Oh,, i wasn't getting in on the greater_than() less_than() compare() equal()
discussion,
for when it comes to comparing strings, what is equal depends on the criteria
being
used in the weighing. If numbers weight more, then any integer atom is more than
any
alphabetic sequence. If length matters, then i'd use length(). etc.. But i
*would* prefer
that if i ask for an index into the sequence, i get a numeral, and if i ask if
it's there, i
get true/false. And i'd like to use anything in an if/then, not just atoms, to
me, any valid
comparison (equal, compare, = < > != etc) should be returning an atom, either an
integer/index or true/false. Problem is at this point, if the core is changed,
and isn't
made smart enough to know what to return based on what it's being compared to,
then
all the code written to date is broken. It *can* be made smart enough, but it
may not
be real easy.
Kat
On 29 Aug 2000, at 16:04, Ben Fosberg wrote:
> Granted - my question was intended to ask which of the many possible arbitrary
> ordering schemes Kat was interested in. I suspect this will be a very
> contentious matter, and that any scheme chosen will, of necessity, only appeal
> to a minority of users. If the ordering scheme implemented doesn't relate to
> your use of sequences, then it's just useless weight in the core language.
> IOW,
> the "rules employed" are extremely material - make or break. For that reason,
> I'd say t it's very much the kind of thing that belongs in application code,
> not
> the core language. However, I wouldn't fight to keep it out, and one of you
> sharp guys (the term is meant sincerely) may come up with an ordering rule
> that
> _is_ widely applicable and quite useful.
>
> Derek Parnell wrote:
>
> > It gives us a way of arbitrarily, yet consistently, ordering a set of
> > sequences. In other words, it enables us to know with 100% accuracy, where a
> > given sequence is in an ordered set of sequences. The rules employed to
> > determine which of two sequences comes first is arguably immaterial, as long
> > as they always work the same way.
> >
> > ----- Original Message -----
> > From: "Ben Fosberg" <BenFosberg at ATT.NET>
> > To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
> > Sent: Wednesday, August 30, 2000 6:12 AM
> > Subject: Re: [If/then and sequences...]
> >
> > > What does it mean to say that one sequence is "greater than" or "less
> > than" another sequence,
> > > given that the constituent "elements" of a sequence could be of any data
> > type, including
> > > other sequences?
> > >
> > > Kat wrote:
> > >
> > > > On 29 Aug 2000, at 7:00, Michael Nelson wrote:
> > > >
> > > > > David Cuny wrote:
> > > > > <snip>
> > > > > > Some people might thing that EE or Win32Lib is my greatest
> > contribution to
> > > > > > Euphoria. Not so! It was haraguing Robert into adding equal() to the
> > > > > > Euphoria. 
> > > > >
> > > > > David,
> > > > >
> > > > > Could you also harrange him into adding greater() and less()? Maybe
> > > > > not_greater(), not_less() and not_equal() could also be added--but I
> > see no
> > > > > real gain over not greaater(), not less() and not equal().
> > > >
> > > > This points out the lack of a couple basic commands we could really use.
> > I still vote for
> > > > goto's, especially since i expect to see a lot of them in the translated
> > C code.
> > > >
> > > > Kat
>
|
Not Categorized, Please Help
|
|
