[m-dev.] For review: minor additions to list.m and string.m

[Ralph Becket]

> > > library/list.m:
> > > 	Added func ++/2 as a synonym for list__append/2.
> > > 	Added func '..'/1 to generate ascending/descending integer range
> > > lists.
> > 
> > Firstly, s/1/2/.

Done.

> > I'm not sure this is the best way to do descending 
> sequences. This means
> > you can never generate an empty list.  Eg, I would find
> > 
> > 	1..0 = []
> > 
> > more natural than
> > 
> > 	1..0 = [1,0]
> 
> I agree.  Also the behaviour of `..' is inconsistent with Haskell,
> for which `[1..0]' evaluates to `[]'.  In Haskell, you need to write
> `[1,0..0]' to get `[1,0].

I, too, agree.  I just remember the last time someone suggested this
they wanted the behaviour I implemented and there were no howls from
the gallery.

How about this: `..`/2 is the inclusive ascending series

	(N `..` M) = Seq
	  <=>  all [I] (index0(Seq, I, X) => (X = N + I, X =< M))

and add

	arith_series(N, M, D) = Seq
	  <=>  (      if D > 0 then
	                all [I] (index0(Seq, I, X) => (X = N + I * D, X =<
M))
	         else if D < 0 then
	                all [I] (index0(Seq, I, X) => (X = N + I * D, X >=
M))
               else /* D = 0 */
                      throw(a_wobbly)
	       )

> Ralph, the `++' methods look fine; unless there are any other 
> objections, I suggest you go ahead and commit those.
> 
> For `..', I suggest you change the semantics to match those 
> in Haskell.

I think arith_series/3 captures that.

Any takers for geom_series/3?  Or is that of too little use?  Mind you, it
would be trivial to include.

Ralph
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------