The only "anomaly" here is the notation for
one-element list constants, where
4-element list 1 2 3 4
3-element list 1 2 3
2-element list 1 2
1-element list impossible
For a one element list, constants are not possible
and you have to say things like ,1 or 1$1 or some
other expression. But it is preferable to have 1
denote an atom.
Regarding q: itself, since the dyad q: has ranks
0 0, the behaviour of x q: y for general x and y
is completely specified by the behaviour for atoms
x and y. The dictionary says:
If x is positive and finite, x q: y is the
list of exponents in the prime decomposition
of positive integer y, based upon the first
x primes; if x is _ , a sufficient number of
primes is used.
Note the word LIST. 1 is positive and finite, so
the result of 1 q: y should be a list when y is an
atom, and so for general y the shape of 1 q: y
should be ($y),1 .
There are several functions where one of the
arguments becomes part of the shape of the result.
For example:
$ 5 # 12
5
$ 5 2 {. 'abc',:'efg'
5 2
$ i.5 4
5 4
$ 5 ? 12
5
$ 1$. 5 4
5 4
----- Original Message -----
From: "R.E. Boss" <[EMAIL PROTECTED]>
Date: Saturday, February 17, 2007 5:34 am
Subject: RE: [Jgeneral] Shape of 1&q:
> This (kind of) anomalies are not unusual in J:
> $&.> 1; 1 2; 1 2 3; 1 2 3 4; 1 2 3 4 5
> ++-+-+-+-+
> ||2|3|4|5|
> ++-+-+-+-+
>
> and so
> $@(5&q:)&.> 1; 1 2; 1 2 3; 1 2 3 4; 1 2 3 4 5
> +-+---+---+---+---+
> |5|2 5|3 5|4 5|5 5|
> +-+---+---+---+---+
>
> Based on this, you could also state
> $@(q:&y)&.> y=:1 2 3 4 5
> +---+---+---+---+---+
> |5 1|5 2|5 3|5 4|5 5|
> +---+---+---+---+---+
>
> is an anomaly.
>
> $@(q:&y)&.> y=:1 2 3 4 5
> +-+---+---+---+---+
> |5|5 2|5 3|5 4|5 5|
> +-+---+---+---+---+
>
> should considered to be (more) consistent.
>
>
> As it is now ($ x q: y) -: ($ y),x
> What other examples exist, apart from the trivial one $ itself,
> where one of
> the parameters becomes part of the shape of the result?
>
>
> I could not detect where ,5 is not consistent with the dictionary.
>
>
> R.E. Boss
>
>
>
> -----Oorspronkelijk bericht-----
> Van: [EMAIL PROTECTED] [mailto:general-
> [EMAIL PROTECTED] Roger Hui
> Verzonden: dinsdag 13 februari 2007 0:28
> Aan: General forum
> Onderwerp: Re: [Jgeneral] Shape of 1&q:
>
> A result shape of 5 1 in that case is consistent
> with the specifications in the dictionary but
> one of ,5 is not.
>
> Also:
> y=: >: i.5
> $@(q:&y)&.> 1 2 3 4 5 6
> +---+---+---+---+---+---+
> |5 1|5 2|5 3|5 4|5 5|5 6|
> +---+---+---+---+---+---+
>
> Making the first one ,5 instead of 5 1 would be
> anomalous.
>
>
>
> ----- Original Message -----
> From: "R.E. Boss" <[EMAIL PROTECTED]>
> Date: Monday, February 12, 2007 1:37 pm
> Subject: [Jgeneral] Shape of 1&q:
>
> > $2 q: >:i.5
> > 5 2
> >
> > seems reasonable, but why is
> >
> > $1 q: >:i.5
> > 5 1
> >
> > and not just 5?
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
