I agree with Fred's comments to Nicholas below, and would like to add a
couple more.
Despite their appearance, { and } are not symmetric: in particular } is an
adverb which has a gerund form for these purposes. The verb { is very
much a "J-style" operator, while } is not.
Defining
get=:4 : 'x.{y.'
looks like it is a step along the way of "give meaningful names to
functions", but it is not quite that: get has a different rank from {.
get b. 0
_ _ _
{ b. 0
1 0 _
You would be better off with
get=:{
(with some code to possibly exclude the monad case).
For the problem at hand, I would avoid the get/set terminology, and just
use the gerund form.
set=:{.@:[`({:@:[)`]}
inc=:>:@:{`[`]}
]a=:2 0 0 0
2 0 0 0
2 3 set a
2 0 0 2
2 inc a
2 0 1 0
Best wishes,
John
Fred Bone wrote:
> On 25 Mar 2006 at 22:20, Nicholas Spies said:
>
>> To conceal some of the visual noise implicit in J, I (foolishly, no
>> doubt) tried to do the following:
>>
>> NB. define an array in which to store data
>> a =: 0 0 0 0 0 0 0 0
>>
>> NB. simple enough
>> get =: 4 : 'x.{ y.'
>>
>> set =: 4 : '({. x.) (({: x.)}) y.'
>>
>> ]a =: 3 0 set a
>> 3 0 0 0 0 0 0 0
>>
>> ]a =: 2 0 set a
>> 2 0 0 0 0 0 0 0
>>
>> NB. defining this was the whole point of my feeble efforts
>> inc =: 4 : '((>:(x. get y.))(x.) set y.)'
>> dec =: 4 : '((<:(x. get y.))(x.) set y.)'
>>
>> NB. above gives an error
>> inc =: 4 : '( 1&+(x. get y.)),( x.) set y.'
>> dec =: 4 : '(_1&+(x. get y.)),( x.) set y.'
>> a
>> 2 0 0 0 0 0 0 0
>> ]a =: 0 inc a
>> 1 0 0 0 0 0 0 0 0
>>
>> NB. OK, it puts the 1: at the 0th position... but adds a zero
>>
>> inc =: 4 : '( 1&+(x. get y.)),( x.) set y.'
>> dec =: 4 : '(_1&+(x. get y.)),( x.) set y.'
>> a =: b
>> a
>> 0 0 0 0 0 0 0 0
>> ]a =: 0 inc a
>> 1 0 0 0 0 0 0 0 0
>> ]a =: 0 inc a
>> 2 0 0 0 0 0 0 0 0 0
>> ]a =: 0 inc a
>> 3 0 0 0 0 0 0 0 0 0 0
>>
>> NB. in this case, incrementing works but the array lengthens...
>>
>> Admittedly, trying to do this may be just plain foolish, but what's the
>> REAL way to do this?
>
> Your problems seem to stem from inappropriate use of parens.
>
> Your second "inc" function:
> inc =: 4 : '( 1&+(x. get y.)),( x.) set y.'
> is not supplying a concatenation of value + position to the "set" verb,
> but is attempting to concatenate the incremented value with the result of
> supplying just a position to "set" (which does nothing). It is not, as
> you suggest, appending a zero:
> 3 inc i.5
> 4 0 1 2 3 4
> 2 inc i.5
> 3 0 1 2 3 4
>
> You can see this more clearly if you use "13 :" instead of "4 :":
> 13 : '( 1&+(x. get y.)),( x.) set y.'
> ([: 1&+ get) , set
>
> I think this does what you want:
> get =: {
> set =: 4 : '({.x.) (}. x.) } y.'
> (Note that this version of "set" won't let you get away with short
> measure in the left operand)
> inc =: 13 : '((>: x. get y.),x.) set y.'
> inc
> (([: >: get) , [) set ]
> 3 inc a
> 0 0 0 1 0 0 0 0
>
> That's if you really want to define "inc" in terms of "get" and "set".
> However, you could use the gerund case of amend:
> inc =: (>:@get)`[`] }
> 3 inc a
> 0 0 0 1 0 0 0 0
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
