you mean ([: v [: u a ]) is not the same as ([: v@:u a ]) but if u incorporates 
a , like in ([: v [: (u a) ]) , then it should be the same as ([: v@:(u a) ]), 
IIRC. 


R.E. Boss


> -----Original Message-----
> From: [email protected] [mailto:programming-
> [email protected]] On Behalf Of 'Pascal Jasmin' via
> Programming
> Sent: donderdag 12 februari 2015 15:13
> To: [email protected]
> Subject: Re: [Jprogramming] Source of frustration
> 
> 
> 
> ([: v [: u ]) is almost but not exactly the same as
> 
> ([: v@:u ])
> 
> basically, if u has any adverbs applied to it (such as "1) then those adverbs
> also apply to v in v@:u.  But don't apply to v in ([: v u)
> 
> A good reason for 13 : to have given you what you got is that you explicitly
> used @ instead of made just a linear phrase.
> 
> 
> 
> 
> ________________________________
> From: Linda Alvord <[email protected]>
> To: [email protected]
> Sent: Thursday, February 12, 2015 3:09 AM
> Subject: Re: [Jprogramming] Source of frustration
> 
> 
> There have been many explanations of y=how @ works, but here is the
> question
> I really am interested in.
> 
> Working code:
>   (i.4)*/>:i.5
> 1 2  3  4  5
> 2 4  6  8 10
> 3 6  9 12 15
> 4 8 12 16 20
> Simple working definition:
>   times=: 13 :'(>:i.x)*/>:i.y'
>    4 times 5
> 1 2  3  4  5
> 2 4  6  8 10
> 3 6  9 12 15
> 4 8 12 16 20
> Definition which also works but  seems cumbersome:
>    times2=: 13 :'(>:@i.x)*/>:@i.y'
>    4 times 5
> 1 2  3  4  5
> 2 4  6  8 10
> 3 6  9 12 15
> 4 8 12 16 20
> Why couldn't my definition of times be converted to what seems to be the
> preferred J version in times2?
>    times
> ([: >: [: i. [) */ [: >: [: i. ]
> 
>    times2
> ([: >:@i. [) */ [: >:@i. ]
> 
> Linda
> 
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Linda
> Alvord
> Sent: Monday, February 9, 2015 5:01 PM
> To: [email protected]
> Subject: Re: [Jprogramming] Source of frustration
> 
> Here's another example of my suggestion:
> 
> 
> 
> These two expressions have the same result.
> 
> 
> 
>            eb=: 13 :'y $<'' '''
> 
>    ([:<eb)"1(>:i.2),"0/>:i.4
> 
> ----T-----T-------T---------┐
> │--┐│--T-┐│--T-T-┐│--T-T-T-┐│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │L--│L-+--│L-+-+--│L-+-+-+--│
> +---+-----+-------+---------+
> │--┐│--T-┐│--T-T-┐│--T-T-T-┐│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │+-+│+-+-+│+-+-+-+│+-+-+-+-+│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │L--│L-+--│L-+-+--│L-+-+-+--│
> L---+-----+-------+----------
> 
>    ([:<eb)"1(>:i.2),"0/>:i.4
> 
> ----T-----T-------T---------┐
> │--┐│--T-┐│--T-T-┐│--T-T-T-┐│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │L--│L-+--│L-+-+--│L-+-+-+--│
> +---+-----+-------+---------+
> │--┐│--T-┐│--T-T-┐│--T-T-T-┐│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │+-+│+-+-+│+-+-+-+│+-+-+-+-+│
> ││ │││ │ │││ │ │ │││ │ │ │ ││
> │L--│L-+--│L-+-+--│L-+-+-+--│ L---+-----+-------+----------
> 
> 
> 
> Define each one as  I  and  j
> 
> 
> 
> 
> 
>    i=: 13 :'(<@eb)"1(>:@i.x),"0/>:@i.y'
> 
>    j=: 13 :'([:<eb)"1(>:i.x),"0/>:i.y'
> 
> 
> 
> Both agree as monads and dyads.
> 
> 
> 
> (i 2)-:j 2
> 
> 1
> 
>    (2 i 4)-:2 j 4
> 
> 1
> 
> In the interpretation of J,  [:  is inserted when appropriate.
> 
> 
> 
> My question is: "why couldn't both have the same
> 
> Definition when trains are involved. It @ is superior, I would suggest the
> definition of  i
> 
> for both.
> 
> 
> 
>     eb
> 
> (<' ') $~ ]
> 
> 
> 
>    i
> 
> [: <@eb"1 ([: >:@i. [) ,"0/ [: >:@i. ]
> 
> 
> 
>    j
> 
> [: ([: < eb)"1 ([: >: [: i. [) ,"0/ [: >: [: i. ]
> 
> 
> 
> Linda
> 
> 
> 
>    eb
> 
> (<' ') $~ ]
> 
> 
> 
> 
> 
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Kip
> Murray
> Sent: Monday, February 9, 2015 11:00 AM
> To: [email protected]
> Subject: Re: [Jprogramming] Source of frustration
> 
> 
> 
> And this:
> 
> 
> 
>    hh =: 3 : '[: >: i. y'
> 
>    hh
> 
> 3 : '[: >: i. y'
> 
>    hh 4
> 
> |domain error: hh
> 
> |       [:>:i.y
> 
> 
> 
> I think Brian's suggestion that you parenthesize the result of 13 : before
> applying arguments is the best way of removing your frustration.  My point
> about parenthsizing [: > i. before applying arguments is specific to verbs
> defined by trains.
> 
> 
> 
> --Kip
> 
> 
> 
> 
> 
> On Monday, February 9, 2015, Linda Alvord <
> <mailto:[email protected]>
> [email protected]> wrote:
> 
> 
> 
> > How about this:
> 
> > h=: 13 :'[:>:i.y'
> 
> >    h
> 
> > [: [: [: >: i.
> 
> >
> 
> >    [: [: [: >: i. 4
> 
> > |domain error: scriptd
> 
> > |   [:[:    [:>:i.4
> 
> >
> 
> > Linda
> 
> >
> 
> >
> 
> > -----Original Message-----
> 
> > From:  <mailto:[email protected]>
> [email protected] <javascript:;>
> 
> > [
> > <mailto:programming-
> [email protected]%20%3cjavascript:;%3e>
> mailto:[email protected] <javascript:;>] On
> 
> > Behalf Of Brian Schott
> 
> > Sent: Monday, February 9, 2015 10:00 AM
> 
> > To: Programming forum
> 
> > Subject: Re: [Jprogramming] Source of frustration
> 
> >
> 
> > Linda,
> 
> >
> 
> > If you are willing to put parentheses around your example verbs --
> 
> > with the data outside the parens, you will not get the domain error.
> 
> >
> 
> > On Mon, Feb 9, 2015 at 9:54 AM, Linda Alvord <[email protected]
> 
> > <javascript:;>>
> 
> > wrote:
> 
> >
> 
> > > Suppose:   ff=: 13 :'>:i.y'
> 
> > >    ff
> 
> > > >:@i.
> 
> > > Then:
> 
> > >   >:@i.4
> 
> > > 1 2 3 4
> 
> > > Which is what you want, more than a domain error.
> 
> > >
> 
> > > Linda
> 
> > >
> 
> > > --
> 
> > (B=)
> 
> > ----------------------------------------------------------------------
> 
> > For information about J forums see
> > <http://www.jsoftware.com/forums.htm>
> http://www.jsoftware.com/forums.htm
> 
> >
> 
> > ----------------------------------------------------------------------
> 
> > For information about J forums see
> > <http://www.jsoftware.com/forums.htm>
> http://www.jsoftware.com/forums.htm
> 
> >
> 
> 
> 
> 
> 
> --
> 
> Sent from Gmail Mobile
> 
> 
> 
> 
> ----------------------------------------------------------------------
> 
> For information about J forums see
> <http://www.jsoftware.com/forums.htm>
> http://www.jsoftware.com/forums.htm
> 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
> 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to