get syntax error



________________________________
From: Jose Mario Quintana <[email protected]>
To: "[email protected]" <[email protected]> 
Sent: Wednesday, March 22, 2017 11:44 PM
Subject: Re: [Jprogramming] introducing j plus plus



Thanks for the feedback Pascal.  I just realized that I might have used
tacit anonymous continuations inadverterdly when I wrote, in Jx, the tacit
recurrent (aka multiple) adverbs   sa   and   sna  which can convert
adverbs taking arguments as a gerund into adverbs taking arguments  as a
strand [0].  Incidentally, I have rewriten similar versions of
adverbs which run in official J interpreters but that should be a subject
for another thread in another time.

In the meantime the following is a little puzzle in the same vain:


WhoAmI '' 'I am presenting myself by executing this sentence :)' ''
'Spoiler coming in,' '' (,.|.i.33) '' (|.'...brev gnicudorper-fles eht
ma I') ''
I am presenting myself by executing this sentence :)
Spoiler coming in,
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
I am the self-reproducing verb...
(<([ ,^:(0:`&) >)@:[ [ 0 0&$@(1!:2&2)@:])&(([ ,^:(0:`&) >)@:[ [ 0
0&$@(1!:2&2)@:])


[0] [Jprogramming] Tacit recursion without $:

*
http://www.jsoftware.com/pipermail/programming/2014-February/035416.html
<http://www.jsoftware.com/pipermail/programming/2014-February/035416.html>*

On Wednesday, March 22, 2017, 'Pascal Jasmin' via Programming <
[email protected]> wrote:

> The f11 script for running a multiline definition from anywhere inside the
> enclosing code is also pretty cool
>
> 0. for (0 {:: ]) 1. for (1 {:: ]) is a big convenience to  me too.  0.~ is
> (0 {:: [)
>
> but quoteless code is the big feature.  It makes it much easier to use
> custom parsers.  Gerunds specifically are easier to input with a higher
> priority than conjunction separator (Q uses ; the way gg uses ` as
> separator)
>
> It should be compatible with existing code as well.  The one caveat is
> that "fancy" multiline code (hides a (: 0) ) should use the magic constant
> ML so the parser can know it is multiline.  or modify the nummultiline verb.
>
> it already handles code lines such as
>
>
> 3 : 0 : (4 : 0)  NB. 2 multiline closing ) are needed.
>
>
>
> As to question on continuations, it is based closely on
>
>
> http://code.jsoftware.com/wiki/Guides/Lexical_Closure
>
> but with the option of being fully anonymous.  Still currently a toy.
>
> One neat feature in the toy unrelated to closures/continuations is SNR -
> self name reference.  This allows genericity in having a function return a
> function of itself.
>
>
> A function "should"/may return a new function and a result.  The best
> example in the toy is
>
> mycycle
> ┌───────────────────────────────────┬──┐
> │┌───────────────┬───────────┬─────┐│11│
> ││┌─┬───────────┐│cycle_base_│┌─┬─┐││  │
> │││0│10 11 12 13││           ││0│2│││  │
> ││└─┴───────────┘│           │└─┴─┘││  │
> │└───────────────┴───────────┴─────┘│  │
> └───────────────────────────────────┴──┘
>
>
> This contains 2 items.  A gerund representing a function, and the last
> result
>
> cont2 'mycycle'
> 12
> cont2 'mycycle'
> 13
>
> these return the function result but also updates the function
>
>
> mycycle
>
> ┌───────────────────────────────────┬──┐
> │┌───────────────┬───────────┬─────┐│13│
> ││┌─┬───────────┐│cycle_base_│┌─┬─┐││  │
> │││0│10 11 12 13││           ││0│0│││  │
> ││└─┴───────────┘│           │└─┴─┘││  │
> │└───────────────┴───────────┴─────┘│  │
> └───────────────────────────────────┴──┘
>
> cycle_base_ is a dyad returns a gerund, and result, but the gerund in
> mycycle has both parameters bound to it.
>
> cycle SNR4.: (x  FNAME aar tiD ((#x) | >: y) tiD ) ,&< (y{x)
>
> the .: glyph quotes both left and right params (until any terminating ")"
> or end if none found)
> SNR expands the definition to
>
> cycle
> 4 : 0
> FNAME =. 'cycle_base_'
> ( x FNAME aar tiD ( ( # x ) | >: y ) tiD ) , & < ( y { x )
> )
>
> It rotates through x items.
>
>
> Real applications include lazy data structures.  Functions that feed one
> data item at a time can work identically with a file or slow-result socket
> connection, or terminate after the right result found.  Anonymous and
> classless versions of this process means there's nothing to cleanup when
> you stop querying the next value(s).
>
>
> ________________________________
> From: Jose Mario Quintana <[email protected] <javascript:;>>
> To: Programming forum <[email protected] <javascript:;>>
> Sent: Wednesday, March 22, 2017 4:16 PM
> Subject: Re: [Jprogramming] introducing j plus plus
>
>
>
> A quoteless J?  That is neat, I do not like quoting code particularly when
> it leads to quotes within quotes within quotes within...
>
> I find the related continuations.ijs script intriguing.  Can you explain
> what is the purpose of the continuations?  How can you use them?  Simple
> examples would be appreciated.
>
>
> On Fri, Mar 17, 2017 at 9:25 PM, 'Pascal Jasmin' via Programming <
> [email protected] <javascript:;>> wrote:
>
> > A preparser for J that autocloses parentheses, and makes embedding quotes
> > easier.  Also has user-defined primitives.
> >
> >
> > https://github.com/Pascal-J/jpp
> >
> > as examples,
> >
> > 4 timespacex: +/ i.1000
> >
> > is equivalent to
> >
> > 4 timespacex '+/ i.1000'
> >
> > the suggested F11 macro is pretty neat.  Runs a section of code in a
> > file.  Unlike ctrl-R, you don't need to be at the top level line in order
> > to run the multiline group that encloses the current line.
> > ----------------------------------------------------------------------
> > 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
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to