In my brute-force search for valid J words, I did find four 3-character
ones, so such things are possible.


On Mon, Oct 14, 2013 at 9:26 PM, Dan Bron <[email protected]> wrote:

> Yes, you're right (verified by checking the source file Don identified
> [1]). But still, seems a shame to have a limit at all. Though I supposed
> it's a nice feature to have a single C structure to represent any J object
> (whether noun, verb, adverb or conjunction), and no one would want to deal
> with (or remember the meaning of) words like &.::.: anyway.
>
> But it would also be nice if the J programmer (as opposed to just the J
> implementer) enjoyed some of the benefits of this tradeoffs, too.  Making
> 15!:6 work with non-noun names again would be a start [2-].
>
> -Dan
>
> [1]  Mapping table for J primitives:
>      https://github.com/openj/core/blob/master/jc.h
>
> [2]  Behavior of 15!:6 changes in J602 (2008), breaks code, including some
> of my favorite utilities
>      Single-assignment J:
> http://www.jsoftware.com/pipermail/general/2008-November/032554.html
>      Extension to f. :
> http://www.jsoftware.com/pipermail/programming/2009-December/017534.html
>      Tracking mem-mapped nouns:
> http://www.jsoftware.com/pipermail/programming/2009-December/017536.html
>
>
> -----Original Message-----
> From: [email protected] [mailto:
> [email protected]] On Behalf Of Henry Rich
> Sent: Monday, October 14, 2013 8:43 PM
> To: [email protected]
> Subject: Re: [Jprogramming] Mystery J function
>
> The limit would be 128 uninflected graphics (not a problem by definition,
> since ASCII is a 7-bit code) and 128 inflected forms, right?
>   That leaves plenty of room.
>
> Henry Rich
>
> On 10/14/2013 7:13 PM, Dan Bron wrote:
> > Does this imply that J is limited to ~128 primitives? Estimating
> quickly, we're pretty close to that limit already.
> >
> > -Dan
> >
> > Please excuse typos; composed on a handheld device.
> >
> >> On Oct 14, 2013, at 12:36 PM, Roger Hui <[email protected]>
> wrote:
> >>
> >> They are special internally so that each verb/adverb/conjunction can
> >> be encoded by a single byte.  It is a bug that the encoding is
> >> visible to the user.
> >>
> >>
> >>
> >>> On Mon, Oct 14, 2013 at 9:34 AM, Devon McCormick <[email protected]>
> wrote:
> >>>
> >>> Why is the top half of a. special?  I would think that all the
> >>> documented words J cares about are in the lower half.
> >>>
> >>>
> >>> On Mon, Oct 14, 2013 at 11:34 AM, Roger Hui
> >>> <[email protected]
> >>>> wrote:
> >>>
> >>>> Convert 232 to hex and start looking for that.
> >>>>
> >>>> It'd be a really bad idea to depend on those encodings as that is
> >>> internal
> >>>> and may change without notice.
> >>>>
> >>>>
> >>>>> On Mon, Oct 14, 2013 at 8:26 AM, Don Guinn <[email protected]>
> wrote:
> >>>>>
> >>>>> The top half of a. is special when executing a sentence. Codes are
> >>>> assigned
> >>>>> to those bytes. I looked for the source but haven't found it yet.
> >>>>>
> >>>>> On Oct 14, 2013 8:50 AM, "Devon McCormick" <[email protected]>
> >>> wrote:
> >>>>>
> >>>>>> I have no expectation of the output.  I only found it because I
> >>>>>> was checking all character combinations in a "try...catch" phrase
> >>>>>> to
> >>>>> determine
> >>>>>> which ones don't give an error.
> >>>>>>
> >>>>>> I did try it in J6 and it behaves the same way.  I don't know if
> >>>>>> I
> >>>> have a
> >>>>>> copy of J5 anywhere.
> >>>>>>
> >>>>>> It's not important - it's just something that passes my simple
> >>>>>> test
> >>> yet
> >>>>> is
> >>>>>> not really a known J word.
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> On Sun, Oct 13, 2013 at 11:47 PM, bill lam <[email protected]>
> >>>> wrote:
> >>>>>>
> >>>>>>> The string is invalid as utf8, what would you expect the output
> >>>>>>> should be?  Not sure how it will run under J504.
> >>>>>>>
> >>>>>>> Вс, 13 окт 2013, Devon McCormick писал(а):
> >>>>>>>> While experimenting with executing randomly-generated snippets
> >>>>>>>> of
> >>>> J,
> >>>>> I
> >>>>>>> ran
> >>>>>>>> across this:
> >>>>>>>>
> >>>>>>>>    (232{a.),'3'
> >>>>>>>> Φ3
> >>>>>>>>    ".(232{a.),'3'
> >>>>>>>>
> >>>>>>>>    $".(232{a.),'3'
> >>>>>>>> 0
> >>>>>>>>    {.".(232{a.),'3'
> >>>>>>>> 0
> >>>>>>>>    '1 2 3',(232{a.),'3 4 5'
> >>>>>>>> 1 2 3Φ3 4 5
> >>>>>>>>    ".'1 2 3',(232{a.),'3 4 5'
> >>>>>>>>
> >>>>>>>> This high-bit character seems to return i.0 for any argument.
> >>>>>>>>
> >>>>>>>> --
> > ----------------------------------------------------------------------
> > 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
>



-- 
Devon McCormick, CFA
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to