Hmm... some questions then:
(1) What would a 5!:1 representation of closures look like? (Is it
basically "just an explicit definition"?)
(2) Can there be adverb/conjunction closures? Or only verb closures?
(3) Are closure symbols local names, or something different? (Can they
be updated?)
(5) W
Like I said: a closure takes the form u &:: (k;v;k;v...), which evaluates u
with bindings k;v;k;v... active. Thereby 5!:1 and 5!:5 trivially obtain.
u can be a proverb, or a tacit verb containing proverbs, but in that case
there's no reason not to eagerly bind, so eg f &:: (f;+`'') is equivale
Like I said, I think it's reasonable to say that {{ ". y }} does not close
over a or b, for performance reasons. You can say {{ ". y [ a [ b }} or
similar if you want to explicitly include them. Happy to hear arguments to
the contrary, though.
On Mon, 23 Jan 2023, Raul Miller wrote:
Also..
Also if this is literally a closure:
example=: {{
a=. 1
b=. 2
{{ ". y }} y
}}
I think this would mean that either 'example' here could not be
represented using 3 : string, or that we would have a lot of closures
popping into existence where they previously did not exist. So... in
additi
eg f &:: (f;+`'') is equivalent to +
(Err, f &:: ('f';+`'').)
--
For information about J forums see http://www.jsoftware.com/forums.htm
Jandroid j904 beta k is available for download now.
On Mon, Jan 23, 2023, 1:10 AM bill lam wrote:
> Me too. I will check.
>
>
> On Sun, 22 Jan 2023 at 10:34 PM Thomas Arneson wrote:
>
>> Same for me.
>>
>> > On 01/22/2023 3:00 AM CST Jan-Pieter Jacobs
>> wrote:
>> >
>> >
>> > I don't know if a
