« first day (2932 days earlier)      last day (5 days later) » 

14:14
@RubenVerg so do I. I did not find an easy solution.
 
7 hours later…
21:23
A thought experiment: What if we gave up being able to specify default values for only some names in ⎕VGET. Then the argument to both ⎕VSET and ⎕VGET would be two items, the second item in ⎕VGET being optional. For ⎕VGET we would have to enclose or mix a set of names only, when not providing defaults. And for both ⎕VGET and ⎕VSET if we were given pairs we would have to ↓⍉↑ before processing. On the upside, we would not have to enclose when we have one item to get or set.
And we would not have to mix a v/v of names ...
Not sure what other implications would follow.
In my opinion, that's a rather heavy price. Enclosing for a single set/item is quite normal. You have it in ⎕NEW object (⊂'Param' value) and in ⎕DT⊂timestamp
I do wonder how often specifying some default values but not others would actually arise as the structure 'Name1' ('Name2' Value2) 'Name3' ... is not very APL-like, and does not arise that easily I think.
I see your point.
 Range←{⎕IO←0 ⋄ (a n b)←⍵ ⎕VGET('from' 0)('step' 1)'to' ⋄ a+n×⍳⌊n÷⍨b-a}
@user28169060 Hi James Hemphill, if you want to participate here, please email [email protected]
@PaulMansour How do you like the above example?
It's nice.
In production to would default to ⍺ ;)
 
2 hours later…
23:50
@Richard I had something like {' \/X'[1+(⊢+2×⌽)∘.=⍨⍳⍵]}

« first day (2932 days earlier)      last day (5 days later) »