« first day (845 days earlier)      last day (56 days later) » 

11:50 AM
@Adám well, none of those names are common in Brazil, but iirc there's a list of names that won't capitalize automatically. I don't have access to that part of the code anymore though, so I can't tell for sure :p
 
 
7 hours later…
6:27 PM
We've been over this before, but why is it again that we can't extend @ with fractional indices meaning insert? Obviously, it can only work with monadic application (array ⍺⍺) and only with 1-dimensional insertion points (simple vector ⍵⍵), but wouldn't that work?
 
ngn
@Adám would 'ab'@0.5 0.5⊢'cd' ⊣⎕io←0 be 'cabd' or 'cbad' or 'cbd'?
 
@ngn not the middle one.
 
@Adám it would work if you define it properly, but imo the whole fractional numbers thing is pretty strange as it takes up an important-to-error domain (i.e. index of 2.0001 rounds to inserting after 2, whereas I'd expect an error as it's probably a floating point mistake. even 2.5 could be a mistake, dividing a (supposedly but not actually) even number by 2; imo the fractionals are an even worse syntax than f[dim] or A~flattened because, well, it's more abuse than syntax)
(if anyone hasn't been able to tell from my many messages before, I very, very, very much like proper errors where they should be)
 
ngn
6:55 PM
@Adám fyi: k7 is planned to have splice for insertions, which is separate from amend
 
7:07 PM
I think this works:
 At←{
     ⍺←⊢
     2∨.≠⎕NC↑'⍺⍺' '⍵⍵':⍺(⍺⍺@⍵⍵)⍵
     ins←~rpl←(⌊=⊢)⍵⍵
     y←(rpl/⍺⍺)@(rpl/⍵⍵)⊢⍵
     idx←⊂⍋(⍳≢y),ins/⍵⍵
     idx⌷y,ins/⍺⍺
 }
 ⋄ At←{⍺←⊢⋄2∨.≠⎕NC↑'⍺⍺' '⍵⍵':⍺(⍺⍺@⍵⍵)⍵⋄ins←~rpl←(⌊=⊢)⍵⍵⋄y←(rpl/⍺⍺)@(rpl/⍵⍵)⊢⍵⋄idx←⊂⍋(⍳≢y),ins/⍵⍵⋄idx⌷y,ins/⍺⍺} ⋄ ⎕←'abxde' At 0.5 0.5 1 0.4 1.5⊢'XY'
 
@Adám
dabxeY
 
@DyalogAPL Hm, can I do multi-line stuff if I terminate or begin lines with ?
⋄ a←⍳3
⋄ ⎕←,⍨a
 
@Adám
1 2 3 1 2 3
 
@DyalogAPL OK, I'll start doing that than.
@ngn 'cabd'
 
@ngn oh, that's a lovely reference
 
7:18 PM
@ngn Yeah, that's amazing. Will study.
@ngn Ooh, your chat room is referenced.
 
ngn
@dogstar all thanks to @chrispsn
 
 
3 hours later…
10:04 PM
@ngn thanks :)
 

« first day (845 days earlier)      last day (56 days later) »