« first day (888 days earlier)      last day (647 days later) » 

7:38 AM
@ngn ah good catch. when does it return in the wrong order? updated
 
 
8 hours later…
ngn
4:04 PM
@chrispsn that works! thanks
i'll replace the old k-string with this
 
 
3 hours later…
7:09 PM
@chrispsn hmm what's wrong with the x@&(!#x)=x?x approach? does it not handle inconsistent rank-ed inputs or something?
 
7:36 PM
I guess a candidate for the k tech tree would be {(0#x){$[0|/x~\:y;x;x,,y]}/x}
 
 
2 hours later…
9:56 PM
@coltim basically I wanted an option that isn't dependent on find
 
10:21 PM
@coltim i like how straightforward your one is, and it doesn't rely on sorting or find. added
 
10:55 PM
@chrispsn new goal: reimplement primitives with the equivalent of for loops =P
there could be the seed of a "here's something that works in the language, but is not idiomatic" thing. would be easier if the over/scan overloads weren't one of the more nuanced parts of the languageā€¦
 

« first day (888 days earlier)      last day (647 days later) »