https://learninghub.kx.com/forums/topic/missing-some-items
We encounter the enlist keyword early in learning q; a reliable way to make a one-item list where 1# wont do. But there is a bit more to enlist than that.
In the first place, it is not only variadic taking various numbers of arguments it will take any number of arguments, even more than eight.
q)enlist[one;two;three;four;five;six;seven;eight;nine] onetwothreefourfivesixseveneightnine
Of course, what it returns is a list, with each argument an item.
q)enlist[one;two;threeandabit;four;five;six;seven;eight;nine]
one two threeandabit four five six seven eight nine
What is less obvious is that enlist is implicated in lists with missing items.
q)type (`one;`two;`three) / symbol vector 11h q)type (`one;::;`three) / mixed list 0h q)type (`one;;`three) / a -- projection? 104h q)(`one;;`three) ~ enlist[`one;;`three] 1b
Is the missing item a generic null? No, a list with one or more missing items is a projection of enlist, and its rank is the number of missing arguments. We can apply and iterate it as any other projection.
q)(`one;;`three;;`five)[`two;`four] `one`two`three`four`five q)(`one;;`three;;`five) . `two`four `one`two`three`four`five q)raze `quick`crafty`cunning(`the;;`brown;;`jumps)/::`fox`cat`dog the quick brown fox jumps the quick brown cat jumps the quick brown dog jumps the crafty brown fox jumps the crafty brown cat jumps the crafty brown dog jumps the cunning brown fox jumps the cunning brown cat jumps the cunning brown dog jumps
Which gives us a few tricks up our sleeves when generating test datasets, or preparing data for loading and ingestion.