I don't really know how to describe what I'm doing, but this example should help:
val vals = Array( (0, true),
(1, true),
(2,true),
(3,true),
(4,false),
(5, true),
(6, true),
(7, false),
(8, true),
(9,true))
I'm looking to identify the first and last elements in each of the 'true' regions, though partitioning the array when the value changes could work as well. I could do this imperatively, but what's the best way to do this in scala?
If you don't mind adding some infrastructure to handle a groupedWhile
feature, you can steal from Rex Kerr's answer on extending scala collection. Use the section that handles array, in the second part of the answer.
Then it's a breeze:
scala> vals.groupedWhile(_._2 == _._2).filter(_.head._2 == true).map{g =>
(g.head, g.last)}.foreach(println)
((0,true),(3,true))
((5,true),(6,true))
((8,true),(9,true))
Edit:
I came up with a solution that does not require groupedWhile
. It's based on using Iterator.iterate
which starts from a seed and repeatedly applies the span
function to extract the next group of elements that have the same boolean property. In this instance the seed is a tuple of the next group and the remainder to process:
type Arr = Array[(Int, Boolean)] // type alias for easier reading
val res = Iterator.iterate[(Arr, Arr)]((Array(), vals)){ case (same, rest) =>
// repeatedly split in (same by boolean, rest of data)
// by using span and comparing against head
rest.span(elem => elem._2 == rest.head._2)
}.drop(1).takeWhile{ case (same, _) => // drop initial empty seed array
same.nonEmpty // stop when same becomes empty
}.collect{ case (same, _) if same.head._2 == true =>
// keep "true" groups and extract (first, last)
(same.head, same.last)
}.foreach(println) // print result
Which prints the same result as above. Note that span
for empty arrays does not call the predicate, so we don't get an exception on rest.head
if rest is empty.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With