Day 4 was an exercise in interval relationships, something I've used before in the day job.
Representation
This followed the problem. An Assignment
is a new data type; I represent pairs of Assignment
s as, well, pairs: 2-tuples. Reading and parsing the datafile is straightforward.
data Assignment = Assignment Int Int deriving (Show, Eq)
type Pair = (Assignment, Assignment)
pairsP = pairP `sepBy` endOfLine
pairP = (,) <$> assignmentP <* "," <*> assignmentP
assignmentP = Assignment <$> decimal <* "-" <*> decimal
Interval relations
We can say a few things about relations between intervals (assignments).
Inverval p contains interval q if p starts no later than q and p finishes no earlier than q.
Interval p is wholly before interval q if p finishes before q starts.
contains (Assignment lower1 upper1) (Assignment lower2 upper2) =
(lower1 <= lower2) && (upper1 >= upper2)
before (Assignment _lower1 upper1) (Assignment lower2 _upper2) = (upper1 < lower2)
A pair of intervals (p, q) has a containment if p contains q or q contains p.
A pair of intervals (p, q) is disjoint if p is before q or q is before p. p and q overlap if they are not disjoint.
hasContainment (assignment1, assignment2) =
(assignment1 `contains` assignment2) || (assignment2 `contains` assignment1)
disjoint (assignment1, assignment2) =
(assignment1 `before` assignment2) || (assignment2 `before` assignment1)
overlaps = not . disjoint
Solution
With these definitions in place, the solutions are just filtering the list of assignment pairs depending on the relations we want.
part1 = length . (filter hasContainment)
part2 = length . (filter overlaps)
See also
The intervals library includes all these relations built-in.
Code
You can get the code from my locally-hosted Git repo, or from Gitlab.