64-Sets

64-set, language concept

Some functions treat values of the long type as 64 separate bits, representing a subset of ${0\ldots 63}$. This acts as a high-performance representation of a set of integers, in cases where using relational logic is not fast enough.

64-sets are created with flag or emptySet.

They can be combined with union and intersection, negated with complement, and compared with isSubsetOf or contains.

A == B and A != B work as expected to check if two sets are exactly equal or not.

The number of elements in a 64-set is given with popCount.

AB = union(flag(1), flag(2))
BC = union(flag(2), flag(3))
ABC = union(flag(1), union(flag(2), flag(3)))

show summary "" with
  union(AB, BC) // "{1,2,3}"
  intersection(AB, BC) // "{2}"
  popCount(ABC) // 3
  isSubsetOf(BC, ABC) // true

Motivation

While typical Envision use-cases involve utilizing tables and vectors for representing sets, these methods can become performance bottlenecks or may be incompatible with certain operations like nested cross functions. 64-sets offer a streamlined, performance-optimized approach to handle boolean representations for a range of scenarios including iterative functions (for, each), Monte Carlo simulations, and more.

64-sets are particularly useful when dealing with temporal sets (weekdays, weeks of the year, months over several years), part bundles, or categories where the overhead of cross-tables could become prohibitive or even infeasible.

While 64-sets offer computational efficiency, they are restricted to representing integers in the range of ${0\ldots 63}$. Therefore, they may not be suitable for all scenarios.

By integrating 64-sets into your Envision workflow, you trade some ease-of-use for computational speed, thereby making them a specialized tool for performance-critical applications.