I am trying to make a function that splits a list of variable length into three lists of even length in order. The following splits it into three, but processes inserts them into each list one at a time.
An example of what I want is:
[1, 2, 3, 4, 5] -> [1, 2], [3, 4], [5]
Another example would be:
[8, 7, 6, 5, 4, 3, 2, 1] -> [8, 7, 6], [5, 4, 3], [2, 1].
The following code splits them by inserting into each list one at a time:
div([], [], [], []).
div([X], [X], [], []).
div([X,Y], [X], [Y], []).
div([X,Y,Z|End], [X|XEnd], [Y|YEnd], [Z|ZEnd]):-
div(End, XEnd, YEnd, ZEnd).
This code outputs:
[1, 2, 3, 4, 5] -> [1, 4], [2, 5], [3]
How can I fix this problem?
The answer by @Boris does not terminate when the length of the list of the first argument is not known. To see this, there is no need to look any further than the first goal with a failure-slice:
On the other hand, your original program had quite nice termination properties. cTI gave the following optimal termination property:
In other words, to ensure termination, you only need a single argument (either
A
orB
orC
orD
) to be a concrete list that is finite and ground (that's whatb(..)
means). That is a very strong termination condition. It's really a pity that the arguments do not fit! Why not generalize your program? The only problem it has is that it restricts the list elements. So I will replace all variable names of list elements by_
s:The very same termination properties hold for this program.
Alas, it is now a bit too general. Boris's solution can now be repurposed:
My preferred way to express the same would rather be:
See other answers for a definition of
seq//1
.