Iteration produces unexpected number of elements

[fardin01]

I am writing a policy that looks for given annotations on a K8s object and compares them to a list of required annotations to see if a required annotation is missing. The code can be found here.

I've tried two different ways:

missing := required - provided	
count(missing) > 0

and

r := required[_]
missing := r; not r in provided

They both have the desired outcome. However, in the first case, count(missing) returns 1 which is the expected number because there is 1 required annotation and it is missing. In the second case, count(missing) returns 10. I do understand that in this case, I am doing a loop but I do not understand where the number 10 comes from.

What's happening here?

[srenatus]

I believe the misunderstanding might be in the meaning of ; in

r := required[_]
missing := r; not r in provided

It does not constitute a restriction or WHERE-clause, or the RHS of a comprehension. It merely is the same as a line-break. So if you look at it like that

r := required[_]
missing := r
not r in provided

it would make sense that missing has the same number of elements as some element of required. Try adding a print(missing) and you'll find that missing is a string, "repository, for which count() returns the number of characters, 10.

I believe you might be looking for a set comprehension here, like

missing := { r | r := required[_]; not r in provided }

or, equivalently,

missing := { r | some r in required; not r in provided }