Alloy - Dealing with unbounded universal quantifiers - scope

Good afternoon,
I've been experiencing an issue with Alloy when dealing with unbounded universal quantifiers. As explained in Daniel Jackson's book 'Software Abstractions' (Section 5.3 'Unbounded Universal Quantifiers'), Alloy has a subtle limitation regarding universal quantifiers and assertion checking. Alloy produces spurious counterexamples in some cases, such as the next one to check that sets are closed under union (shown in the aforementioned book):
sig Set {
elements: set Element
}
sig Element {}
assert Closed {
all s0, s1: Set | some s2: Set |
s2.elements = s0.elements + s1.elements
}
check Closed for 3
Producing a counterexample such as:
Set = {(S0),(S1)}
Element = {(E0),(E1)}
s0 = {(S0)}
s1 = {(S1)}
elements = {(S0,E0), (S1,E1)}
where the analyser didn't populate Set with enough values (a missing Set atom, S2, containing the union of S0 and S1).
Two solutions to this general problem are suggested then in the book:
1) Declaring a generator axiom to force Alloy to generate all possible instances.
For example:
fact SetGenerator{
some s: Set | no s.elements
all s: Set, e: Element |
some s': Set | s'.elements = s.elements + e
}
This solution, however, produces a scope explosion and may also lead to inconsistencies.
2) Omitting the generator axiom and using the bounded-universal form for constraints. That is, quantified variable's bounding expression doesn't mention the names of generated signatures. However, not every assertion can be expressed in such a form.
My question is: is there any specific rule to choose any of these solutions? It isn't clear to me from the book.
Thanks.

No, there's no specific rule (or at least none that I've come up with). In practice, this doesn't arise very often, so I would deal with each case as it comes up. Do you have a particular example in mind?
Also, bear in mind that sometimes you can formulate your problem with a higher order quantifier (ie a quantifier over a set or relation) and in that case you can use Alloy*, an extension of Alloy that supports higher order analysis.

Related

Understanding recurive let expression in lambda calculus with Haskell, OCaml and nix language

I'm trying to understand how recursive set operate internally by comparing similar feature in another functional programming languages and concepts.
I can find it in wiki. In that, I need to know Y combinator, fixed point. I can get it briefly in wiki.
Then, now I start to apply this in Haskell.
Haskell
It is easy. But I want to know behind the scenes.
*Main> let x = y; y = 10; in x
10
When you write a = f b in a lazy functional language like Haskell or Nix, the meaning is stronger than just assignment. a and f b will be the same thing. This is usually called a binding.
I'll focus on a Nix example, because you're asking about recursive sets specifically.
A simple attribute set
Let's look at the initialization of an attribute set first. When the Nix interpreter is asked to evaluate this file
{ a = 1 + 1; b = true; }
it parses it and returns a data structure like this
{ a = <thunk 1>; b = <thunk 2>; }
where a thunk is a reference to the relevant syntax tree node and a reference to the "environment", which behaves like a dictionary from identifiers to their values, although implemented more efficiently.
Perhaps the reason we're evaluating this file is because you requested nix-build, which will not just ask for the value of a file, but also traverse the attribute set when it sees that it is one. So nix-build will ask for the value of a, which will be computed from its thunk. When the computation is complete, the memory that held the thunk is assigned the actual value, type = tInt, value.integer = 2.
A recursive attribute set
Nix has a special syntax that combines the functionality of attribute set construction syntax ({ }) and let-binding syntax. This is avoids some repetition when you're constructing attribute sets with some shared values.
For example
let b = 1 + 1;
in { b = b; a = b + 5; }
can be expressed as
rec { b = 1 + 1; a = b + 5; }
Evaluation works in a similar manner.
At first the evaluator returns a representation of the attribute set with all thunks, but this time the thunks reference an new environment that includes all the attributes, on top of the existing lexical scope.
Note that all these representations can be constructed while performing a minimal amount of work.
nix-build traverses attrsets in alphabetic order, so it will evaluate a first. It's a thunk that references the a + syntax node and an environment with b in it. Evaluating this requires evaluating the b syntax node (an ExprVar), which references the environment, where we find the 1 + 1 thunk, which is changed to a tInt of 2 as before.
As you can see, this process of creating thunks but only evaluating them when needed is indeed lazy and allows us to have various language constructs with their own scoping rules.
Haskell implementations usually follow a similar pattern, but may compile the code rather than interpret a syntax tree, and resolve all variable references to constant memory offsets completely. Nix tries to do this to some degree, but it must be able to fall back on strings because of the inadvisable with keyword that makes the scope dynamic.
I guess several things by myself.
In eagar evaluation language, I must declare before use it. So the order of declaration is simple.
int x = 10;
int y = x;
Just for Nix language
In wiki, there isn't any concept comparision with Haskell though let ... in is compared with Haskell.
lexical scope
all variables are lexically scoped.
mutual recursion
https://en.wikipedia.org/wiki/Let_expression#Mutually_recursive_let_expression

Unexpected results in playing with relations

/*
sig a {
}
sig b {
}
*/
pred rel_test(r : univ -> univ) {
# r = 1
}
run {
some r : univ -> univ {
rel_test [r]
}
} for 2
Running this small test, $r contains one element in every generated instance. When sig a and sig b are uncommented, however, the first instance is this:
In my explanation, $r has 9 tuples here and still, the predicate which asks for a one tuple relation succeeds. Where am I wrong?
An auxiliary question: are these two declarations equivalent?
pred rel_test(r : univ -> univ)
pred rel_test(r : set univ -> univ)
The problem is that with the Forbid Overflow option set to No the integer semantics in Alloy is wrap around, and with the default scope of 3 (bits), then indeed 9=1, as you can confirm in the evaluator.
With the signatures a and b commented the biggest relation that can be generated with scope 2 has 4 tuples (since the max size of univ is 2), so the problem does not occur.
It also does not occur in the latest build because I believe it comes with the Forbid Overflow option set to Yes by default, and with that option the semantics of integers rules out instances where overflows occur, precisely the case when you compute the size of the relation with 9 tuples. More details about this alternative integer semantics can be found in the paper "Preventing arithmetic overflows in Alloy" by Aleksandar Milicevic and Daniel Jackson.
On the main question: what version of Alloy are you using? I'm unable to replicate the behavior you describe (using Alloy 4.2 of 22 Feb 2015 on OS X 10.6.8).
On the auxiliary question: it appears so. (The language reference is not quite as explicit as one might wish, but it begins one part of its discussion of multiplicities with "If the right-hand expression denotes a unary relation ..." and (in what I take to be the context so defined) "the default multiplicity is one"; the conditional would make no sense if the default multiplicity were always one.
On the other hand, the same interpretive logic would lead to the conclusion that the language reference believes that unary multiplicity keywords are only allowed before expressions denoting unary relations (which would appear to make r: set univ -> univ ungrammatical). But Alloy accepts the expression and parses it as set (univ -> univ). (The alternative parse, (set univ) -> univ, would be very hard to assign a meaning to.)

Behavior of `=` in alloy fact

I was experimenting with alloy and wrote this code.
one sig s1{
vals: some Int
}{
#vals = 4
}
one sig s2{
vals: some Int
}{
#vals = 4
}
fact {
all a : s1.vals | a > 2
all i : s2.vals | i < 15
s1.vals = s2.vals
}
pred p{}
run p
It seems to me that {3,4,5,6} at least is a solution however Alloy says no instance found. When I comment s1.vals = s2.vals or change i < 15 to i > 2, it finds instances.
Can anyone please explain me why? Thanks.
Alloy's relationship with integers is sometimes mildly strained; it's not designed for heavily numeric applications, and many uses of integers in conventional programming are better handled in Alloy by other signatures.
The default bit width for integers is 4 bits, and Alloy uses twos-complement integers, so your run p is asking for a world in which integers range in value from -8 to 7. In that world, the constraint i < 15 is subject to integer overflow, and turns out to mean, in effect, i < -1. (To see this, comment out both of your constraints so that you get some instances. Then (a) leaf through the instances produced by the Analylzer and look at the integers that appear in them; you'll see their range is as I describe. Also, (b) open the Evaluator and type the numeral "15"; you'll see that its value in this universe is -1.)
If you change your run command to provide an appropriate bit width for integers (e.g. run p for 5 int), you'll get instances which are probably more like what you were expecting.
An alternative change, however, which leads to a more idiomatic Alloy model, is to abstract away from the specific kind of value by defining a sig for values:
sig value {}
Then change the declaration for vals in s1 and s2 from some Int to some value, and comment out the numeric constraints on them (or substitute some other interesting constraints for them). And then run p in a suitable scope (e.g. run p for 8 value).

meaning of Alloy predicate in relational join

Consider the following simple variant of the Address Book example
sig Name, Addr {}
sig Book { addr : Name -> Addr } // no lone on Addr
pred show(b:Book) { some n : Name | #addr[b,n] > 1 }
run show for exactly 2 Book, exactly 2 Addr, exactly 2 Name
In some model instances, I can get the following results in the evaluator
all b:Book | show[b]
--> yields false
some b:Book | show[b]
--> yields true
show[Book]
--> yields true
If show was a relation, then one might expect to get an answer like: { true, false }. Given that it is a predicate, a single Boolean value is returned. I would have expected show[Book] to be a shorthand for the universally quantified expression above it. Instead, it seems to be using existential quantification to fold the results. Anyone know what might be the rational for this, or have another explanation for the meaning of show[Book]?
(I'm not sure I have the correct words for this, so bear with me if this seems fuzzy.)
Bear in mind that all expressions in Alloy that denote individuals denote sets of individuals, and that there is no distinction available in the language between 'individual X' and 'the singleton set whose member is the individual X'. ([Later addendum:] In the terms more usually used: the general rule in Alloy's logic is that all values are relations. Binary relations are sets of pairs, n-ary relations sets of n-tuples, sets are unary relations, and scalars are singleton sets. See the discussion in sec. 3.2.2 of Software Abstractions, or the slide "Everything's a relation" in the Alloy Analyzer 4 tutorial by Greg Dennis and Rob Seater.)
Given the declaration you give of the 'show' predicate, it's easy to expect that the argument of 'show' should be a single Book -- or more correctly, a singleton set of Book --, and then to expect further that if the argument is not actually a singleton set (as in the expression show[Book] here) then the system will coerce it to being a singleton set, or interpret it with some sort of implicit existential or universal quantification. But in the declaration pred show(b:Book) ..., the expression b:Book just names an object b which will be a set of objects in the signature Book. (To require that b be a singleton set, write pred show(one b: Book) ....) The expression which constitutes the body of show is evaluated for b = Book just as readily as for b = Book$0.
The appearance of existential quantification is a consequence of the way the dot operator at the heart of the expression addr[b,n] (or equivalently n.(b.addr) is defined. Actually, if you experiment you'll find that show[Book] is true whenever there is any name for which the set of all books contains a mapping to two different addresses, even in cases where an existential interpretation would fail. Try adding this to your model, for example:
pred hmmmm { show[Book] and no b: Book | show[b] }
run hmmmm for exactly 2 Book, exactly 2 Addr, exactly 2 Name

Is there any object-oriented static typed language with variables with few types?

I like reading about programming theories, so could you tell me if there is any object-oriented static typed language that allow variables to have a few types?
Example in pesudocode:
var value: BigInteger | Double | Nil
I think about way of calling methods on this object. If object value have type BigInteger | Double language could allow user to call only shared methods (lake plus, minus) but when the type is BigInteger | Double | Nil then object of Nil hasn't methods plus and minus, so we can't do anything usefull with this object, because it has only few shared methods (like toString).
So is there any idea how should work calling methods on variable with few types in static typed object-oriented language?
What you are describing is an intersection type. They do exist in Java, for example, but they only arise within the type-checker as the result of capture conversion and type-inference. You cannot write one yourself.
I don't know of any language which uses them directly, but they are often used to describe or analyze type systems of languages, espececially languages which don't actually have a type system. For example, Diamondback Ruby, which is a static type system and type-inferencer for the dynamically typed Ruby programming language, uses both union and intersection types.
Note that the syntax you are using is generally used to denote union types, which are the dual of intersection types. Intersection types are generally written A & B & C.
I am not aware of any language that does this... sadly, I'd love to play around with it (but first, they should adopt type inference and parametric polymorphism ;) ).
Although it is alreapossible: Relatively elegantly in a structural type system (type a is a subtype of type b if a has everything b has), simply by specifying a type for value that is a structural subtype of BigInteger and of Double and of Nil and slightly less elegantly in a nominative type system (type a is a subtype of type b if and only if it inherits from it, directly or indirectly) by specifying a common ancestor of all three (if all else fails, object). Of course we'd need to go recursive - what is the type of toString? And what's the typ of (Integer | Double | BigInteger).+?!? This is far from trivial (in fact, looking for a solution made my head hurt a bit). I can't say if it is impossible, but no mainly-OO-language's type system is anywhere sophisticated enough for a possible solution.
The bottom line is: It'd be really cool if some whizz came along and sorted out the issues it raises. Propably not worth the effort...
Edit: Do you know algebraic data types? They are similar to your idea (but much older ;) ) in that an algebraic data type is composed of several types and can therefore contain e.g. a BigInteger, a Double and Nil - the actual value is one of these and a tag (as in tagged union) says which. But to use the value stored in an algebraic data type, you have to use pattern matching to extract it safely. This concept is very powerful, and still "simple" enough to be understood tools - e.g. type inference and static typechecking work.
It has not much to do with OO but (as far as I understand it) what you describe looks much like polymorphism as implemented by C++.
Yes, OCaml has these in the form of polymorphic variants:
type my_var = Integer of int | Float of float;;
let x = Integer(10);;
let y = Float(3.14);;
Pike has them, as does Magpie, an optionally-typed language I'm working on. Google's Closure compiler for Javascript allows you to annotate types in Javascript using |.
They crop up frequently in languages that bridge static and dynamic typing because a lot of expressions in a dynamic language can yield one of a couple of types:
var a = 123;
if (foo) { a = "string"; }
bar(a);
The statically-determined type being passed to bar() is Number | String.
I'm not so sure if we really have a complete definition of what a static typed language is but I also hope that the language you describe wouldn't qualify as one.
One of my concerns is that if you add type T1 and T2 to be a part of your BigInteger | Double | Nil, how would they know about each other and how to handle the operations you defined? Now I realize you never said that the language would allow expanding the "implicit" conversion definition.
Come to think of it, C# does something that resembles this in its string handling
string s = -42 + '+' + "+" + -0.1 / -0.1 + "=" + (7 ^ 5) +
" is " + true + " and not " + AddressFamily.Unknown;
=> "1+1=2 is True and not Unknown"
string str = 1 + 2 + "!=" + 1 + 2;
=> "3!=12"
And I do not like it.

Resources