Search Results

It has tons of breaking changes. For example, if they've found an inconsistency, they'll try to restrict some features, so code surely breaks. For instance, they used to have a special treatment for w …

So, if you have installed Agda, remember the version and download the suitable version of the libraries. OTOH, if you intend to use a certain library, make you install the suitable version of Agda. … For example, the compiler of Cedille does not work with latest Agda (see my pull request. …

Then disabling eta equality for a record saves time for Agda. …

First of all, you may simplify your code with the ∀ keyword: data Σ' ∀ {i} {j} (A : Set i) (B : A → Set j) : Set (i ⊔ j) where pair : (x : A) → B x → Σ' A B You may also make use of the variable …

This is a problem of the implementation of Cubical Agda. … compPath-filler: Refl∙ {x = x} {y = y} p = sym (compPath-filler p refl) A simple explanation of the error message: hcomp (λ { j (i = i0) → x ; j (i = i1) → refl j }) (p i) is compPath p refl, and this is what Agda …

I would say in Agda, coinductive records is a technique of implementing codata types. … Records in Agda are by default non-recursive, say, you may think of them as a named version of sigma types (so they have η-rules, etc.), but coinductive records are totally different, either in the semantic …

Is there a way to write a function like this, in which the result depends on x? foo : Test → Bool foo point = true foo (loop false i) = true foo (loop true i) = true The logic is equivalent to your …

Apart from @Trebor's answer, there is another way to specify the implicit argument: (x , u) ≡ (_,_ {B = B} y (transp (λ i → B (p i)) i0 u)) You may also remove the parentheses on the right hand side: …

Nowadays, thanks to the developers (I think primarily Jesper Cockx and Andreas Abel), the Agda LHS elaborator can infer dotted patterns automatically, so you don't have to mark it with a dot anymore if … However, if you want to put expressions there, you will need the dot, like if you remove the dot above and Agda 2.6.5-9e4afb1 will still complain about it. …

After some research I have found a counterexample by Jesper Cockx: data ⊥ : Prop where data Bad : Prop where b : ((P : Prop) → P → P) → Bad bad : Bad bad = b (λ P p → p) no-bad : Bad → ⊥ no-bad ( …

There are three possible approaches: Use a different algorithm, like division in stdlib, see div-helper. Use the well-founded induction library. There are many online resources with examples, like ht …

It means that your implementation of the squash case, let's call it u (where u may refer to i, j, p1, p2, etc., and can perform recursion), must satisfy the following definitional equalities: u [i ↦ …

Apart from judgmental equality, MLTT and CIC also differ in the following: The existence of an impredicative universe. CIC has Prop, and that's what makes it a part of the lambda cube. This universe …

Core term definition In the latest version of Aya, we use locally nameless approach where names are represented by Java object identity (see this), and bound variables are integers (see this). Bound v …