Data Types

fn main() {
    const age = 31
    const height = 1.68
    const price: Decimal = 499.99
    const name = "Asha"
    const initial = 'A'
    const member = true
    print("{name} ({initial}), {age}, {height} m, pays {price}, member: {member}")
}

Asha (A), 31, 1.68 m, pays 499.99, member: true

You didn't write a single type: the compiler works each one out from the value.

Built-in types

Types built from other types

No null: none instead

Kavach has no null that any value could secretly be. A value that may be missing has a ? in its type, as in String?, and is none when missing. The compiler then makes you handle that case, here with ??, which gives a fallback: see Optionals.

JSON does have null. Read as a Json value it is Null, and a key that isn't there reads as Null too. Decoded into your own type, it becomes none:

use std.json

type User {
    name: String
    nickname: String?
}

fn main() ! Error {
    const text = "\{\"name\": \"Asha\", \"nickname\": null}"
    const doc = try json.parse(text)
    print(doc["nickname"].is_null())
    print(doc["missing"].is_null())
    const user = try json.decode<User>(text)
    print(user.nickname ?? "no nickname")
    print(user.nickname == none)
}

truetrueno nicknametrue

Writing null (or nil, None, undefined) in Kavach code is an error that says what to write:

fn main() {
    let nickname: String? = null
    print(nickname ?? "no nickname")
}

error: Kavach has no `null`: a value that may be missing has an optional type (`String?`), and the missing value is `none` (spec 2.7)

Where a Json value is expected, none becomes JSON's null; more on Json values.

Kavach never turns one number type into another behind your back. An Int plus a Float is an error until you say which you want: see Converting types.

Limits and sizes

Every limit in Kavach, on one page: Limits and Sizes.