nessa-language 0.9.1

An extensible programming language with a strong type system
Documentation
Let's create arguably one of the most useful constructs when working with iterable structures:
list comprehensions. This can be done thanks to Nessa's syntax definition features.

## What is *list comprehension*?


List comprehensions are a syntax option in some languages that allow the programmer to
**transform** the values of a list or even filter them without having to create an explicit for loop and
a temporary variable. In this case, we will not deal with filtering the list.

## Creating the syntax


The syntax for list comprehensions then to be something like this:

```
[expr for elem in container]
```

This is problematic for us because Nessa is a strongly typed language. We will have to adapt it like this:

```
[expr for elem: ElemType in container]
```

This can be expressed in *NDL* like this:

```
"[" 
    [s] Arg(<expr>, map) [s]                                        // Map expression
    "for" [s] Arg(<ident>, it) [s] ":" [s] Arg(<type>, type) [s]    // Element
    "in" [s] Arg(<expr>, container) [s]                             // Container
"]"
```

Note that you need `Arg`s to mark the variables that we will use inside the body.

## Generating the body


Internally, a list comprehension is just a **for loop** with a transformation function. There are multiple ways to do this, but
this is one of them (again, note that we have to escape the for's closing brace):

```
syntax list_comprehension from [...] {
    let res = arr<$type>();
    let func = ($it: $type) -> $type $map;

    for _it_ in $container {
        res.push(func(*_it_));
    \}
    
    return move(res);
}
```

## Example


Making use of an `Array` initialization syntax, you can build the following code:

```
let array = [i * 2 for i: Int in <Int>[1, 2, 3, 4, 5]]; // [2, 4, 6, 8, 10]
```

This would compile to:

```
let array = do {
    let res = arr<Int>();
    let func = (i: Int) -> Int i * 2;

    for _it_ in <Int>[1, 2, 3, 4, 5] {
        res.push(func(*_it_));
    }

    return *res;
};
```

In this example `<Int>[1, 2, 3, 4, 5]` would also be recursively compiled into a new expression based on a lambda expression, but
the exact compilation is not important.