Skip to content

Latest commit

 

History

History
249 lines (194 loc) · 8.02 KB

README.md

File metadata and controls

249 lines (194 loc) · 8.02 KB
Raw

L-Interpreter

Description

L Interpreter is a Haskell-writen CLI Application that parse and execute code written to L language

Syntax

Types

L Language has types:

  • Float (as a number)
  • String
  • Boolean

Variables

x := 5
y := 15.423
z := x * y + 1
stringVar := "Example"

Expressions

Expression examples

L language support expression in the format below

exprNum := -8 * 91
exprBool := True || True && False
exprStr := "Example of " + "string concatenation"

L language supports brackets "(" and ")" in expression. Below you can find examples of valid expressions

exprNum1 := 8 * (3 + 6)
exprNum2 := (-(8 * -(3 + 6)) + (7 * 6)) * 2
exprBool := (True || True) && False

Priorities for operation in expressions

L language has different operation priority within expressions:

  • Priority "4" garantee that operation will be executed firstly;
  • Priority "0" garantee that operation will be executed at the end.
Operation Priority
* 4
/ 4
+ 3
- 3
== 2
!= 2
> 2
>= 2
< 2
<= 2
&& 1

Types of Expressions

Expressions in L language does not have strict type. Variables could changes the type depending on defition. Below you can find example of valid expression named expr and reassigning this expr.

getInput varName <- {
    write "Input value for var named " + varName
    read
}

a := getInput "a"
b := getInput "b"

expr := a + b
expr := 5 + 7
expr := 12.58
expr := "String again"
expr := True

Lazy computations

Expressions provide mechanism of lazy computations

Example below demonstates the application on this mechanism. In this case, expr is not going to be evaluated fairy, because the first part is always True. Then there is no need to evaluate the other options with the || operator.

a := 5
b := 7

expr := True || a * 35.48 / 41 > b * b - 44 || ...

The verse logic works for False with && operator. Variable expr would take False value without evaluation the rest of whole expression.

a := 5
b := 7

expr := False && a * 35.48 / 41 > b * b - 44 || ...

Conditional computations

Not strict types provide light and usefull mechanism to use conditional computations. The code below demonstates the easy behavior of if-else statement without else branch.

notification message <- {
    write "A problem with appeared with: " + message
}

a := 5
b := 7

a < b || notification ("5 is not greater than 7")

As you can see, the left part is simply dropped. This means the whole expression could have as Boolean type (if the left part is True/False with corresponding functions), as any other type depending on the rest part of expression.

If - else

If-else statement exists in L language and describes possibility of program behavior depending on the condition. In L language both "then" and "else" branches are required.

a := 5
b := 6

if a == b then { 
    write "Math does not work"
 } else {
    write "Math, hopefully, works"
 }

While loop

If-else statement exists in L language and describes possibility of program behavior depending on the condition. In L language both "then" and "else" branches are required.

getUserInput type <- {
    write "Please, input your, " + type
    read
}

userInput := getUserInput("name")

while userInput != "Nikita" do { 
    write "You are not Nikita! Please, call author of L Interpreter"
    userInput := getUserInput("name")
}

Skip

Keyword that provide you to skip the branch or the body is "skip"

Write

Command "write" provide you to write any kind of information to the file. The argument of command write is an expression, so write firstly evaluates it and then writes to the file.

The file, that "write" writes anything, has default value of "output.txt". However, it's possibly to redirect out of the program. The only thing you need to do is override variable "outputFile". This variable is accessible everywhere

It's possible to override it during some scope to provide logging, and any other kind of interaction. See the example below with evaluation of fib.

fib n <- 
    {   
        outputFile := "logs.txt" 
        write n
        if n == 0 || n == 1 then { 1 } else { fib(n - 1) + fib(n - 2) }
    }

outputFile := "output.txt" 
write fib(15)

Moreover, you can easily override it to the value "str.out" to provide program write to console.

Read

Command read provide ability to interact with user througth the terminal. It returns the user input. As you can see, read is a statement, not an expression. Therefore we need to wrap it into the function and then call a funtion to provide interaction. Otherwise it would not work, because variable binding accepts expression part on the left (not statement)

getUserInput type <- {
    write "Please, input your, " + type
    read
}

x := getUserInput "example"

Functions

Function is the core element of execution You can declare it througth the sytax

mySum a b <- {
    a + b
}

The last expression in the function define the return value in the whole function

Functions can have their own function declaration and private variable inside of this scope. For example this code will throw Error, because abs(-5) does not exist outside of the scope "mySum"

mySum a b <- {
    abs value <- {
        if value > 0 then { value } else { -value }
    }

    abs(a) + abs(b)
}

mySum(1, -1)
abs(-5)

Lexical environment

Interaction with the lexical environment is implemented using a stack.

The main entry point to the program is the "main" function, which is created automatically. It already contains variables. For example, output File. He is responsible for which file the information will be output to using the write command. This variable can be easily overwritten.

Let's look at the mechanism of the interpreter's work with lexical environments: 0. The input program is being parsed

  1. All user code is wrapped in the main function
  2. The functions declared by the user inside main are packed into a list and become available inside main. Also, for each internal function, the declared functions are packed into a list, which becomes available only to the internal function.
    • Consider the example below. The innerFunction1 and innerFunction2 functions are declared inside the main code (main). There is a function named myFunc for each of them, but these are different functions.
        innerFunction1 x <- {
        myFunc y <- {
            y
        }

        myFunc(x)
    }

    innerFunction2 x <- {
        myFunc y <- {
            y * y
        }

        myFunc(x)
    }

    out1 := innerFunction1(2)
    out2 := innerFunction2(1)
    • Let's take a look at the conditional scheme of the AST tree. We will omit information about the available variables, we will pay attention only to the functions.
    • main - [ innerFunction1 - [myFunc], innerFunction2 - [myFunc] ]
    • Only two functions are available inside main: innerFunction1 and innerFunction2
    • Inside of innerFunction1: myFunc
    • Inside of innerFunction2: myFunc
  3. Code execution starts (note that at this stage the code has already been analyzed and all functions have surfaced).
  4. The code is executed line by line
  5. When calling a function, we add the lexical environment of this function to the execution stack. Now we can access local functions (declared inside the executable function) as well as functions inside main. The same principle applies to variables.
  6. When exiting a function, we remove the lexical environment of that function. The execution of the "main" code continues
  7. When executing all commands from "main", the lexical environment of main is removed from the execution stack. The execution of the program is completed.