Basic Syntax

identifiers

Identifiers can (currently) use

  • alphanumeric A-Za-z
  • digits 0-9
  • underscore _
  • Other special characters such as ~@#$^&;? should not be used since they may be used in the future.

    Whitespace is not significant, but is not allowed inside of indentifiers, since they are used for tokenization.

    numbers

    Numbers are the same as in most programming languages (integer, floating point, hex). Octal is not supported. Binary is not currently supported (may revisit).

  • integer -? [0-9]+ ex: 42
  • decimal -? [0-9]+ . [0-9]* ex: 42.42
  • scientific notation decimal e integer ex: 42.42e42
  • hex: 0x [0-9] [A-F] [a-f] ex: 0xFFAAbb. Maybe also use # to indicate hex?
  • Underscores may be placed anywhere within numbers to help with readability. They will be stripped before evaluation.

    42_000_000 is easier to read than 42000000

    Units are word suffixes. They may or may not be separated from the number part by a single space. Separating by newlines is not allowed.

    strings

    Single, double, and back quotes are all allowed for creating text literals. There is no distinct character type. smart quotes or curly quotes will be replaced with their standard equivalent.

    Operators

    Boolean Operators

  • < less than
  • > greater than
  • <= less than or equal
  • >= greater than or equal
  • = equal
  • <> boolean NOT equal
  • != boolean NOT equal? What about unicode:
  • and boolean AND
  • or boolean OR
  • not boolean NOT
  • Math Operators

  • + add(a,b), addition
  • - sub(a,b) subtraction
  • * mul(a,b) multiplication
  • / div(a,b) division
  • ** pow(a,b) exponentiation
  • - neg(a) negative sign (depending on context)
  • ! fact(a) factorial
  • mod mod(a) modular (remainder ) division. % is used for percentages
  • All the rest

    () parenthesis for grouping and function calls [] square brackets for list literals, maps to list(a,b,c,...) function << assign >> assign and pipeline operator (depending on the context) as convert to a specified unit. ex: 10cm as inches . works on lists.

    Specifically Excluded

  • bit shifting << >>
  • bitwise booleans | &
  • increment and decrement shortcuts: -- -= += ++
  • symbols for boolean AND, OR, NOT, XOR
  • ranges: 1..10 , 1..<10 and other comprehensions (may revisit)
  • % for modular division. use mod instead
  • curly braces {} are currently not used and are often harder to type on non English (QWERTY) keyboards.
  • Functions

    Definition

    Functions are named by standard identifiers. They are called using parenthesis with one or more arguments. Arguments may be referred to by position (indexed) or name (keyword).

    Consider the chart function.

    myfun << (data:?, type:'point', color:'red') -> { }

    myfun is an identifier. A function which takes x and y is declared and assigned to myfun. Essentially all functions are anonymous functions. Each argument includes a default value. The default of '?' means it has no default and must be provided by the caller.

    Parameter Resolution

    The arguments may be called by order or name. ex:

    call by order

    myfun( [1,2,3], 'point', 'red')

    call by keyword

    chart( data=[1,2,3], color='red')

    call by order and keyword

    chart( [1,2,3], color='red')

    Parameter resolution is as follows.

  • all parameters may be specified by name or position (index)
  • the parameters are prepared by the runtime from the call site at runtime. all named args are filled in first, from left to right in the parameters def. any indexed arguments are filled into missing parameters, left to right. if there are no required parameters left, run the function if there are unfilled required parameters, throw an error
  • no varargs. use lists instead.
  • inside the function you just use the named parameters. don't care about positional args or indexed vs keywords.
  • pipeline operators

    The << and >> operators are called pipelines. They take something and pass it into something else. They are used to assign values to variables, and to pass the output of one function into the first parameter of the next.

    assign 42 to answer

    answer << 42
    42

    is the same as

    42 >> answer
    42

    pass the output of range into the print function

    range(10) >> print()

    is the same as

    print(range(10))

    Control Flow

    If statement

    The if statement works as follows

  • The if keyword followed by a boolean expression, followed by then , followed by a block.
  • Optional else block.
  • blocks may be replaced by single expressions.
  • ex:

    if (x = 5) then { 5 } else { 10 }

    using single expressions

    if x = 5 then 5 else 10

    the if statement itself is an expression, so it's result can be assigned to a variable or used inside another expression

    x << 56 is_even << if x mod 2 = 0 then true else false

    =================

    notes

    if( test: exp, then: exp exp exp, else: exp exp exp )

    is_even << if( test: x mod 2 = 0 then: true else: false )

    is_even << def (x:?) -> start x mod 2 = 0 end

    Match statement

    Still being defined. I'm leaning towards something like

    match x start 5 -> expression 6 -> expression x when x < 5 -> expression end

    Misc

    Unicode operators and identifiers

    While any unicode characters can be used in strings and identifiers, these will likely have special support in the interface to make them easier to enter. They will also be shown in their unicode form in the "print ready" syntax.

  • theta ø or θ
  • pi π or
  • alpha α
  • sigma σ and Σ
  • not equal , greater than or equal, and less than or equal
  • right arrow replaces >>
  • left arrow replaces <<
  • curved arrow or replaces return ??
  • Issues

    I'd really like to get rid of the curly braces for blocks. I'm not sure how to do that. The if statement already has delimiters in the form of then and else . By adding end I could get rid of the braces. I don't know about function definitions and match expressions yet.

    Match could be

    match exp start exp fun exp fun exp where exp fun else exp end

    if could be

    if exp then exp exp else exp exp end

    Function definitions could be

    myname << define args start exp exp exp end

    Unicode stuff isn't working in docs. I don't know why. The symbols always get garbled. Ideally you'd write with some sort of abbreviation and the editor would immediately replace it with the unicode form.