ANONYMOUS wrote:
> Hi, would the following code as a .ml file be syntactically valid?
>
> Additionally, can comments be indented with a '/t' outside of a function or will this cause an error? Chris has said randomly indented lines outside of a function is not permitted, but I am not sure if this applies to comments as well.
It does apply to comments. He made the change to make sure the project isn't too hard.
> This file also assumes that whitespace anywhere except at the start of the line is valid (therefore any line starting with a whitespace (even if it is a commented out line with just #), including any lines with whitespace before or after a '\t' character within a function, causes an error).
> Could I please get a clear cut yes or no whether or not this is true instead of a link to another help forum post, I have read all of the posts regarding whitespaces and still am not certain if this is correct. The post that is always linked says 'Whitespace can only appear at the beginning of a line if it is a tab under a function definition' is true, but whitespace and tab are not the same thing so it is confusing.
>From Oxford English Dictionary:
"white space n. (a) Typography the blank areas of a page or other piece of printed matter, as margins, gutters between columns, etc., regarded collectively as an element of layout and design; (b) Computing (frequently as one word) blank space in electronic text produced by one or more keyed characters, as spaces, tabs, line-breaks, etc.; often attributive as whitespace character."
The only allowed white space at the beginning of a statement is a single tab (if in a function body). Otherwise, white space is optional but not necessary. Except for key words like print, function, return where there should be at least one space between them and an expression.
>
> ```
> # this file should print the following (only the numbers):
> # 9
> # 16
> # 0
> # 2
> # 2.500000
> # 3
> # 1
> #
> #
> #
> # additional whitespace except at the start of a line is valid (? still waiting on clarification ?)
Yes (however statements in a function must only start with a single tab)
> # 9 is printed
> one <- 1
> #
> function increment value
> return value + one
> #
> print increment( 3 ) + increment ( 4 )
> #
> #
> #
> # from
[help2002]⬈
> # functions can call another function as a parameter
> # 16 is printed
> function printsum a b
> print a + b
> print printsum(increment( 5 ) , 10)
> #
> #
> #
> # variables do not need to be defined before being used in an expression, and are automatically initialised to the (real) value 0.0
> print d
> #
> #
> #
> $ from
[help2002]⬈
> # functions do not need to end with a return or print statement, if there is no tab the function has ended
> # variable reassignment is allowed
> # 2 is printed
> function none
> one -> 2
> none( )
> print one
> #
> #
> #
> # comments can start in the middle of a line
> # 2.500000 is printed
> print 2.5 # comment
> #
> #
> #
> # from
[help2002]⬈
> # 3 is printed then 1 is printed, since noreturn(1, 2) would return 0
> function noreturn x y
> print x + y
> print 1 + noreturn(1, 2)
> ```
Only thing that might be invalid is:
> function none
> one -> 2
> none( )
> print one
> #
The "one" in the function would likely be a local variable, after the function ends it would likely not exist (and so 0 would be printed). I'm not sure if global variables can be accessed by functions in ml. It would be best to ask Amitava to clarify.
Also, the arrow it the wrong way around.
