Struct wlambda::vval::VValFun

pub struct VValFun {
    pub fun: FunType,
    pub upvalue_pos: Rc<Vec<VarPos>>,
    pub upvalues: Vec<VVal>,
    pub local_size: usize,
    pub min_args: Option<usize>,
    pub max_args: Option<usize>,
    pub err_arg_ok: bool,
    pub syn_pos: Option<SynPos>,
    pub label: VVal,
}

This structure is the runtime representation of a WLambda function value.

Fields

fun: FunType

The closure or vm program that runs the function.

upvalue_pos: Rc<Vec<VarPos>>

The positions of the upvalues that are being captured by this function.

upvalues: Vec<VVal>

Contains any caught upvalues.

local_size: usize

The number of local variables defined in this functions.

This value is used to reserve stack space for storing them.

min_args: Option<usize>

Min number of arguments this functions requires.

max_args: Option<usize>

Max number of arguments this functions requires.

err_arg_ok: bool

If true, then this function accepts error values without panic. Functions by default don’t accept errors as argument. It needs to be explicitly enabled.

syn_pos: Option<SynPos>

The location of the definition of this function.

label: VVal

The return label of the function:

Implementations

impl VValFun

pub fn new_fun<T>(
    fun: T,
    min_args: Option<usize>,
    max_args: Option<usize>,
    err_arg_ok: bool
) -> VVal where
    T: 'static + Fn(&mut Env, usize) -> Result<VVal, StackAction>, 

Creates a new VVal containing the given closure with the given minimum and maximum parameters (see also add_func of GlobalEnv).

There is also a new more convenient (because provided by VVal itself) function: VVal::new_fun which has the same parameters.

The err_arg_ok parameter specifies whether the function accepts error values as arguments. If it doesn’t, the program will panic once an error value is encountered. This makes programs more maintainable.

This is usually useful if you want to add functions to the EvalContext. at runtime.

 use wlambda::compiler::EvalContext;
 use wlambda::vval::{VVal, VValFun, Env};

 let mut ctx = wlambda::compiler::EvalContext::new_empty_global_env();

 ctx.set_global_var("xyz",
     &VValFun::new_fun(
         move |env: &mut Env, argc: usize| {
             Ok(VVal::new_str("xyz"))
         }, None, None, false));

 assert_eq!(ctx.eval("xyz[]").unwrap().s_raw(), "xyz")

pub fn new_fun_with_pos<T>(
    fun: T,
    min_args: Option<usize>,
    max_args: Option<usize>,
    err_arg_ok: bool,
    spos: SynPos
) -> VVal where
    T: 'static + Fn(&mut Env, usize) -> Result<VVal, StackAction>, 

pub fn new_val(
    fun: ClosNodeRef,
    upvalues: Vec<VVal>,
    env_size: usize,
    min_args: Option<usize>,
    max_args: Option<usize>,
    err_arg_ok: bool,
    syn_pos: Option<SynPos>,
    upvalue_pos: Rc<Vec<VarPos>>
) -> VVal

Internal utility function. Use at your own risk, API might change.

pub fn new_prog(
    prog: Rc<Prog>,
    upvalues: Vec<VVal>,
    env_size: usize,
    min_args: Option<usize>,
    max_args: Option<usize>,
    err_arg_ok: bool,
    syn_pos: Option<SynPos>,
    upvalue_pos: Rc<Vec<VarPos>>,
    label: VVal
) -> VVal

Internal utility function. Use at your own risk, API might change.

pub fn new_dummy() -> Rc<VValFun>

Returns a dummy function that does nothing.

pub fn dump_upvals(&self) -> VVal

Dumps captured up values of this function. Useful only if you want to debug functions/closures creates by WLambda code.

Trait Implementations

impl Clone for VValFun

fn clone(&self) -> VValFun

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for VValFun

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.