Struct mozjs_sys::jsapi::JS::Value

#[repr(C)]
pub struct Value {
    pub asBits_: u64,
}

[SMDOC] JS::Value type

JS::Value is the interface for a single JavaScript Engine value. A few general notes on JS::Value:

• JS::Value has setX() and isX() members for X in

  { Int32, Double, String, Symbol, BigInt, Boolean, Undefined, Null, Object, Magic }

  JS::Value also contains toX() for each of the non-singleton types.

• Magic is a singleton type whose payload contains either a JSWhyMagic “reason” for the magic value or a uint32_t value. By providing JSWhyMagic values when creating and checking for magic values, it is possible to assert, at runtime, that only magic values with the expected reason flow through a particular value. For example, if cx->exception has a magic value, the reason must be JS_GENERATOR_CLOSING.

• The JS::Value operations are preferred. The JSVAL_* operations remain for compatibility; they may be removed at some point. These operations mostly provide similar functionality. But there are a few key differences. One is that JS::Value gives null a separate type. Also, to help prevent mistakenly boxing a nullable JSObject* as an object, Value::setObject takes a JSObject&. (Conversely, Value::toObject returns a JSObject&.) A convenience member Value::setObjectOrNull is provided.

• Note that JS::Value is 8 bytes on 32 and 64-bit architectures. Thus, on 32-bit user code should avoid copying jsval/JS::Value as much as possible, preferring to pass by const Value&.

Spectre mitigations

To mitigate Spectre attacks, we do the following:

• On 64-bit platforms, when unboxing a Value, we XOR the bits with the expected type tag (instead of masking the payload bits). This guarantees that toString, toObject, toSymbol will return an invalid pointer (because some high bits will be set) when called on a Value with a different type tag.

• On 32-bit platforms,when unboxing an object/string/symbol Value, we use a conditional move (not speculated) to zero the payload register if the type doesn’t match.

Fields

asBits_: u64

Implementations

impl Value

fn asBits(&self) -> u64

pub fn is_undefined(&self) -> bool

pub fn is_null(&self) -> bool

pub fn is_null_or_undefined(&self) -> bool

pub fn is_boolean(&self) -> bool

pub fn is_int32(&self) -> bool

pub fn is_double(&self) -> bool

pub fn is_number(&self) -> bool

pub fn is_primitive(&self) -> bool

pub fn is_string(&self) -> bool

pub fn is_object(&self) -> bool

pub fn is_object_or_null(&self) -> bool

pub fn is_magic(&self) -> bool

pub fn is_symbol(&self) -> bool

pub fn is_bigint(&self) -> bool

pub fn is_gcthing(&self) -> bool

pub fn to_boolean(&self) -> bool

pub fn to_int32(&self) -> i32

pub fn to_double(&self) -> f64

pub fn to_number(&self) -> f64

pub fn to_string(&self) -> *mut JSString

pub fn to_object(&self) -> *mut JSObject

pub fn to_object_or_null(&self) -> *mut JSObject

pub fn to_symbol(&self) -> *mut Symbol

pub fn to_bigint(&self) -> *mut BigInt

pub fn to_private(&self) -> *const c_void

pub fn to_gcthing(&self) -> *mut c_void

pub fn is_markable(&self) -> bool

pub fn trace_kind(&self) -> TraceKind

Trait Implementations

impl Clone for Value

fn clone(&self) -> Value

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for Value

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

Formats the value using the given formatter.

impl GCMethods for Value

unsafe fn initial() -> Value

Create a default value

unsafe fn post_barrier(v: *mut Value, prev: Value, next: Value)

Place a post-write barrier

impl PartialEq for Value

fn eq(&self, other: &Value) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl RootKind for Value

type Vtable = ()

const VTABLE: Self::Vtable = ()

const KIND: RootKind = JS::RootKind::Value

impl Copy for Value

impl StructuralPartialEq for Value